This paper presents a high-accuracy single-eye vision measurement system using digital cameral, which can substitute for universal tool maker's microscope. Based on the fact that the measurement of structural parameters ends in the operation of the coordinates of feature points, three main measures were put forward to improve the measurement accuracy, i.e., the method of correcting the image distortion by cameral calibrating; the method of fine adjusting by image the position of the camera to make the camera perpendicular to the surface of a measured object; the method of sub-pixel edge-detection technique that combines the B-spline and the least squares fitting. The experimental tests confirm that the measurement system is of high accuracy, effectiveness and real time response.

In order to develop a kind of universal controller that is very convenient in actual applications, it is necessary to utilize the universal approximation characteristic of an intelligent control algorithm. But it is very hard to implement control in real-time because of the large amount of computation needed. A fuzzy control, with human simulated intelligent weight coefficient, was selected as the control algorithm, since its computations are less than that of the other intelligent approaches. To improve the performance in real time, a high-speed memory address mapping approach was adopted as the hardware implementation. The entire ideas, including the algorithm and approach, are introduced in detail. Based on theoretical studies, corresponding hardware circuits were designed, and this kind of fuzzy controller was developed. In the experimental study, the experimental device of machine control was selected, and the fuzzy controller developed to manipulate the machine's speed was applied. Under the same experimental conditions, the control performance of the fuzzy controller was excellent with respect to its speed. The control quality was found to be better than that of PID hardware circuit control system.

This paper proposed a method which was based on DSP and FPGA and applied in the fully digital and high-performance permanent magnet synchronous motor (PMSM) AC servo system. In the system, the position loop control combined PI control and feedforward control was implemented by DSP unit, while the speed loop and current loop control were implemented by FPGA unit. Since FPGA provided a kind of hardware design scheme for the control system, it had the features of a special control IC. The functions of vector control algorithm, velocity and current sampling, space vector pulse width modulation and PI algorithm were included in the special control IC. In the meanwhile, it also provided host communication interfaces that could realize the real-time modification of the control parameters in the IC. Finally, the experimental results indicate that this system has high-speed response and well static state control performance.

For the purpose of cooling the electronic device rapidly, a cooling system based on Peltier elements has been designed and manufactured. The mathematic model of the system was built by theoretic deduce. The system is able to actively control the euthermic device's virtual thermal resistance to its surroundings and efficient cooling can be achieved. According to the proposed model, the relation between the input power of the Peltier cell and the power loaded on the euthermic device was simplified, moreover, a new multi-modal control strategy based on low power consumption was presented. The experiments proved that the use of this cooling system with our control strategy increased the control precision up to 0.8% when the power loaded on the euthermic device was at 12W.

The overall performance of Particle Swarm Optimizer lies on its ability to harmonize global and local search process. By dividing the whole swarm into equal sub-swarms with iterative cooperation, and taking a series of Sugeno functions as inertia weight decline curves for each sub-swarm, an adaptive strategy was proposed to adaptively select different inertia decline curve according to the vary rate of the sub-swarm's fitness value. Experimental results on several benchmark functions show that the modified algorithm can effectively balance global and local search ability to avoid premature problem, and obtain better solutions with higher convergence speed.

To reduce the low-frequency hydraulic-mechanical oscillations when a hydraulic crane boom system (HCBS) works, an improved PI control strategy based on fuzzy scheduling set-point weighting was proposed for the active damping control of a HCBS. By using a range camera and an electronic feedback control, the damping on the crane tip can be adjusted artificially. A co-simulation technique of HOPSAN and MATLAB is used in the controller design. Simulation results show that the proposed PI control system is with less overshoot and fast response. And the tip damping during HCBS working is improved.

This new design method is aimed at the disadvantages that the energy cannot feed back into the power network with a non-electric dynamometer, and a DC dynamometer is influenced by a commutator, a method of dynamometer design, based on direct torque control, is introduced. The dynamometer has the characteristic of constant torque when the rotation's speed has a downward rating value, and it has the characteristic of constant power when the rotation's speed has an upward rating value. The system experiment shows that direct torque control can effectively reduce torque pulsation and obtain satisfactory dynamic performances, and that the torque can be controlled accurately and quickly. The AC dynamometer has better performance and it has important application value.

This paper engages to study the 3-DOF attitude control problem of the kinetic interceptor. When the kinetic interceptor enters into terminal guidance it has to maneuver with large angles. The characteristic of interceptor attitude system is nonlinearity, strong-coupling and MIMO. A kind of inverse control approach based on adaptive wavelet neural networks was proposed in this paper. Instead of using one complex neural network as the controller, the nonlinear dynamics of the interceptor can be approximated by three independent subsystems applying exact feedback-linearization firstly, and then controllers for each subsystem are designed using adaptive wavelet neural networks respectively. This method avoids computing a large amount of the weights and bias in one massive neural network and the control parameters can be adaptive changed online. Simulation results betray that the proposed controller performs remarkably well.

The complex dynamics characters of the mechanical centrifugal flywheel governor system are studied. The dynamical equation and the state equation of the mechanical centrifugal flywheel governor system are established according to Lagrangian equations. By the phase portraits, the motions of the nonautonomous system are studied under the definite parameters. And by Poincare map, the route from periodic motion to chaos is studied under the presented system parameters. An adaptive differential feedback is constructed to control the chaos of the mechanical centrifugal flywheel governor system. The chaos of the system is controlled via x|x|, too. And the periodic motion of the system is anticontrolled to chaotic motion by add a linear feedback item. The phase portraits of the controlled and anticontrolled system are obtained. The advantages of the two controlled methods are that the collect of the control signals are simple and can put on any time and the chaotic system can be asymptotically stabilized to equilibriums with small control, and the control disturbance will tend to zero at the same time. Chaos anticontrolling is helpful to make a non-chaotic dynamical system chaotic. The orbits of the system can be controlled by these three methods according to our target.

Based on errors of torque and stator flux, the reference voltage vector was determined by the principle that the error is zero during one periodic time; then the voltage vector can be composed by using Space Vector Pulse Width Modulation (SVPWM). Because the magnitude and direction of output voltage vector is arbitrary in this method, it is able to track the setting value of torque and flux linkage primely, the voltage and error of flux linkage at the time of sampling can be offset in the next period, so that the pulsation of torque and flux linkage is reduced effectively. Through imitation and experiment, it was proved that this control strategy is correct and effective.

A novel engraving machine (NUT-1A) is presented, in which fully digital controlled AC system based on DEC4DA was used to improve the machining precision and sensitivity. This engraving machine was constructed around AC servo motor with encoder, controlled by a servo motor control card - DEC4DA. As the upper unit of AC servo motor, DEC4DA was a numerical control generator, which received pulses form CPU by ISA bus, and these pulses were amplified and converted to drive AC servo actuator. This novel engraving machine can achieve a higher positioning accuracy of ±0.01mm and positioning repetition of ±0.005mm, and its resolution is 0.001mm/0.0001mm. Moreover, because of multi-closed loops were used in the system, the steady and transient performances are more excellent. This system ensures a much quicker current regulation in closed-loop operation, of acceleration and braking in both directions, as well as stable speed characteristics. Amplifier boards are protected against excessive current, excessive temperature and short circuiting of the motor supply cables.

To investigate the dynamic characteristics of an automotive and develop a control method for its safe braking, a dynamic model of the car with active suspension devices during its braking was established. The dynamic model was with 4 degrees of freedom, that is, two degrees of freedom with the car body and others with forefront wheels and rear wheels. For the purpose of reliable use of the control method in most cases, a wavy road was included in the problem. Considering random road surface, braking deceleration and traveling speed of the car when braking, a fuzzy control method was used. Based on a real car made in China, computer simulation was adopted to investigate the dynamic characteristics of the car and the fuzzy control method. The results show that under the condition of an elaborately developed fuzzy control method used in the 4 active suspension devices, the displacement of the mass center and the turnover angle of the car body are greatly reduced and about only one half of their original values are remained.

A nonlinear dynamic model of a class of underactuated mechanical systems was built using the Lagrangian method. Some system properties such as the system passivity were analyzed. A GA(Genetic Algorithms)-based stable control algorithm was proposed for the class of underactuated mechanical systems. The Lyapunov stability theory and system properties were utilized to guarantee the system's asymptotic stability to its equilibrium. A real-valued GA was used to adjust the parameters of a stable controller to improve the system performance. An underactuated double-pendulum-type overhead crane system is used to validate the proposed control algorithm. Simulation results illustrate the validity of proposed control algorithm under different conditions.

Laser Shock Forming (LSF) is a new forming method for sheet metal, and a settle of LSF equipment carry out technics and forming process, which are made up of high power pulse laser, 5-axis coordinated movement worktable, Industrial Personal Computer, SERCOS field bus and control software etc. In order to satisfy request to adjust the work surface which needs forming to be perpendicular to laser beam, put forward a kind of new B-spline interpolation arithmetic for 5-axis movement, and according to emulation by MATLAB, it testified that maximum error when interpolated a dimensional curve was not more than one micron if instruction speed is 600 millimeters per second.

Today most movement control cards need special control software of topper computers and are only suitable for fixed-axis controls. Consequently, the number of axes which can be controlled is limited. Advanced manufacture technology develops at a very high speed, and that development brings forth. New requirements for movement control in mechanisms and electronics. This paper introduces products of the 5th generation of movement control cards, PMAC 2A-PC/104, made by the Delta Tau Company in the USA. Based on an analysis of PMAC 2A-PC/104, this paper first describes two aspects relevant to the hardware structure of movement control cards and the interrelated software of the topper computers. Then, two methods are presented for solving these problems. The first method is to set limit switches on the movement control cards; all of them can be used to control each moving axis. The second method is to program applied software with existing programming language (for example, VC ++, Visual Basic, Delphi, and so forth). This program is much easier to operate and expand by its users. By using a limit switch, users can choose different axes in movement control cards. Also, users can change parts of the parameters in the control software of topper computers to realize different control axes. Combining these 2 methods proves to be convenient for realizing multi-axis control in numerical control systems.

Because the parameters of car-engines are of the discreteness, nonlinear and uncertain characteristics, a conventional control model can not accurately control the excess air coefficient and the spark advance angle in dynamic procedures. A new auto-optimized fuzzy control model based on genetic arithmetic was introduced. The controller design scheme based on using an embedded operating system to achieve such a control model was discussed. Based on a real-time simulation, using GA resulted in the fuzzy control parameters being optimized automatically. The results of experiments showed that the dynamic indexes and the fuel economy indexes can be improved greatly. The model has the obvious advantage of reducing fuel consumption.

Railway blocking system is the system with the high demanding of real-time performance. Firstly, the tasks and the time limits, which had to be handled for the blocking system, were introduced. The FCFS and the Markov chain were used to set the model for it. By analyzing the performance of the system with the FCFS model found out that it was not satisfied to the real-time performance. Secondly, NPPR model to evaluate the software real-time performance of the blocking processor was proposed. By evaluation, analysis and comparison, the results indicate that the NPPR model is prevail over the model of (M/M/1): (N/N/FCFS) in real-time performance. And the priorities of the tasks in the system should be given according to their time limit. With the principle of (M/M/1): (N/N/NPPR), if the priority was given to the tasks properly, the satisfied real-time performance will be gotten. The models were tested in forms software and the satisfied result has been gotten in practice.

A position following control system was discussed to synchronize the rotational position of the dome with that of the theodolite. There were two loops in the servo control system. Open loop control was for speed adjustment, while closed loop control was for position following control. The core controller of the servo control system adopted TMS320F2812 DSP (Digital Signal Processing) chip and an absolute type encoder was used as a position feedback unit. A converter DA8562 converted the digital angular position calculated in DSP to an analogy voltage to control the dome rotation through a series of devices such as frequency conversion actiyator, servomotor and gear case. The experimental results show that the system synchronous following error less than ±2° at 0.1°/s to 35°/s.

