Teruo Tsuji

Terminal sliding mode control of second‐order nonlinear uncertain systems

In this paper, a terminal sliding mode control scheme is proposed for second-order nonlinear uncertain systems. By using a function augmented sliding hyperplane, it is guaranteed that the output tracking error converges to zero in finite time which can be set arbitrarily. In addition, the proposed scheme eliminates the reaching phase problem so that the closed-loop system always shows the invariance property to parameter uncertainties. Copyright

On-line identification of continuous time-delay systems combining least-squares techniques with a genetic algorithm

This paper proposes a new approach to on-line identi® cation of continuous timedelay systems from sampled input± output data. In order to track the time-varying time-delay and system parameters, the linear recursive least-squares (RLS) method is combined in a bootstrap manner with the genetic algorithm (GA) which has a high potential for global optimization. The time-delay is coded into binary bit strings and searched by the GA, while the system parameters are updated by the RLS method. Since only the time delay is searched by the GA, a small population size for the GA is suc cient and hence it is possible to implement the algorithm on line on the digital computers. Furthermore, this method (GALS method) is hybridized with the sequential nonlinear least-squares method which is eŒective in local search, to improve the speed of convergence. Simulation results show that both the GALS and the hybrid methods are quite ec cient. It is also veri® ed that, since the hybrid method is eŒective in both global and local optimizations, it has superior tracking performance over the GALS method especially in the case where the system parameters and time delay vary continuously with time.

Parameter Identification and Adaptive Control of Continuous Systems with Zero-Order Hold

Abstract This paper discusses the digital implementation techniques of parameter identification and adaptive control for continuous systems with zero-order hold (ZOH) focusing on the bilinear transformation based on the block pulse functions (BPFs). It is found that the emphasized discretizing method yields very excellent performances of parameter identification and adaptive control even in the case of large sampling intervals.

Sensorless speed control of synchronous reluctance motor using RTLinux

In this paper a novel sensorless speed control of a synchronous reluctance motor (Syn.RM) is presented. The sensorless control is based on the modified flux linkage observer which is proposed by the authors for a permanent magnet synchronous motor (PMSM), and the theory is applied to Syn.RM here. The observer is able to estimate the modified flux linkage and the electromotive force (EMF) simultaneously. The motor speed and the rotor position are calculated from these estimated values. The validity of the proposed method is verified by experiment using the RTLinux.

A control design of robotics using the genetic algorithm

This paper presents a new and practical method for a control design of a robotic system. In general, actuators in robotic systems are set with gears whose characteristics are elastic. Since a state feedback-type digital controller is usually used for such a robotic system, the design of the feedback gain of the controller is important, because undesirable vibrations or an overshoot in responses occur for high gains. Therefore the desired response, the output of a reference model, is designed first, and the feedback gains are determined so that the response will coincide with the desired response, which is an optimization problem. The gradient method works to some extent, but it takes a long time to get a satisfactory result. Thus we applied the genetic algorithm (GA) to this nonlinear optimization problem, which gave the very first convergence. The gains obtained have many useful applications. The results of a simulation are also given.

Frictionless control of robot arm with sliding mode observer

In this paper, the authors consider a one-link robot arm, and the purpose of this study is to realize a frictionless joint of the arm of a robot. In order to do so, compensation of not only friction but also spring effect in harmonic gear which is set on the motor shaft is required. For their compensation, the authors consider a nonlinear sliding mode disturbance observer, and removing the effect of the disturbance by the observer, the parameters of the system are identified by a GA (genetic algorithm). The disturbance observer is also implemented in a controller in an online manner, so that frictionless jointing of the arm of the robot is realized. In order to verify the frictionless joint by compensation of the disturbance, the authors try to set the spring characteristics for the joint by a control design. The effectiveness of their proposal is verified by simulation and experimental results.

Sensorless speed control of synchronous reluctance motor using a novel flux estimator based on Recursive Fourier Transformation

This paper proposes a sensorless control of SynRM (synchronous reluctance motor). It can calculate the motor flux from an armature voltage and a current directly. The proposed estimator is able to eliminate both the dc offset and harmonic component of the armature current which cause the estimation error. State observer based Fourier Transformation called RFT (Recursive Fourier Transformation) is effective to estimate the fundamental component of the flux. Also a block pulse approximation is used to convert a continuous time state equation to a difference equation to reduce the calculation time. The validity of the proposed method is shown in the experimental results.

Vector approximation method with parameter adaptation and torque control of CSI-fed induction motor

The concept of a vector approximation method for a CSI-fed induction motor consists of the vector control and the shaping of a stator current wave. The vector approximation means that a desirable current space vector is approximated by two realizable current vectors which have only six directions, and by DC link current control. The main purpose of the vector approximation method is not only to decrease torque ripple but also to control instantaneous torque, which is different from that of a general PWM method. Moreover, in this paper, a parameter adaptation system containing a magnetic saturation table and a model reference adaptive system (MRAS) is newly added to the system, which may bring more precise vector approximation and torque control to induction motor drive systems. The simulation and experimental results are shown in this paper.<<ETX>>

CS1 drive induction motor by vector approximation

A new method decreasing torque ripple for a current source inverter (CSI) drive induction motor system is shown. Unlike a PWM system, this method uses a detailed online simulation to compute an optimal stator current which