In the first portion of this study a linear double inverted pendulum is modeled and an LQR controller is designed to control the pendulum. Following this, the feasibility of a real-time control system based on the Microsoft Windows operating system is analyzed. A multimedia timer is adopted to realize the control of the pendulum based on the Windows system, and the real-time control codes are encapsulated into a single Dynamic Link Library (DLL). LabVIEW is used as the programming tool to design the user interface on the server. Then remote measurement and control system of the linear double inverted pendulum, including starting, stepping, publishing to the web, is realized. While publishing to the web proceeds, the user can control the linear double inverted pendulum in the browser and realize the remote control of the linear double inverted pendulum based on LabVIEW.

The aircraft arresting system is a complicated nonlinear system. This paper analyzes the mechanical-hydraulic structure of aircraft arresting system composed of electro hydraulic valve and establishes the dynamic equation of the aircraft arresting system. Based on the state-feedback linearization of nonlinear system, a PD-based controller is synthesized. Simulation studies indicate, while arresting the different type aircraft, the proposed controller has fast response, good tracking performance and strong robustness. By tuning the parameters of the PD controller, a satisfactory control performance can be guaranteed.

A brief introduction of an optic-electronic tracking system using a fast reflection mirror is presented, the design of the fast reflection mirror control system, based on immune genetic PID algorithm, is described. Besides the ability of a stochastic global searching of simple genetic algorithm, the immune genetic algorithm used involves mechanisms which exist in biological immune systems, such as antigen memory, antibody encouragement and restraint, antibody diversity keeping, etc. By combining an algorithm of biological immunity with a genetic algorithm, a kind of optimal computing model is proposed based on a genetic mechanism of immunity. Taking immune genetic algorithm as a base, the optimal parameters of a PID controller are obtained. The simulation results of an immune genetic PID algorithm for the tracking system using a fast reflection mirror are presented, which show that the fast reflection mirror control system based on an immune genetic PID algorithm possesses superior qualities, indicating good prospects of application.

Now, MATLAB/SIMULINK software is popularly used by automotive electronic control designers to develop automotive electronic control systems and perform numerical simulations. But they will face problems, such as value initialization in the "integrator" block, conversion among different data types, selection of "if" block and "switch" block, realization of the "if-clause" under multiple options and the auto-switching control, etc. Taking as an example the designing of an Automated Mechanical Transmission (AMT) system, this paper discusses some techniques and methods for modeling the automotive electronic control system with MATLAB/SIMULINK, offering designers some successful examples.

In order to achieve JIT (Just-In-Time) level and clients' maximum satisfaction in logistics collaboration, a Virus Evolutionary Genetic Algorithm (VEGA) was put forward under double constraints of logistics resource and operation sequence. Based on mathematic description of a multiple objective function, the algorithm was designed to schedule logistics tasks with different due dates and allocate them to network members. By introducing a penalty item, make span and customers' satisfaction were expressed in fitness function. And a dynamic adaptive probability of infection was used to improve performance of local search. Compared to standard Genetic Algorithm (GA), experimental result illustrates the performance superiority of VEGA. So the VEGA can provide a powerful decision-making technique for optimizing resource configuration in logistics network.

In this paper an active random noise control using adaptive learning rate neural networks with an immune feedback law is presented. The adaptive learning rate strategy increases the learning rate by a small constant if the current partial derivative of the objective function with respect to the weight and the exponential average of the previous derivatives have the same sign, otherwise the learning rate is decreased by a proportion of its value. The use of an adaptive learning rate attempts to keep the learning step size as large as possible without leading to oscillation. In the proposed method, because of the immune feedback law change a learning rate of the neural networks individually and adaptively, it is expected that a cost function minimize rapidly and training time is decreased. Numerical simulations and experiments of active random noise control with the transfer function of the error path will be performed, to validate the convergence properties of the adaptive learning rate Neural Networks with the immune feedback law. Control results show that adaptive learning rate Neural Networks control structure can outperform linear controllers and conventional neural network controller for the active random noise control.

For non-minimum phase plants, Ishihara et al. have proposed a partial LTR method for the design of integral controllers based on the disturbance cancellation. Their method can be termed as a singular partial LTR method since the conventional LTR method can not be used for the disturbance cancellation design. This paper gives a complete characterization of the recoverable targets.

In this paper, we propose a method, which uses Coloured Petri Net (CPN) and genetic algorithm (GA) to obtain an optimal deadlock-free schedule and to solve re-entrant problem for the flexible process of the cluster tool. The process of the cluster tool for producing a wafer usually can be classified into three types: 1) sequential process, 2) parallel process, and 3) sequential parallel process. But these processes are not economical enough to produce a variety of wafers in small volume. Therefore, this paper will propose the flexible process where the operations of fabricating wafers are randomly arranged to achieve the best utilization of the cluster tool. However, the flexible process may have deadlock and re-entrant problems which can be detected by CPN. On the other hand, GAs have been applied to find the optimal schedule for many types of manufacturing processes. Therefore, we successfully integrate CPN and GAs to obtain an optimal schedule with the deadlock and re-entrant problems for the flexible process of the cluster tool.

A control system and its environment are said to be matched if the control specifications are satisfied for all exogenous inputs generated by the environment. This paper considers the so-called matching condition for a system-environment match for sampled-data critical control systems with delay of input. The environment is characterized by the exogenous inputs restricted in rate of change. Fast-discretization technique is used to obtain a discrete-time model including intersample behavior. The matching condition is given in the form of the inequality constraint of the unit step response. The interesting issue on selection of a sampling period is also discussed.

This paper presents a system identification process and control system design of an artificial neural network based suspension assembly with self-sensing micro-actuator for dual-stage hard disk drive. Artificial neural networks can be used effectively for the identification and control of nonlinear dynamical systems such as a flexible micro-actuator and self-sensing system. Three neural networks are developed for the self-sensing micro-actuator, the first for system identification, the second for inverse model for control using laser sensor signal, and the third for inverse model for control using only self-sensing piezoelectric signal. And we use a neural network inverse model to control the suspension assembly which includes the micro-actuator pair. Simulation and experimental results show that good control performance can be achieved by using artificial neural networks approach.

This paper presents a comparison study of performance of three motion controllers such as conventional PID controller, LQR(Linear Quadratic Regulator) and RIC(Robust Internal-loop Compensator). All of comparison studies were verified by using a Industrial Servo Emulator (Model 220) made by ECP Corp. To evaluate the disturbance rejection feature, the various disturbances are applied into a motion system.

The aim of this work is to present an advanced method of estimation of the system defects coming up in robot links and actuators. Considering a fictitious model, a new extended system equation is set up. Based on this extended model an observer is designed for the detection of faults and an observer bank is proposed of faults.

In this paper, we propose a novel algorithm can exactly solve the inverse kinematics of a robot manipulator which does not have the closed form solution. The conventional numerical methods suggested by others have a weak point that they do not consider the feasible joint region of solution. Moreover they are likey to saddle on local points not on the real optimal points of solutions. In this work, we consider the inver kinematics problem as a kind of fixed point problem and propose a solution approach based on unidirectional search and boundary reflection algorithm which can guarantee the global optimal solution and the feasible range of joint limits. Futhermore we present an acceleration idea based upon Steffenson iteration which can speed up the search. Simulation results show the validity and efficiency of our approach.

In the NCS (networked control systems) whose control loops are closed through the serial communication channel, some important issues are network-induced time delay, multiple-packet transmission and data packet dropouts, which are caused by MAC (Medium Access Control) protocol and/or scheduling algorithm, packet size constraint of applied network and fault of the communication channel, respectively. Among these issues, the most important problem which should be overcome from the point of controller design in NCS is the network-induced time delay because the multiplepacket transmission and the data packet dropout problems can be solved by proper assumptions. As is well known, the network-induced time delay can be constant, time-varying, or random variable depending on protocol and/or scheduling algorithm, but it is needed to compensate any kinds of time delay by a proper controller. This paper focuses on LMI(Linear Matrix Inequality) based output feedback controller design in order to compensate network-induced time delay which is the limited time-varying delay in NCS. In our study, the effect of time-varying delay is regarded as a variable in a discrete-time model of NCS within sampling period. The effectiveness of the proposed output feedback controller is shown to be useful by example.

In this paper, we proposed the electric circuit using one common arm of H-Bridge inverters to reduce the number of switching components in multi-level PWM inverter combined with H-Bridge inverters and transformers. And we proposed a switching method that can get same rate of usage at each transformer. We tested the proposed prototype inverter to clarify the proposed electric circuit and reasonableness of the control signal.

This paper proposes a practical design method for ship-to-ship missiles' autopilot. When the pre-designed analogue autopilot is implemented in digital way, they generally suffer from severe performance degradation and instability problem even for a sufficiently small sampling time. Also, aerodynamic uncertainties can affect the overall stability and this happens more severely when the nonlinear autopilot is digitally implemented. In order to realize a practical autopilot, two main issues, digital implementation problem and compensation for the aerodynamic uncertainties, are considered in this paper. MIMO (multi-input multi-output) nonlinear autopilot is presented first and the input and output of the missile are discretized for implementation. In this step, the discretization effect is compensated by designing an additional control input. Finally, we design a parameter adaptation law to compensate the control performance. Stability analysis and 6-DOF (degree-of-freedom) simulations are presented to verify the proposed adaptive autopilot.

Final products of hot rolling process are produced to the form of hot coil. The quality of hot coil to be produced is decided by tension control of finishing rolling in hot rolling process. The fluctuation of the strip tension in conventional hot rolling process is controlled through a looper located in between stands. But, the looper can cause a motor trip and tension hunting in hot rolling process, therefore, alternative method is essential. Hence, in this paper, the mathematical modeling of tension mechanism is implemented to estimate tension and to control tension using the proposed SVR algorithm without looper in hot rolling process. Also, results of the simulation show a performance of estimation and a possibility of tension control using SVR technique.

The control objectives in this paper are to move the gantry of a container crane to its target position and to suppress the transverse vibration of the payload. The crane system is modeled as an axially moving nonlinear string, in which control inputs are applied at both ends, through the gantry and the payload. The dynamics of the moving string are derived using Hamilton's principle. The Lyapunov function method is used in deriving a boundary control law, in which the Lyapunov function candidate is introduced from the total mechanical energy of the system. The performance of the proposed control law is compared with other two control algorithms available in the literature. Experimental results are given.

In most physical processes, the transfer function includes a time-delay, and in the general distributed control system using a computer network, an inherent time-delay exists due to the spatial separation between controllers and actuators. Under the circumstance where an input time-delay exits, the system response overshoots and tends to diverge. For this reasons described above, a controller design method is proposed for a discrete nonlinear system including input time-delay, which adopts the time-discretization using Taylor series. Controllers are synthesized using an input/output linearization method. Finally, several cases of the computer simulations were conducted, and the results validate the proposed methods.

The four rotor flying robot has a simple mathematical model of dynamics and a simple mechanism of mechanical design beside the classic type of helicopter with main rotor and tail rotor. This paper presents a 5-DOF simulator for a four rotor flying robot. And this paper shows the mathematics model including kinematical analysis for the 5-DOF simulator.

In this paper, the design concept of the smart type motor operator valve for nuclear power plant was described. The development objective of the smart valve is to achieve superior accuracy, long-term reliability, and ease of use. In this reasons, developed smart valve has fieldbus communication such as deviceNet and Profibus-DP, auto-tuning PID controller, self-diagnostics, and on-line calibration capabilities. And also, to achieve pressure, temperature, and flow control with internal PID controller, the pressure sensor and transmitter were included in this valve. And, temperature and flow signal acquisition port was prepared. The developed smart valve will be performed equipment qualification test such as environment, EMI/EMC, and vibration in Korea Test Lab. And, the valve performance is tested in a test loop which is located in Seoul National University Lab. To apply nuclear power plant, the software is being developed according to software life cycle. The developed software is verified by independent software V and V team. It is expected that the smart valve can be applied to an existing NPPs for replacing or to a new nuclear power plants. The design and fabrication of smart valve is now being processed.

The focus pull-in behavior analysis in a Blu-Ray disk (BD) and various control approach to improve the performance of it are discussed. The focus pull-in work, which locks the beam spot of an optical pick-up to the information layer in an optical disk, is an essential work in an optical disk drive (ODD), and the error tolerance of it depends on the kind of an ODD. In a BD system, the focus pull-in work should be done within a very narrow beam height region (λ/NA =0.4765μm) and a very short period. Therefore, the moment of the focus pull-in is a transient phenomenon, and more careful analysis of it is necessary. In this paper, the focus full-in work is analyzed in depth, and various control approaches to improve the focus pull-in performance are discussed. First, various detected servo signal during the focus pull-in work are summarized briefly. Then, a novel swing arm type actuator which is developed by our center for a small form factor BD is introduced. Finally, three control approaches to perform the focus pull-in are designed and applied to the system: (i) conventional lead-lag, (ii) fuzzy control, (iii) sliding mode control. From the experiments results, we find that two control approaches, fuzzy control and sliding mode and, have more good focus pull-in performances than the conventional lead-lag approach.

This paper presents mathematical model of six degree-of-freedom (6-DOF) helicopter (ERGO50) in hover, and H₂ feedback controller which is a powerful technique for the MIMO system as a helicopter. Mathematical model of the helicopter is multi-input multi-output (MIMO) and linearized system which accommodates aerodynamics. H₂ controller based on optimal control theory is used in a myriad application and plays an important role as a valuable precursor to other advanced methods for future work, when we need to improve stability of the helicopter. We design linear-quadratic-gaussian controller as H₂ controller. Simulation results show good performance.

This paper addressed mathematical modeling and control of liquid rocket propulsion system (LRPS). Dynamic simulation of LRPS shows that control difficulty is rooted from the slow dynamics and high nonlinearity of the rocket motion. As a consequence a conventional feedback control technique is unsuitable for controlling the rocket motion. In this paper, it is proposed to use both cascade and iterative learning control techniques including feedback controllers. Using the cascade control algorithm it is enabled to overcome the control difficulty to a certain degree caused by the variation of time constants. Iterative learning control strategies improve the tracking performance and guarantee the safety through repetition. The LRPS control system which tracks the changes in the set-point and also diminishes the disturbance under modeling error has been synthesized. Overall performance of proposed control system has been demonstrated by means of thorough numerical simulation of LRPS control system.

In home appliances, electric energy is optimally controlled by using power electronics technology, creating a comfortable environment in terms of energy saving, low sound generation, and reduced time consumption. Usually simplicity and robustness make the three phase induction motor attractive for use in domestic appliance, including washing machines. Two main types of domestic washing machine have evolved. We focus on efficiency of the front loading machine favored in Europe, which has a horizontal drum axis. This paper presents the control algorithm for optimal efficiency drives of an induction motor for drum washing machine. This system uses a simple model of the induction motor that include equations of the iron losses. The proposed optimal efficiency control algorithm calculates commands of the reference torque and flux currents for the flux oriented control of the induction motor. The proposed algorithm is verified through digital simulation.

The purpose of this study is to develop the fuzzy logic RSC(Roll Stability Control) system to prevent the rollover for the SUV(sport utility vehicle). The SUV model used in this study is the 8-DOF model considering the longitudinal, lateral, yaw and roll motions. The longitudinal and transversal weight transfers are considered in the computation of the vertical forces acting on a wheel. The engine torque is obtained from the throttle position and the r.p.m. of the engine map. The fuzzy logic controller input consists of the roll angle error and its derivative. The output is the brake torque and the throttle angle. The engine torque controller controls the throttle valve angle. The brake controller independently controls both right and left wheels. When the roll angle is ±4.5° defined as the critical roll angle, the front inner tire experiences the 1/100 ~ 1/50 of the total vertical forces, and the rollover starts. To prevent the rollover in advance, the target angle ±4.5° is adopted to control the vehicle stability. The RSC system begins operating at ±4.5° and stops at 0°. The simulations are conducted to evaluate the controller performance at right turns for the excessive steering angle. When the roll angle error and its derivative exceed the limited point, the RSC system makes the longitudinal velocity of the SUV decrease the brake torque and adjusts the throttle angle. The roll motion of the SUV is then stabilized.

A closed-loop sensorless stroke control system for a linear compressor has been designed. the motor parameters are identified as a function of the piston position and the motor current. they are stored in rom table and used later for the accurate estimation of piston position. some experimental results are given in order to show the feasibility of the proposed control schemes for linear compressors.

Tension control of loom system are usually achieved by using loadcell sensor and powder clutch, which require additional mounting space, reduce the reliability in harsh environments and increase the cost of a loom system. Moreover, the physical properties of textile fabrics are very sensitive to several factors(temperature, humidity, radius change of warp beam etc.) which result in tension change. In this paper, a novel sensorless tension control of a shuttleless loom system based on SVR(Support Vector Regression) is presented. The sensorless tension algorithm of shuttleless loom system driven by servo motor which is robust to disturbance and tension variation. First, the modeling and dynamic behaviors of a shuttleless loom system is described. Then, different tension control strategies are analyzed and discussed. And finally, the validity and the usefulness of proposed algorithm are thoroughly verified through numerical simulation.

The current control for interior permanent magnet(IPM) machines is more complicate than surface permanent magnet (SPM) machine because of its torque characteristic depending on the reluctance. For high performance torque control, it requires state decoupling between the d-current and q-current dynamics. However the variation of the inductances, which couples the state dynamics of the currents, makes the state decoupling difficult. So some decoupling methods have developed to cope this variations and each current can be regulated independently. This paper presents a novel approach for the decoupling of the states cross-coupling using sliding mode control. The sliding mode control method is based on the error between reference currents and the actual currents. The proposed method has the advantages of PI control performance and the sliding mode control robustness. Its first design step is to design PI controller, then the sliding mode control input term is added to it. This makes actual implementation of the controller easier.

In this paper, we propose a novel stability criterion for switched linear systems. The proposed method employs the results on the upper bound of the solution of LME (Lyapunov Matrix Equation) and on the stability of hybrid systems. The former guarantees the existence of Lyapunov-like energy functions and the latter shows that the stability of switched linear systems by using these energy functions. The proposed criterion provides us with easy implementation and less severe restriction than existing methods.

Much of the study on the controller that guarantees the stability and performance for the system under irregular disturbances has been carried out. Normally those irregular disturbances are regarded as noise having random characteristics and originated externally and internally or in combined manner. Those noises decrease the system stability and performance as well. In the paper only internal noises that come from elements such as resistor, diode and transistor etc. in the circuit system and numerical operation in the digital control are considered. "Dual Stochastic Controller" is proposed to reduce the random internal noises in the model-tracking controller. The proposed "Dual Stochastic Controller" is consisting of two controllers. One is stochastic controller to suppress internal random noise and the other one is model-tracking controller to follow the control law. Those internal random noises in the system are transformed into stochastic domain via F-P-K(Fokker-Planck-Kolmogorov) equation approach. Simulation reveals the performance of proposed "Dual Stochastic Controller" is quite promising.

The output delay is always existent in the practical systems. Analysis of the delay phenomenon in a continuous-time domain is sophisticated. It is appropriate to obtain its corresponding discrete-time model for implementation via digital computer. In this paper a new method for the discretization of nonlinear systems using Taylor series expansion and the zero-order hold assumption is proposed. This method is applied to the sampled-data representation of a nonlinear system with constant output time-delay. A performance of the proposed method is evaluated using a nonlinear system with time-delay output.

In this paper, we propose a new scheme for the discretization of nonlinear systems using Taylor series expansion and the zero-order hold assumption. This scheme is applied to the sample-data representation of a nonlinear system with constant state time-delay. The mathematical expressions of the discretization scheme are presented and the effect of the time-discretization method on key properties of nonlinear control system with state time-delay, such as equilibrium properties and asymptotic ability, is examined. The proposed scheme provides a finite-dimensional representation for nonlinear systems with state time-delay enabling existing controller design techniques to be applied to them. The performance of the proposed discretization procedure is evaluated using a nonlinear system.

Designing of the control system for synchronous generator is difficult task. The problem is that the generators usually work under uncertain load. The generator model is unknown. In this situation the adaptive control approach is accepted. In this paper, a new methodology to design an adaptive control system was propose. The controller was build based on RBF network, which approximates the optimal control law found by solving Euler's equation. The parameters of the controller were correcting on-line by an adaptive law. The second Lyapunov method was used to proof the stability of the control system. Simulation results show that the control system has a good quality.

Unified power flow controller (UPFC) consists two converters. There are three purposes of this paper, firstly to illustrate the UPFC device based VSC designs, then to describe a decoupling method the UPFC's controller into two separate control systems of the shunt and the series converters respectively in realizing an appropriate coordination between them. Finally, using the Matlab tool to build a discrete simulator for the UPFC with 12 pulse converters. The simulation results show that the developed UPFC model is reflected the static and dynamic characteristics of the UPFC. The harmonics of the output of the model were analyzed. Using the simple power system with UPFC as an example, the dynamics characteristics were studied. The fault status of the system with UPFC was analyzed too.

The object of this research was to compare the performances of the active suspension system with three different control strategies. At first, a two degree-of-freedom vehicle model was established in order to study the basis feature of the active suspension system. Then three different control strategies including fuzzy control, optimal control and fuzzy-PID control were designed for the model. Each control strategy has its own characteristics. Therefore, their advantages and disadvantages were discussed in details. The simulation results show that the performances of the active suspension with three different control strategies are improved in riding comfort obviously and the fuzzy-PID control takes the advantages of both fuzzy control and PID control. It can be concluded that the fuzzy-PID control is a useful strategy for the active suspension system.

Improving the starting characteristics of a motor is an important part of the motor control. An intelligent soft starting technique was adopted in the starter and used in the present study because of its many advantages compared with conventional starting processes. The core of the soft starter was a single chip (Atmel 8098), its soul was the software and its control object was a Silicon Controlled Rectifier (SCR). The starter achieved not only current-limit starting, but also closed-loop control with a stator current detection circuit. In conclusion, as a result of digital control, starting characteristic can be conveniently chosen according to the load. In addition the starter is of small size, and starting is smooth and reliable due to current feedback.

Performance analysis can help us reduce requests' congestions in a workflow network, cut servers' cost or strike a best balance between them. Analysis for requests' remaining time at a workflow network plays an important role in the per-formance analysis for time-constrained workflow. Because requests' remaining time has complex relations with the arrival rate of users' service requests, the service rate of each server for each activity, servers' number and the structure of the workflow network, it is difficult to give a quantificational analysis for requests' remaining time. We develop a method to determine probability density function of requests' remaining time at a workflow network so that we will know the accurate proportion of requests can be executed without delay. An experiment illustrates our method can be effectively utilized in practice.

With the rapid development of the Internet/Intranet, distributed databases have become a broadly used environment in various areas. It is a critical task to mine association rules in distributed databases. The algorithms of distributed mining association rules can be divided into two classes. One is a DD algorithm, and another is a CD algorithm. A DD algorithm focuses on data partition optimization so as to enhance the efficiency. A CD algorithm, on the other hand, considers a setting where the data is arbitrarily partitioned horizontally among the parties to begin with, and focuses on parallelizing the communication. A DD algorithm is not always applicable, however, at the time the data is generated, it is often already partitioned. In many cases, it cannot be gathered and repartitioned for reasons of security and secrecy, cost transmission, or sheer efficiency. A CD algorithm may be a more appealing solution for systems which are naturally distributed over large expenses, such as stock exchange and credit card systems. An FDM algorithm provides enhancement to CD algorithm. However, CD and FDM algorithms are both based on net-structure and executing in non-shareable resources. In practical applications, however, distributed databases often are star-structured. This paper proposes an algorithm based on star-structure networks, which are more practical in application, have lower maintenance costs and which are more practical in the construction of the networks. In addition, the algorithm provides high efficiency in communication and good extension in parallel computation.

A fuzzy controller based on analytic rules, which can self-adjust the fuzzy rules online, has good performance. That can change the output of the controller by modifying the fuzzy rules. An improved structure of a fuzzy controller based on analytic rules was proposed, and a modifying function aiming to regulate the fuzzy rules dynamically was introduced. The novel approach can effectively alleviate the contradictions between speediness and overshoot. Moreover, the genetic algorithm was applied to optimize five parameters of the fuzzy controller simultaneously. The steps required in seeking optimized parameters are presented. Simulation was conducted to show the efficiency of the proposed approach.

Based on the study of individual spatial behavior, the authors developed a tourists' spatial behaviors simulator (TSBS). TSBS employs Geographic Information System (GIS) to manage the spatial data, and Multi-Agent system (MAS) to simulate tourists' behaviors. By utilizing TSBS, tourists' travel patterns such as location, cost, and state can be analyzed and predicted. So TSBS can be an efficient tool for the management and planning of tourist resorts. TSBS is composed of four modules: data customizing, database, agent computation, and simulation results analysis. By use of a data customizing module, the user can customize the simulation content based on the realities. The database of TSBS is composed of spatial database, knowledge base and Agent status database. The agent computation module calculates tourists' speed, location and attributes at every clock step. By utilizing the simulation results analyzing module, the user can assess the simulation results by different indexes, or visualize the simulation process by means of a graphical interface. Based on a Client/server computation model, the local area network, even the internet, can be used as a computing platform for TSBS. The results obtained show that TSBS will be an effective tool to analyze tourists' spatial behaviors in a tourist resort.

Based on image sequences analysis, a new automatic counting method of passengers getting on and off a bus is proposed. In particular, it addresses the problem of counting when bus is crowded and illumination changes. By extracting and tracking passengers' heads appearing in image sequences and their moving direction, passengers boarding and alighting a bus are counted respectively. The multi-objects recognition and tracking is the key task of passenger counting. After foreground is separated from background subtraction, true objects are extracted through analysis on connected binary foreground image and their features are also extracted. Then, block matching algorithm is presented for the tracking multi-objects represented by regions and moving direction is judged according the result of matching. The experiment results show that the proposed method can realize counting passenger getting on and off a bus in high density correctly and improve the accuracy of counting.

This paper discusses the faults diagnosis of diesel engine systems. This research aims at the optimization of the diagnosis results. Inspired by Bayesian Network (BN) possessing good performance in solving uncertainty problems, a new method was proposed for establishing a BN of diesel engine faults quickly, and diagnosing faults exactly. This method consisted of two stages,namely the establishment of a BN model, and a faults diagnosis of the diesel engine system using that BN mode. For the purpose of establishing the BN, a new algorithm, which can establish a BN quickly and easily, is presented. The Fault Tree (FT) diagnosis model of the diesel engine system was established first. Then it was transformed it into a BN by using our algorithm. Finally, the BN was used to diagnose the faults of a diesel engine system. Experimental results show that the diagnosis speed is increased and the accuracy is improved.

Networked control systems are one type of distributed real-time control systems where sensors, actuator and controllers are interconnected by communication networks. In networked control systems, the insertion of the communication network makes the analysis and design of a networked control system more complex. In the past, control system design and CPU scheduling or network scheduling design have normally been separated. But from the point of both the communication protocols and the interacting controlled system, they should not be treated as separate. Thus, a co-design approach is proposed. Firstly, transmission model for the networked control systems is defined when a set of networked control systems is connected to the network, the Rate Monotonic (RM) scheduling algorithm and the Earliest Deadline First (EDF) scheduling algorithm are introduced. Then, applying EDF scheduling algorithm with the Integral of the Absolute Error (IAE) to schedule a set of real-time networked control systems. Finally, a simulation example is given to implement EDF-IAE performance protocol. The simulation results show that EDF-IAE performance protocol is feasible.

Magnetic-bearings, which support shafts with magnetic levitation rather than mechanical contact, have been in industrial use for decades. Recent technological developments, especially in digital processing and control, have made magnetic bearings a more-robust and cost-effective design solution than ever. The dynamic characteristic of electromagnetic bearing depends upon adopted controller; the active control can makes the electromagnetic bearings to realize complex control and special control. With the development of signal processing technology and modern control theory, the main parts of the control system are the digital signal-processing (DSP) electronics, a power supply, and amplifiers. An Active Magnetic Bearing (AMB) controller is mainly discussed in the paper, which is to be solved to realize this flexible control by hardware design based on DSP using TMS320C32 processor. It is proved by experiment that this kind of controller can optimize for this system, improve its stability and also have a very important referential value on the further study of AMB system.

An AC motor belongs to the category of a controlled object that is multi-variable, nonlinear and strong correlation, complex to mathematical model, and whose control performance is affected by a time-changing parameter. Therefore, it is very difficult to obtain the desired static and dynamic characteristic through a general fixed regulator. In this paper, the authors present a compound intelligent control strategy, combined with a neural network and fuzzy control. Considering that a neural network is good at self-learning, and a single fuzzy control algorithm is rapid in its response characteristics, the compound control strategy can compensate for a disadvantage of fuzzy control, which is associated with poor stability and precision and also requires solving a puzzle in the time-changing parameters in the controlled object. On the basis of a dynamic model of the permanent magnetic synchronous motor and its working principle, the authors designed the block diagram of a control system, combined a neural PID control and fuzzy control, and studied the corresponding control algorithm in detail. The simulation results show that the compound intelligent control system is good in dynamic performance and robustness.

A high-precision energized power supply was designed and developed based on the thought of expert self-tuning control, which was a combination of expert system and conventional PID (Proportional-Integral-Derivative) control. During the control process, the algorithms of increment PID and hierarchical integral were adopted to eliminate the phenomenon of integral saturation; and expert tuning rules were used to implement parameters trimming for the higher precision of the system. The advantages of the expert system and the conventional PID control were incorporated in the expert PID control system, namely the expert experience and the adaptive characteristics were sufficiently applied for the control performance of human-like empirical techniques. Therefore, higher steady-state precision and better dynamic response performance could be obtained. Experimental results show that the output voltage precision of the power supply is ±0.05%, much higher than the ones based on conventional PID control.

In the process of road surface construction, dispatchers' scheduling was experiential and blindfold in some degree and static scheduling restricted the continuity of the construction. Serious problems such as labor holdup, material awaiting and scheduling delay could occur when the old scheduling technique was used. This paper presents ant colony algorithm based on MAS that has the abilities of intelligentized modeling and dynamic scheduling. MAS model deals with single agent's communication and corresponding in engineering machinery of clusters firstly, next we apply ant colony algorithm to solve dynamic scheduling in the plant. Ant colony algorithm can optimize the match of agents and make the system dynamic balance. The effectiveness of the proposed method is demonstrated with MATLAB simulations.

The fault self-diagnosis system is an important component of an the automotive electronic control system. Designers of automotive electronic control systems urgently require or need a complete understanding of the self-diagnosis designing method of the control system in order to apply it in practice. Aiming at this exigent need, self-diagnosis methods of designing sensors, electronic control unit (ECU), and actuators, which are the three main parts of automotive electronic control systems, are discussed in this paper. According to the fault types and characteristics of commonly used sensors, self-diagnosis designing methods of the sensors are discussed. Then fault diagnosis techniques of sensors utilizing signal detection and analytical redundancy are analysed and summarized respectively, from the viewpoint of the self-diagnosis designing method. Also, problems about failure self-diagnosis of ECU are analyzed here. For different fault types of an ECU, setting up a circuit monitoring method and a self-detection method of the hardware circuit are adopted respectively. Using these two methods mentioned above, a real-time and on-line technique of failure self-diagnosis is presented. Furthermore, the failure self-diagnosis design method of ECU are summarized. Finally, common faults of actuators are analyzed and the general design method of the failure self-diagnosis system is presented. It is suggested that self-diagnosis design methods relative to the failure of automotive electronic control systems can offer a useful approach to designers of control systems.

Integration of process planning with scheduling by considering the manufacturing system's capacity, cost and capacity in its workshop is a critical issue. The concurrency between them can also eliminate the redundant process and optimize the entire production cycle, but most integrated process planning and scheduling methods only consider the time aspects of the alternative machines when constructing schedules. In this paper, a fuzzy inference system (FIS) in choosing alternative machines for integrated process planning and scheduling of a job shop manufacturing system is presented. Instead of choosing alternative machines randomly, machines are being selected based on the machines reliability. The mean time to failure (MTF) values is input in a fuzzy inference mechanism, which outputs the machine reliability. The machine is then being penalized based on the fuzzy output. The most reliable machine will have the higher priority to be chosen. In order to overcome the problem of un-utilization machines, sometimes faced by unreliable machine, the particle swarm optimization (PSO) have been used to balance the load for all the machines. Simulation study shows that the system can be used as an alternative way of choosing machines in integrated process planning and scheduling.

To reduce the Design time of a Lamp product, the development of a case-based reasoning (CBR) system for lamp modeling design is dealt with. One effective way to design a new part is to retrieve a relevant case of exacting similar to a new desired part and then adapt the retrieved case to meet the new requirements. A CBR mechanism for retrieval of lamp part is presented. Its kernel contains: (1) a object-oriented representation of a part; (2) indexing of a part; (3) a feature hierarchical structure based on lamp modeling; (4) a similarity metric to measure the similarity between a new desired part and any existing part in the case base. (5) Adaptation/Evaluation the solution. A prototype system has been developed to verify the usefulness of CBR in lamp modeling design.

Defects on the surface of hot rolled steel strips are main factors to evaluate quality of steel strips, an improved image recognition algorithm are used to extract the feature of Defects on the surface of steel strips. Base on the Machine vision and Artificial Neural Networks, establish a defect recognition method to select defect on the surface of steel strips. Base on these research. A surface inspection system and advanced algorithms for image processing to hot rolled strips is developed. Preparing two different fashion to lighting, adopting line blast vidicon of CCD on the surface steel strips on-line. Opening up capacity-diagnose-system with level the surface of steel strips on line, toward the above and undersurface of steel strips with ferric oxide, injure, stamp etc of defects on the surface to analyze and estimate. Miscarriage of justice and alternate of justice rate not preponderate over 5%.Geting hold of applications on some big enterprises of steel at home. Experiment proved that this measure is feasible and effective.

The automatization in power plat will enter the era of Digitalization and Information from that of Distributed Control System (DCS) and Network with the realization of enterprise informatization as measure and improvement of enterprise profit as the destination. Maintenance Based on Condition (MBC) is analyzed to be more rational than the traditional MBC, which can reduce the maintenance cost, prolong the life of device and improve the reliability and available rate of the unit. The intelligent instrument based on fieldbus is also discussed to be more intellectualized than the traditional one, such as self-calibration, self-compensation, self-validation, etc, which has more powerful function and can provide more accurate and sufficient information and data that can reflect the state of process and instrument itself. The MBC based on fieldbus and intelligent instrument represents the tendency of MBC in power plant.

The super saturation of magma, a main parameter in the boiling process control, usually is measured by electrical conductivity, but the output of sensor is influenced with the pure degree and temperature of magma. To eliminate that interference, the output would be done with by data fusion method based on the neural network. Using the measured value of electrical-conductivity sensor, syrup's Brix and temperature as the inputs, the network is trained by BP algorithm to achieve the fused output. The results of test shows that the data fusion method based on the artificial neural network could effectively eliminate the change of Brix and temperature has influenced on the supersaturation, and then could attain precise and stable output value.

Hydrostatic transmission with secondary regulation is a new kind of hydrostatic transmission that can regenerate inertial and gravitational energy of load. On the basis of an in-depth analysis of the working principles and energy-saving mechanisms of the parking equipment lifting systems with, secondary regulating technology, this paper proposes a method of regenerating a lifting system's inertial energy by controlling rotational speed and reclaiming the gravitational energy by use of a constant rotational speed. Considering large changes of the parameters of lifting systems and then non-linearity, a fuzzy control was adopted to control the lifting system, and a mathematical model of the system was established. By simulation and experiment of the lifting system, the conclusion was reached a lifting system's braking achieved by controlling rotational speed is reliable and stable at a definite speed. It also permits changing the efficiency of recovery by changing the rotational speed. The design power of the lifting system can be chosen in terms of the system's average power, so the system's power can be reduced and energy savings achieved.

To improving unilateral defect in CAE technology of injection molding, the paper advances collaborative design method based on knowledge and numerical simulation case for injection molding design. According to the application of feature technique to the modeling of injected part and injection molding more and more comprehensively, this paper puts forward a knowledge model of injection molding design based on feature, which is called FEATURE, and FEATURE is also applied to express the model of CAE numerical simulation case successfully. The collaborative reasoning model includes case-based reasoning (CBR), rule-based reasoning (RBR) and the both collaborative mechanism together, which includes the information exchange during the reasoning and the optimized choice of the reasoning result. The study shows that the technique of collaborative design based on knowledge and numerical simulation case has achieved effectively the design reuse of CAE results and shortened the design cycle of injection molding.

The vibration equation of piezo-smart beams were proposed with Heviside function and modal analysis method, then, the problem of active vibration control and suppression of piezo-smart beams were investigated under modal negative velocity feedback control law. It is concluded that triangular progression of Heviside function can make vibration equation of smart beams simple. There is no direct relationship between calculating modals' number and amount of PZT patches, modal negative velocity feedback increases modal damping of smart beams to suppress structure vibration. At last, numerical example is given to illustrate the validity of the method proposed in the paper.

The design method of a control system for weighing-distribution equipment was introduced, with detailed description of hardware connection and software programming. A special chip AD7730, high-precision bridge transducer A/D converter, was used to simplify task in hardware circuit. Simultaneously, array management method in hierarchical software makes it convenient to process numerous parameters and run all kinds of test programs.

With the development of the internet and the application of databases, seas of storage of data have become available. How to use the data for humans is the task of data mining. But in the process of data mining, a problem often encountered is that mining association rules on the quantitative attributes in RDBMS (Relational Database Management System) or Web logs. A genetic algorithm is proposed in the present paper to solve the clustering problem which can be solved by FCM (Fuzzy Clustering Method), so as to avoid the local optimization that often occurs in FCM. The quantitative attributes can be converted into categorical attributes, and then the categorical attributes are mapped into Boolean attributes, so that many association algorithms can be used to mine significant association rules.

Wavelet neural network was used in the area of curve grinding. The prediction model of surface machining quality in curve grinding based on wavelet neural network was founded. The work piece feed amount, rotation speed of grinding wheel and vibration frequencies were chosen as input variables of wavelet neural network. The roughness was used to assess grinding surface quality. Prediction results were feedback to adjust machining parameters. In order to solve disadvantages of "dimension disaster", slow rate of convergence and easily falling into local minimum point caused by multi-input and output. A new local evolutionary algorithm was used to train wavelet neural network. From some experiments, it can be seen that this method increase rate of convergence effectively. The surface quality of curve grinding process can be obtained.

Based on Fuzzy controller and expert system (ES), a real-time control system is proposed for improving the stability of the weld pool width in aluminum alloy metal inert gas (MIG) welding. A vision sensing system for taking the image of pool of aluminum alloy has been setup and corresponding image-processing algorithm has been developed to acquire characteristic parameters of the weld pool. The experiments show that the real-time and precision requirements for detecting and control of weld pool width of aluminum alloy MIG welding process could be satisfied by the established system.

A compromised color management method was proposed that was modeled after visual perception instead of conventional color management method that using spectrum analysis. The proposed scheme used a computer color scanner to obtain a bit map image from dyed original color sample, and conduct pixel analysis using an image histogram. From the image histogram, it can extract the color and RGB value that shows the dominant pixel distribution of the sample image. An RGB color model was constructed for dyestuff manufacturing with a rule-based algorithm. Constructed color model can minimize additional experiment to obtain more color and produce the color recipe information from basic data. And performance of color matching was measured as a Euclidean distance that compares the RGB value of the original color with RGB value of the produced color. The matching range was confirmed by threshold of Euclidean distance.

In this paper a magnetic flywheel system is studied with a magnetic bearing, which is able to support the shaft without mechanical contacts, and it is also able to control the rotational vibration. Magnetic flywheel system is composed of position sensors, a digital controller, actuating amplifiers, electromagnets and a flywheel. This work applies the fuzzy neural network (FNN) algorithm to control the vibration of a magnetic flywheel system. It proposes the design skill of an optimal controller when the system has the uncertainty, i.e. it has a difficulty in extracting the exact mathematical expressions. Two controllers are designed for the FNN in order to reduce the rotor vibration effectively. Unbalance response, which is a serious problem in rotating machineries, is improved by using a magnetic bearing with a FNN algorithm.

Data storage related with writing and retrieving requires high storage capacity, fast transfer rate and less access time. Today any data storage system cannot satisfy these conditions, however holographic data storage system can perform faster data transfer rate because it is a page oriented memory system using volume hologram in writing and retrieving data. System can be constructed without mechanical actuating part therefore fast data transfer rate and high storage capacity about 1Tb/cm³ can be realized. In this paper, to reduce errors of binary data stored in holographic data storage system, a new method for bit error reduction is suggested. First, find cluster centers using subtractive clustering algorithm then reduce intensities of pixels around cluster centers and fuzzy rules. Therefore, By using this error reduction method following results are obtained; the effect of Inter Pixel Interference noise is decreased and the intensity profile of data page becomes uniform therefore the better data storage system can be constructed.

A full fuzzy-logic-based vector control for a permanent-magnet synchronous motor(PMSM) was proposed. The high-performance of the proposed fuzzy logic control(FLC)-based PMSM drive were investigated and compared to those obtained from the conventional proportional-integral(PI) controller at different conditions such as step change in command speed and load, etc. In the simulation and experimental results, the fuzzy-logic controller was employed in the speed and current controller. The comparative experimental results show that the FLC is more robust and, hence, found to be a suitable replacement of the conventional PI controller for the high-performance drive system.

This paper deals with development of a kinematics model, a trajectory tracking, and a controller of fuzzy-genetics algorithm for 2-DOF Wheeled Mobile Robot (WMR). The global inputs to the WMR are a reference position, P_r= (x_r,y_r,θ_r)^t and a reference velocity q_r=(v_r,ω_r) ^t, which are time variables. The global output of WMR is a current posture P_c= (x_c,y_c,θ_c)^t. The position of WMR is estimated by dead-reckoning algorithm. Dead-reckoning algorithm can determine present position of WMR in real time by adding up the increased position data to the previous one in sampling period. The tracking controller makes position error to be converged 0. In order to reduce position error, a compensation velocities q=(v,ω)^t on the track of trajectory is necessary. Therefore, a controller using fuzzy-genetic algorithm is proposed to give velocity compensation in this system. Input variables of two fuzzy logic controllers (FLCs) are position errors in every sampling time. The output values of FLCs are compensation velocities. Genetic algorithms (GAs) are implemented to adjust the output gain of fuzzy logic. The computer simulation is performed to get the result of trajectory tracking and to prove efficiency of proposed controller.

This paper proposes a population structure of random signal-based learning (PRSL), merged with simulated annealing (SA), to optimize the fuzzy logic controller (FLC). Random signal-based learning (RSL) exploits (local search) the search space very well, but it can not explore (global search) the search space because of its serial nature. To overcome these difficulties, PRSL, which consists of serial RSL as a population, was considered. Moreover, SA was added to RSL to help the exploration. The validity of the proposed algorithm was conformed by applying it to the optimization of a FLC for the inverted pendulum.

For an intelligent vehicle driving which uses magnetic markers and magnetic sensors, it can get every kind of road information while moving the vehicle if we use the code that is encoded with N, S pole direction of makers. If there make it an only aim to move the vehicle, it becomes easy to control the vehicle the more we put markers close. By the way, to recognize the direction of a marker pole it is much better that the markers have no interference each other. To get road information and move the vehicle autonomously, the method of arranging magnetic sensors and algorithm of recognizing the position of the vehicle with those sensors was proposed. The effectiveness of the methods was verified with computer simulation.

The rotor position and speed estimation of permanent-magnet synchronous motor(PMSM) was dealt with. By measuring the phase voltages and currents of the PMSM drive, two diagonally recurrent neural network(DRNN) based observers, a neural current observer and a neural velocity observer were developed. DRNN which has self-feedback of the hidden neurons ensures that the outputs of DRNN contain the whole past information of the system even if the inputs of DRNN are only the present states and inputs of the system. Thus the structure of DRNN may be simpler than that of feedforward and fully recurrent neural networks. If the backpropagation method was used for the training of the DRNN the problem of slow convergence arise. In order to reduce this problem, recursive prediction error(RPE) based learning method for the DRNN was presented. The simulation results show that the proposed approach gives a good estimation of rotor speed and position, and RPE based training has requires a shorter computation time compared to backpropagation based training.

The hybrid neural-genetic multi-model parameter estimation algorithm was demonstrated. This method can be applied to structured system identification of electro-hydraulic servo system. This algorithms consist of a recurrent incremental credit assignment(ICRA) neural network and a genetic algorithm. The ICRA neural network evaluates each member of a generation of model and genetic algorithm produces new generation of model. To evaluate the proposed method, electro-hydraulic servo system was designed and manufactured. The experiment was carried out to figure out the hybrid neural-genetic multi-model parameter estimation algorithm. As a result, the dynamic characteristics were obtained such as the parameters(mass, damping coefficient, bulk modulus, spring coefficient), which minimize total square error. The result of this study can be applied to hydraulic systems in industrial fields.

Recently, the Korean Railroad took a new stage with the opening of KTX (Korea Train Express). Korean National Railroad was planning using the on-board signaling system for putting KTX and existing equipment to a good. Provided a on-board signal mode was used, it can improve headway more than the classical ground signal mode but there can be places where the improvement of headway is impossible in a certain section because of physical constraints. So a new algorithm for train separation control is necessary to balise. Train separation control system by movement authority and balise does not depend on conventional track circuit. Therefore, position and distance control of train to prevent collision with leading train may become important safety factor. This paper suggests an investigation of train separation control algorithm using movement authority of several units efficiently for this. It is come up with an algorithm for this situation.

Unmanned aerial vehicle (UAV) was designed to fulfill its mission autonomously. Its core technology for successful development of the UAV is to design of the flight control system (FCS) consisting of the flight control computer, the actuator, the sensor in order to replace the pilot. Each development procedure of FCS involves a verifying test for each step and was completed by an evaluation test of the comprehensive performance verification for the overall systems. The development tool, which is applied to this objective, is the HILS (Hardware-In-the-Loop-Simulation). That is, HILS is a cost-effective method for testing complex, flight-critical hardware before it is used in the real world. The real-time HILS is shown to develop/verify the performance of SURV UAV's controller in the lateral/directional control systems.

Tracking position control of a linear magnetic actuator implemented on distributed control network was studied. In the distributed control system involving a discrete-time controller and a linear magnetic actuator, time delay happens as the controller reads the sensor data or sends out the control input to the actuator located at a remote site. In many cases the time delay is not ignorable, and has significant effect on the control system performance. In order to compensate for the adverse effect of the time delay, the modified Smith predictor was introduced to the control system. The results of simulation and experiment show that the performance of linear magnetic actuator can be improved by augmenting the modified Smith predictor onto the existing control system. With the modified Smith predictor, the position error is reduced and the constant disturbance is suppressed.

Early detection and diagnosis of induction machine incipient faults are desirable for increasing machinery availability, reducing consequential damage, and improving operational efficiency. However, fault detection using analytical method is not always possible because it requires perfect knowledge of a process model. A neural network based expert system was proposed for diagnostic problems of the induction motors using vibration analysis. The short-time Fourier transform (STFT) was used to process the quasi-steady vibration signals, and the neural network was trained and tested using the vibration spectra. The efficiency of the developed neural network expert system was evaluated. The obtained results lead to a conclusion that neural network expert system can be developed based on vibration measurements acquired online from the machine.

An analog compound orthogonal neural network is presented which is based on digital compound orthogonal neural networks. The compound neural network's control performance was investigated as applied to a robot control problem. The analog neural network is a Chebyshev neural network with a high speed-learning rate in an on-line manner. Its control algorithm does not relate to controlled plant models. The analog neural network is used as the feedforward controller, and PD is used as the feedback controller in the control system of robots. The excellent performance in system response, tracking accuracy, and robustness was verified through a simulation experiment applied to a robotic manipulator with friction and nonlinear disturbances. The trajectory tracking control showed results in satisfactory effectiveness. This analog neural controller provides a novel approach for the control of uncertain or unknown systems.

The robotics soccer game is a standard task platform, the soccer robot is one form of the artificial life body. Based on this platform, in terms of the action-selection model of artificial life, carries a comparable and valid study. Firstly, this paper analyses the action layer and the action type of the soccer robot. Then, according to the behavior selection model of artificial life body based on priority degree, improves the action selection model of the soccer robot. Finally, builds the decision-making system of soccer robot based on the model, and this system has been applied in the actual games of soccer robot, the result indicates that it is workable and efficient.

Based on the analogies between multi autonomous robots system (MARS) and immune system, a synthesized immune network is proposed, and used to solve the navigation and coordination problem on MARS. Individual robot was regarded as small-scaled immune networks (SN). Task was regarded as antigen, and behavior tactics were deemed to the antibodies respectively. Behavior tactic to a robot sensor was taken as B cell. Navigation and coordination problem is transformed into the interaction mechanism among antibody, antigen and small-scaled immune networks. The pursuit problem was used to validate the hypothesis. Simulation results suggest that the proposal is promising.

A new type of mobile robot that is used to search and locate missing miners after mine disasters is designed according to the complex environment underground mine. The general structure and main parameters of the Missing Miner Searching Robot (MMSR) are presented. Four tracked units drive MMSR under independent control and cooperative driving mode. Each tracked unit is a leg-tracked function moving mechanism. The moving and legged rotating of a unit are respectively controlled by two motors. Surmounting obstacles capability, steering performance, toppling stability and kinematical reconfigurability of the MMSR are analyzed. The results show that MMSR adapted itself to terrains and topographic environment underground mine. It could directly surmount a raceway with width or an obstacle with height less than one tracked unit length. It had recovery ability of moving function.

It is a current tendency of development in automation technology to replace manpower with manipulators in working places where dangerous, harmful, heavy or repetitive work is involved. The sort manipulator is installed in a crystal-angle sort machine to take the place of manpower, and engaged in unloading and sorting work. It is the outcome of combing together mechanism, electric transmission, and pneumatic element and micro-controller control. The step motor makes the sort manipulator operate precisely. The pneumatic elements make the sort manipulator be cleverer. Micro-controller's software bestows some simple artificial intelligence on the sort manipulator, so that it can precisely repeat its unloading and sorting work. The combination of manipulator's zero position and step motor counting control puts an end to accumulating error in long time operation. A sort manipulator's design in the practice engineering has been proved to be correct and reliable.

Robotic blimps present an enormous potential for applications in low-speed and low-altitude exploration, surveillance, and monitoring, as well as telecommunication relay platforms. To make our lighter-than-air platform a robotic blimp with significant levels of autonomy, the decoupled longitude and latitude dynamic model are developed, and the hardware and software of the flight control system are designed. The onboard hardware consists of blimp state observer, actuators, MCU, etc. The software functions include signals processing, data filtering and fault tolerance, ground command execution, etc. Based on decoupled dynamic model, the control system architecture is presented, and navigation strategy for waypoint flight problem is discussed. The paper gives results of a flight experiment using the designed flight control system, and the results manifests that the system is applicable and initial machine intelligence of robotic blimp is achieved.

In this paper, a novel 2-DOF high-speed/high-accuracy robotic system is proposed, which consists of two permanent linear DC servomotors and a planar parallel mechanism. Analysis of robotic dynamical model verifies that load variation is violent. Due to the direct-drive principle of a linear motor, the load disturbance can have a significant effect on position accuracy. For this purpose, a discrete-time disturbance observer, feedback controller and ZPECT feedforward controller is presented. The disturbance observer, which is applied to the plant velocity loop, provides robustness for design of both the feedback controller and the feedforward controller. The feedback controller is intended to make the command following properties fast, and the feedforward term realized as a Zero Phase Error Tracking Controller improves tracking. In the experiment, robotic max acceleration is equal to 20m/s2, max velocity is more than 1 m/s, and resolution is 0.5 μm. While the tracking errors are less than 50 μm, the steady-state errors are less than 1 μm, and steady time is less then 42 ms. Experiment results show that: high-performance motion system can be obtained with this discrete-time controller, even in large load variance motion system.

The scarcity of the proper tool for fast developing intelligent toy is the one of problem that affects developing the top grade toys of our country. This paper presents a method that is based on the tree structure for developing intelligent toys, which is supported by a friendly man-machine interface, is able to realize restructuring and modularizing design of the program, debugging and controlling intelligent toy smartly, and to avoid some mistakes of a program. This method with the help of Problem Analysis Diagram (PAD) transforms a programming task of the intelligent toy into the tree structure, translates and executes every node of the tree structure. The corresponding commands and data can be transmitted to the controller of an intelligent toy, also be generated into a downloaded program for the chip microprocessor.

On the basis of the research on nanometer drive parts under large feeding force and the thermal expansion principle, the nano precision drive parts based on the semiconductor refrigeration equipment was designed. In order to get a better feed precision, considering that the nano feeding under large feeding force has some open air temperature random disturbance phenomenon which influence the feed precision, LMI-based robust H∞ control scheme was presented. And the LMI-based robust H∞ control scheme used here can realize the purpose of obtaining a good control of nano feeding under large feeding to enhance the response speed and the robustness. When the open air temperature changes within 0.3°C, the control precision without LMI-based robust H∞ control scheme is 15 nm and the control precision under LMI-based robust H∞ control scheme has arrived within 10 nm. This proves the LMI-based robust H∞ control feasible and usable.

Developed the foot mechanism, which is one of the key parts of three-limb permanent magnetic adsorption robot HIT-RMT. With this foot mechanism, the limbs of the robot can be transformed into adsorption feet to walk when needed, as well as operators to carry out assignments. Three-dimension force/moment sensor has been employed in the design of the foot, which is used to detect force information between the robot and the walking surface or operating objects. One permanent magnetic adsorption mechanism has been designed, and with it, the robot is easy to realize the adsorption and departure with the working surface with comparatively small energy consuming and ensured safety.

"Tooth-arrangement Three-fingered Hands" is a brand-new oral cavity repairing robot. It has the functions of fine grasping and arranging artificial tooth, manufacturing complete denture, etc. A new optimization design method is proposed in this paper that solves the problems in optimization structure parameter confirming and optimal dexterous design. The bionic theorem behind duplicating the dexterity of the human finger is incorporated into the structure parameter optimization algorithm. Through analyzing the dexterity of the single finger, adopting the design criteria of dexterous degree with optimum index and combining the characteristic of grasping the smallest object, this design method can confirm the best dexterity area for single finger, relations between every rod length and rotation range of every joint. Using MATLAB optimization toolbox to optimize above-mentioned structure parameter, it gains the optimal dimension that meets characteristic of human fingers and can finish grasping. A quantitative method is also proposed in this paper to calculate the relative position between fingers. It is applied to the real design of Tooth-Arrangement Three-fingered Hands and obtained optimal flexible performance.

The design of a kind of pneumatic squirming robot is presented. It is based on the use of flexible pneumatic actuator. The flexible pneumatic actuator was made of caoutchouc. Its working principle is described. The structure, working principle, pneumatic and electrical control system of the pneumatic squirming robot are designed. All of the actuator's driving and squirming parts are composed of pneumatic elements. The vacuum osculums, which act as feet, are connected to the flexible pneumatic actuator. When the vacuum pumps operate, vacuum will be produced in the corresponding osculums, which can adsorb on the contacting surface and orient the robot. The actuator, operating under air pressure, drives the robot. By controlling the vacuum pumps and the actuator, straight and bending squirming of this robot can be obtained.

A neuro-fuzzy network based approach for robot motion in an unknown environment was proposed. In order to control the robot motion in an unknown environment, the behavior of the robot was classified into moving to the goal and avoiding obstacles. Then, according to the dynamics of the robot and the behavior character of the robot in an unknown environment, fuzzy control rules were introduced to control the robot motion. At last, a 6-layer neuro-fuzzy network was designed to merge from what the robot sensed to robot motion control. After being trained, the network may be used for robot motion control. Simulation results show that the proposed approach is effective for robot motion control in unknown environment.

The architecture of ergonomics simulation and evaluation for the automobile was described. Ergonomics analysis and evaluation is one of the most important processes in product design at present. This ergonomics simulation system based on the elements of ergonomics analysis and evaluation can provide an excellent solution to take human element into account earlier in the design phase and make proactive choices in automobile design than those traditional methods. Thinking of the characteristics of the automobile industry, this system adopted the anatomy-based and parameterized human model for Chinese, the simulation technique using motion editing and the mathematical models of ergonomics to solve real ergonomic design problems in the design phases.

The automatic gelatinizing system described in this paper adopts a DRC robot controller and a programmable controller (PLC) as its core. The DRC controller receives signals from the glue gun, cylinder sensors on the worktable, arms' limit switches of the SCARA robot, control signals from PLC, and so on. After systematic calculation and analysis, the DRC drives the robot's motors of every axis, electromagnetic valves, indicator lights, ancillary components on the claw, etc. The PLC receives input signals from the robot controller and detection sensors to control the air cylinder of clamps, loading and unloading devices, support panel and electrical motors, and it can also control the motor of the feeding glue system precisely. An automatic off-line programming system is developed with Visual C++ 6.0 programmable language to solve problems in the course of applying automatic gelatinizing robot, by which the users can easily design their own gelatinizing schemes at their will according to different gelatinizing tasks.

Familiarity is the crucial requirement for the human symbiosis robot. This paper presents a new concept of a robot avatar to increase familiarity of a care service robot. The robot avatar is a small robot mounted on a main robot and equipped with minimum function to play some gestures during task execution of the main robot. By looking at the avatar, user feels as if it is controlling the main robot. A prototype of the avatar (CHIRIS) is developed and installed to a service robot IRIS. Several psychological tests about the impression of IRIS were carried out using video. Test results showed that CHIRIS is effective to give more familiar impression to the users. It is also shown that CHIRS is useful to announce IRIS's following behavior to the user preliminarily.

Safing of a robot system used for research and education has been carried out based on the international safety standards in the paper. The following two situations are considered 1) operation without contact to a user. 2) operation with physical contact to a user. By using a simple two-link robot manipulator system, the difficulty of the safing in mechanical parts is examined experimentally. Safety of electrical circuit is also carried out and it is evaluated by using Boolean algebra and experiments.

In the present day, there is demand for making robot being assistant in many kinds of cooperative tasks with human being, for example, in the medical service and welfare work. Such robot is supposed to be without sense of incongruity felt against human being and with the ability of executing smooth movement in human-robot cooperative task. In view of every motion of the human body being in naturally perfect smoothness, it is essential to analyze the human motional characteristics by introducing human model for establishing the law of control on the cooperative robot with human being. In this study, master-slave cooperation more appropriate for human-robot cooperative task was adopted. The experiment was carried out founded on the hypothesis that the human arm around the elbow joint was considered as a simple mass-spring-damper dynamic system with one end fixed. Upon main investigation into the slave's relaxed arm in clockwise motion, by virtue of examining the physical parameters in the mathematical model of the human arm, its impedance characteristics were recognized. It was found that as for human ordinary motion, the damping factor was low enough to be disregarded and the stiffness possessed individually linear property. And besides, the friction-like force was discovered to exist.

A control method of a robot manipulator using EMG (electromyogram) from human face is proposed. The robot has three-degree-of-freedom arm and an en effector on the tip of the robot. The amplitude of the EMG signal generated from masseter, which is the muscle of jaw, corresponds to the velocity of the robot hand. The difference of the EMG level between right and left masseter determines the direction of the movement of the hand. The open/close control of the robot hand is activated by detecting double clenches performed within 0.6 seconds. Optimal filtering is applied to the measured EMG to enable effortless robot task. Picking, transferring and placing of a brick object are performed and the effectiveness of the proposed method has been confirmed.

In this paper, we propose SSMs (Safety Service Manipulator system) which is developed for achieving high safety and dexterity. The SSMs combines four novel techniques, BRM (Base Redundant Machanism), APCS (Air Pressure Collision detection System), CTLB (Controllable Torque limiter in Base), and a base redundancy, to reduce harmful forces and to achieve high velocity, force, and failure avoidance ability on a service process. BRM is a safer mechanism than SRM (Serial Redundant Mechanism) on pushing force and impact force because it needs less torque for base motors and has globally small equivalent mass. APCS is a flexible whole-body tactile sensor, which is equipped on the surface of the arm-part of SSMs. APCS can reduce impact force effectively not only by giving contact information but also the flexibility itself. Besides, it gives enough space for a brake to decelerate impact velocity before crash. CTLB is a controllable torque limiter installed in the base of SSMs. CTLB enables the arm-part of SSMs to slip when excessive external force is applied to it, which brings the reduction of pushing force. A base redundancy which is provided by BRM can be utilized in control and motion planning to improve the safety and dexterity. We actually develop a prototype of SSMs and MiniSSMs, and execute some experiments to confirm the effectiveness of APCS and CTLB with the MiniSSMs.

This paper addresses the issue of dynamic parameter identification of the robot manipulator with servo-controlled electrical motors. An assumption is made that all kinematical parameters, such as link lengths, are known, and only dynamic parameters containing mass, moment of inertia, and their functions need to be identified. First, we derive dynamics of the robot arm with a linear form of the unknown dynamic parameters by taking dynamic characteristics of the motor and servo unit into consideration. Then, we implement the parameter identification approach to identify the unknown parameters with respect to individual link separately. A pseudo-inverse matrix is used for formulation of the parameter identification. The optimal solution is guaranteed in a sense of least-squares of the mean errors. A Direct Drive (DD) SCARA type industrial robot arm AdeptOne is used as an application example of the parameter identification. Simulations and experiments for both open loop and close loop controls are carried out. Comparison of the results confirms the correctness and usefulness of the parameter identification and the derived dynamic model.

Collaboration mainly consists of two tasks; one is each partner's task that is performed by the individual, the other is communication with each other. Both of them are very important objectives for all the collaboration support system. In this paper, a collaboration support system for 3D shape evaluation in virtual space is proposed on the basis of both studies in 3D shape evaluation and communication support in virtual space. The proposed system provides the two viewpoints for each task. One is the viewpoint of back side of user's own avatar for the smooth communication. The other is that of avatar's eye for 3D shape evaluation. Switching the viewpoints satisfies the task conditions for 3D shape evaluation and communication. The system basically consists of PC, HMD and magnetic sensors, and users can share the embodied interaction by observing interaction between their avatars in virtual space. However, the HMD and magnetic sensors, which are put on the users, would restrict the nonverbal communication. Then, we have tried to compensate the loss of nodding of partner's avatar by introducing the speech-driven embodied interactive actor InterActor. Sensory evaluation by paired comparison of 3D shapes in the collaborative situation in virtual space and in real space and the questionnaire are performed. The result demonstrates the effectiveness of InterActor's nodding in the collaborative situation.

A fault-tolerant dead reckoning system is presented for a modular vehicle, which consists of one chassis unit and several wheel units. The units locally estimate the vehicle position based on their own internal sensors. The local estimates are exchanged among the units via an inter-communication system, and they are fused in a decentralized manner. The units can then determine the vehicle position accurately. The decentralized dead reckoning algorithm is formulated based on the information filter and the covariance Intersection method. For enhancing the reliability of the dead reckoning a multi-model based fault detection and diagnosis (FDD) of the internal sensors is incorporated into the dead reckoning system. The units diagnose their sensors with the FDD system, and they apply only the normal sensors for the vehicle localization. In this paper two fault modes (hard fault and noise fault modes) of the sensors are handled; on the hard fault the sensor output is stuck at a constant value. On the noise fault it is disturbed by a large noise. The FDD algorithm is based on the variable structure interacting multiple-model estimator. The fault-tolerant dead reckoning algorithm was implemented on our indoor test-vehicle, which consists of one chassis unit and four wheel units. Experimental results show that our dead reckoning provided better localization accuracy than the conventional one (i.e., the dead reckoning without sensor FDD system) did even though the sensors partially failed.

This paper proposes a human body posture estimation method based on analysis of human silhouette and Kalman filter. The proposed method is based on both the heuristically extraction method of estimating the significant points of human body and the contour analysis of the human silhouette. The 2D coordinates of the human body's significant points, such as top of the head, and tips of feet, are located by applying the heuristically extraction method to the human silhouette, those of tips of hands are obtained by using the result of the contour analysis, and the joints of elbows and knees are estimated by introducing some heuristic rules to the contour image of the human silhouette. The estimated results are optimized and tracked by using Kalman filter. The proposed estimation method is implemented on a personal computer and runs in real-time. Experimental results show both the feasibility and the effectiveness of the proposed method for estimating human body postures.

A human-symbiotic robot needs to provide both safety for human and function of avoiding dameges for the objects around the robot along with task performing ability. It is, however, not easy to be compatible with each other. We suggest a two wheeled inverted pendulum-type mobile robot which consists of a dual-manipulator system, a body with two links and a middle joint, and an inverted pendulum-type mobile base. The following three tasks are assigned to the robot as its basic functions: handling a small object on a desk or a floor, carrying a heavy object, and sitting down and standing up. The goal of this research is to realize above tasks with taking account of safety for human and the task performance ability. This paper presents the concept of the research and the specific way to realize these tasks and shows the mechanical design of the robot.

In this paper, we propose a soft finger-joint mechanism that gives a robot mult-directional passive compliance and the ability of the estimation of external torque. In this joint, we utilize silicone gum cushions as elestic elements to transmit motion from an actuator to an output link. The cusion contributes to both the generation of passive compliant motion for external torque and the estimation of the external torque. Based on the proposed mechanism, we developed the metacarpal-phalangeal joint (MP joint) of the finger of a robot for human-care services. In addition, it is confirmed in several experiments that the proposed mechanism exhibits the expected characteristics.

The purpose of this paper was to present the study and design idea for entertainment robot with educational purpose (IRFEE). The robot has been designed for home life considering dependability and interaction. The developed robot has three objectives - 1. Develop autonomous robot, 2. Design robot considering mobility and robustness, 3. Develop robot interface and software considering entertainment and education functionalities. The autonomous navigation was implemented by active vision based SLAM and modified EPF algorithm. The two differential wheels, the pan-tilt were designed mobility and robustness and the exterior was designed considering esthetic element and minimizing interference. The speech and tracking algorithm provided the good interface with human. The image transfer and Internet site connection is needed for service of remote connection and educational purpose.

In this work, important components for a high-fidelity system bandwidth are force reflecting device and that all the computations including contact determination and response computation have to be performed in less than a millisecond. This paper described a force-reflecting device and an embedded controller. The realized force-reflecting device is based on a novel serial type mechanical structure, and features compactness, high sustained output force capability, low friction, zero backlash, and enough workspace. The embedded controller reduced software computational load via main processor and simplifies hardware strictures by the time-division control. The device was integrated with existing dynamic simulation algorithms running separate workstation, so that objects can be manipulated in real time and the corresponding forces felt back by the operator.

By using a new type of silver robot called SpiderBot^TM, posture balancing and muscle strengthening exercises were presented. Unlike mobile robots on a floor, SpiderBot^TM is conceptualized to generate full mobility in a room and to have higher payload while generating six degrees-of-freedom for assisting daily lives of the elderly and the handicapped. A preliminary prototype of SpiderBot^TM has been constructed by using incompletely restrained wire mechanism and simple control algorithms were implemented for two types of exercises for elderly people's health improvement. One is a posture balancing exercise that prevents from falling to any directions. And the other is muscle strengthening exercise that uses the SpiderBot^TM as a weight lifting mechanism. Preliminary experimental results show that the proposed new silver robot platform is used for such exercises.

This paper described a map-building for the mobile robot based on photo sensor system. The navigation module and Hough transform used to find line segments in sensor data were explained. A fuzzy inference engine was proposed for Wall-following algorithm, and the algorithm was used to gather the geometric scanning data of unknown environment. For map-building, Hough-Transform algorithm was developed. As inputs of the inference engine, emulated photo distance sensor values were used. Some experiments were illustrated for validity of the proposed algorithm.

Considering personal robot as a special class of autonomous distributed system, the several requirements have to be met. These are scalability, portability, heterogeneity and real-time. An architecture of Object-oriented Middleware for Robot's network Integration (OMRI) was proposed. The OMRI explores idea of component-based software engineering. Architecture of two main components was proposed. The first is network handling component and the second is application management component. Components are independent and communicate through defined interfaces. The features of OMRI are support of heterogeneous networks, handling applications according to the object-oriented paradigm and possibility for extensions by additional components. Middleware was implemented and ported onto different platforms.

Most of the recently developed robots are human friendly robots which imitate animals or humans such as entertainment robot, bio-mimetic robot and humanoid robot. Interest for these robots are being increased because the social trend is focused on health, welfare, and graying. Autonomous eating functionality is most unique and inherent behavior of pets and animals. Most of entertainment robots and pet robots make use of internal-type battery. Entertainment robots and pet robots with internal-type battery are not able to operate during charging the battery. Therefore, if a robot has an autonomous function for eating battery as its feeds, the robot is not only able to operate during recharging energy but also become more human friendly like pets. Here, a new autonomous eating mechanism was introduced for a biomimetic robot, called ELIRO-II(Eating LIzard RObot version 2). The ELIRO-II is able to find a food (a small battery), eat and evacuate by itself. This work describe sub-parts of the developed mechanism such as head-part, mouth-part, and stomach-part. In addition, control system of autonomous eating mechanism is described.

This paper presented a novel bio-mimetic quadruped walking robot with 2-DOF (Degree Of Freedom) waist joint, which connects the front and the rear parts of the body. The waist-jointed walking robot can guarantee more stable and more animal-like gait than that of a conventional single-rigid-body walking robot. The developed robot, called ELIRO-II (Eating LIzard RObot version 2), can bend its body from side to side by using 1-DOF passive waist joint while the legs is transferred, thereby increasing the stride and speed of the robot. In addition, ELIRO-II has one more active DOF to bend its body up and down, which increases the mobility in irregular terrain such as slope and stairs. We design the mechanical structure of the robot, which is small and light to have high mobility. This research described characteristics of the 2-DOF waists joint and leg mechanism as well as a hardware and software of the controller of ELIRO-II.

In recent years, many mobile service robots have been developed. These robots are different from industrial robots. Service robots were confronted to unexpected changes in the human environment. So many capabilities were needed to service mobile robot, for example, the capability to recognize people's face and voice, the capability to understand people's conversation, and the capability to express the robot's thinking etc. This research considered face detection, face tracking and face recognition from continuous camera image. For face detection module, it used CBCH algorithm using openCV library from Intel Corporation. For face tracking module, it used the fuzzy controller to control the pan-tilt camera movement smoothly with face detection result. A PCA-FX, which adds class information to PCA, was used for face recognition module. These three procedures were called face feature processor, which were implemented on mobile service robot OMR to verify.

Complex "lifelike" behaviors are composed of local interactions of individuals under fundamental rules of artificial life. In this paper, fundamental rules for cooperative group behaviors, "flocking" and "arrangement" of multiple autonomouse mobile robots were represented by a small number of fuzzy rules by Subtractive clustering algorithm and DNA coding method. Fuzzy rules in Sugeno type and their related parameters were automatically generated from clustering inputoutput data obtained from the algorithms for the group behaviors. Simulations demonstrate the fuzzy rules successfully realize group intelligence of mobile robots.

The 7G LTR (7th generation LCD Transfer Robot) is the robot that is capable of transport the 7th generation LCD glass (1870×2200mm). This paper described a specification and performance of the 7G-LTR. The robot needs precise path accuracy to improve stack efficiency of glass cassette which affects the productivity of LCD plant. This paper presented a method of improving the path accuracy by kinematic compensation for the static deflection. Firstly, the 7G-LTR and its requirements were introduced. Next, the robot arm's kinematic model including the static deflection and the complete kinematic analysis were presented. Finally, the method of improving the path accuracy was provided and its optimization problem was solved. Various tests have been conducted to demonstrate the validity and feasibility of the proposed method.

Using the ultrasonic and the radio frequency, a method for the robot localization and calibration was presented. The distance between the receiver and a beacon can be computed by using the difference between times of flight. The presented method uses the gradient of the maximum amplitude of the ultrasonic in order to measure the time of flight precisely. The measured three distances between the receiver and the beacon were used to compute the robot position by the direct inverse method and the iterated least square approximation method. This paper defines the calibration as the problem to find the location of 3 beacons and 3 robots, and presents 3 methods for it and found the 2B2R method as the best among them.

A method for kinematic modeling of a mobile robot with rocker-bogie link mechanism was described. By using the well-known concept of the instantaneous coordinates, it derives the kinematic model for the full six degree of freedom motion including the x, y, and z motions and the pitch, roll, and yaw rotations. The kinematic model here implies both of the forward and the inverse kinematic equations. The forward kinematic equation with the wheel Jacobian matrices can be used to obtain the robot position and orientation from the measured wheel velocities and the rocker-bogie joint angles. On the contrary, the inverse kinematic equation implies a resulting robot motions consisting of body velocity and turning rate from the individual wheel velocities. Through the computer simulation, the kinematic model of the mobile robot was verified.

The tracking control of an automatic pipe cutting robot, called APCROM, with a magnetic binder was studied. Using magnetic force APCROM, a wheeled robot, binds itself to the pipe and executes unmanned cutting process. The gravity effect on the movement of APCROM varies as it rotates around the pipe laid in the gravitational field. In addition to the varying gravity effect other types of nonlinear disturbances including backlash in the driving system and the slip between the wheels of APCROM and the pipe also cause degradation in the cutting process. To maintain a constant velocity and consistent cutting performance, the authors adopted a repetitive learning controller (MRLC), which learns the required effort to cancel the tracking errors. An angular-position estimation method based on the MEMS-type accelerometer was also used in conjunction with MRLC to compensate the tracking error caused by slip at the wheels. Experimental results verify the effectiveness of the proposed control scheme.

This paper presented a framework of the integrated planning and control for mobile robot navigation. Unlike the existing hybrid architecture, it learns topological map from the world map by using MST (Minimum Spanning Tree)-based SOFM (Self-Organizing Feature Map) algorithm. High-level planning module plans simple tasks to low-level control module and low-level control module feedbacks the environment information to high-level planning module. This method allows for a tight integration between high-level and low-level modules, which provided real-time performance and strong adaptability and reactivity to outer environment and its unforeseen changes. This proposed framework was verified by simulation.

A new method was described for computing trajectories from zero velocity to zero velocity for a second order system. In order to reduce jerk without great efforts, this contribution attempts to quantify the relative performances of different trajectory computation methods for a given maximum acceleration and deceleration over a given path. To compare the jerk between the proposed and conventional trajectory computation method, an industrial robot controlled by transputer network was used.

Humanoid robots usually consist of a number of joint actuators. Thus, it is essential to adequately control the torques of the joint actuators for achieving desirable motion. Also, torque profile needs to be properly generated especially when a robot is to be moved in cooperation with external environment. For the purpose, this paper proposed a method of optimizing humanoid-robot cooperative motion using a genetic algorithm to find an optimized posture in consideration of torque distribution. An eighteen degree-of-freedom humanoid robot was adopted in this study, and a motion pushing an object was considered. When the robot pushing an object moves from a position to one of the next candidate positions, the torques for all joints were calculated. Then, the torques obtained was reflected to an objective function, which was to be minimized by a genetic algorithm. Here, the objective function was designed in such a way that the surplus torques for all joints should be maximized. As a result of successive minimization processes, a posture optimized for the cooperative motion was found. To show the effectiveness of the proposed method, a series of simulations were carried out for a humanoid robot, and the results were analyzed. The simulation results show that the proposed method can be extended to the control of torques for humanoid robots cooperative with external environment.

This paper proposed PDA based control of multiple heterogeneous robots by sharing information to perform surveillance and reconnaissance. Since heterogeneous groups were equipped with different sensor systems, they in particular allow a potentially greater degree of fault-tolerance compared to homogeneous groups. In this paper, unknown environment around robots was recognized efficiently by multiple robots using internet based PDA, where communication between PDA(user interface) and host PC(a controller for robots) is based on TCP/IP protocol. Each of the robots collects sensor data regarding its own motion and environment to share this information with the rest of he team during he update cycles.

Based on a particle filter, a method that mobile robots can track a person in complex environment was presented. The problem of person following for mobile robot has been researched many different areas. The main issues of following a person are real time constraint, motion change of person during the tracking and occlusion with other objects. Using appearance-adaptive models in a particle filter, a robust visual tracking algorithm was realized. The appearance-adaptive model can handle occlusion with other people while the target is moving.

In this paper, a real-time reactive mechanism in the hybrid deliberate/reactive control architecture was proposed. The real-time reactive layer consists of resources, behaviors, an action coordinator, and a controller. Each resource offers an independent sensor information. To improve real-time characteristics, individual behaviors and the action coordinator were designed to fulfill the realtimeness in executing each component in RTAI (Real-Time Application Interface for Linux). The effectiveness and the real-time characteristics of the proposed reactive mechanism were verified by computer simulations of preplanned scenarios.

The research described a method for localizing a mobile robot in the working environment using a vision system and VRML. The robot identifies landmarks in the environment and carries out the self-positioning. The image-processing and neural network pattern matching techniques were employed to recognize landmarks placed in a robot working environment. The robot self-positioning using vision system was based on the well-known localization algorithm. After self-positioning, the 2D scene of the vision is overlaid with the VRML scene. How to realize the self-positioning was described. Also the result of overlapping between the 2D scene and VRML scene was shown. In addition, the advantage expected from overlapping both scenes described.

On developing the personal robot system presently, the internal architecture is every module those occupy separated functions are connected through heterogeneous network system. This module-based architecture supports specialization and division of labor at not only designing but also implementation, as an effect of this architecture, it can reduce developing times and costs for modules. Furthermore, because every module is connected among other modules through network systems, we can get easy integrations and synergy effect to apply advanced mutual functions by co-working some modules. In this architecture, one of the most important technologies is the network middleware that takes charge communications among each modules connected through heterogeneous networks systems. The network middleware acts as the human nerve system inside of personal robot system; it relays, transmits, and translates information appropriately between modules that are similar to human organizations. The network middleware supports various hardware platform, heterogeneous network systems (Ethernet, Wireless LAN, USB, IEEE 1394, CAN, CDMA-SMS, RS-232C). This paper discussed some mechanisms about our network middleware to intercommunication and routing among modules, methods for real-time data communication and fault-tolerant network service. There have designed and implemented a layered network middleware scheme, distributed routing management, network monitoring/notification technology on heterogeneous networks for these goals. The main theme is how to make routing information in our network middleware. Additionally, with this routing information table, we appended some features. Now we are designing, making a new version network middleware (we call 'OO M/W') that can support object-oriented operation, also are updating program sources itself for object-oriented architecture. It is lighter, faster, and can support more operation systems and heterogeneous network systems, but other general purposed middlewares like CORBA, UPnP, etc. can support only one network protocol or operating system.

This paper discussed the whole body cooperated work to generate large impact energy which is generated by arm when hit a wall. The maximum power of arm of the humanoid robot is lower than that of leg. But humanoid robot can generate more impact power by whole body cooperated motion than only used a arm. The humanoid robot falls easily down by force of the repulsion of the impact energy. But the robot can generate much of the impact energy by the stabilization of the ZMP. This paper discussed not only a relationship between the stabilization of the ZMP and the impact energy, but also a relationship between the whole bodies cooperated force control and impact energy.

Dextrous manipulation is a fundamental problem in the study of multifingered robotic hands. Given a hand and an object to be manipulated by the hand in an environment filled with obstacles, the main objectives of dextrous manipulation are to have the hand grasp the object and transfer it from a starting configuration to a goal configuration without collision. To fulfill such a task in general, we will need: (a) a manipulation planner to generate a "feasible" path for the hand; and (b) a controller to implement the planned path. In this overview paper, we define the manipulation planning problem and present a unified Control System Architecture for Multifingered Manipulation (CoSAM²). By incorporating the various kinematic and static relationships of a multifingered robotic hand system with proper sensory data inputs at different stages, CoSAM² achieves the various objectives of dextrous manipulation. Theoretical background of the system design along with real-time experimental results are described.

Contact surface of cross-wedge rolling is a complicated space surface and distribution rule of contact surface stress is very complicated. So far, its analyzed result was still based on slippery line method. Designing mould and actual production mainly depend on experiential factor. Application and development of cross-wedge rolling was baffled seriously. Based on the forming characteristics of cross-wedge rolling with flat wedge-shape, the ANSYS/DYNA software was developed secondly on the basis of itself, and the corresponding command program was compiled. Rolling process of cross-wedge rolling with flat wedge-shape was simulated successfully. Through simulation, space surface shape of contact surface was achieved, and distribution rule of contact surface stress was analyzed detailed and obtained. The results provide important theoretical foundation for avoiding appearing bug on surface of rolled part, instructing to design cross-wedge mould and confirming force and energy parameter.

The development of a fieldbus based electro-hydraulic converter with functions of non-sensor fault self-diagnostic, fault prediction, fault-tolerant and fault self-removing is described. Non-sensor fault self-diagnostic and fault prediction were implemented through the detection of an oscillation current. The current was selected as the criterion of jam fault. Redundancy construction was adopted to realize the fault-tolerant function. Also, the switching principle of the redundancy construction was discussed. A concept of "electrical hammer" was presented to resolve the jam fault on-line automatically. A CEBus based powerline communication network was developed to fulfill the communication demands of field devices which are normally interconnected by "low voltage" wiring. The effectiveness of the proposed converter was demonstrated through simulations and experiments. It was concluded that the converter has a promising future in engineering applications, especially in turbine governing systems that suffer from jam fault.

A kind of actuator, which is used in the high frequency Vibratory Stress Relief (VSR), was researched. The actuator is based on the technology of giant magnetostrictive materials. The design principle of the actuator was firstly analyzed, which consists of the analysis of giant magnetostrictive materials and a force generator. Then the design criterion of magnetostrictive actuators was deeply discussed, which includes the dimension design of magnetostrictive materials, the design of magnetic field and the design of elimination of heat. Finally, a real actuator was developed, which has been used in the high frequency VSR. The experimental results show that the developed actuator works very well. Large exciting force but small vibration amplitude will make it widely used in the VSR.

Gravity center position is one important index to evaluate the stability of vehicles. A high precision gravity center measure system is proposed in this paper, which can be used in weight and gravity center measuring for large heavy vehicles. Based on a static gravity measuring principle with three measuring points, a hydraulic driven 2-DOF motion platform is developed. By means of a degree-of-freedom control, the platform can realize accurate positioning to construct two intersected planes and work out the three-dimensional coordinates of the vehicle gravity center. Experiments demonstrate that the system's measurement error in weight is less than 0.3%, in Gravity center position measurement is less than 3mm in X and Y direction, and in Z direction is no greater than 5mm.

Flexure hinges manufacturing error and hysteresis nonlinearity of piezoelectric actuator are two key factors that affect control precision of micromanipulator. In order to evaluate error affects on micromanipulator accuracy, a partial differentiation error model of manufacturing error was presented. And the flexure hinges hole radius error, cross-sectional width error and minimum thickness error were respectively modeled; besides, we used finite element simulation to validate the analytical error model. The analytical and simulation result shows that the analytical error model is correct, and minimum thickness error induces biggest displacement precision error. In addition, the nonlinear error of piezoelectric actuator was considered, and support vector machine for regression (SVR) was used to model hysteresis nonlinearity. The experiment results show that SVR nonlinear model can accurately describe the nonlinear error. The obtained error model and data are used to establish a theoretical limit for the design and precision control of micromanipulator.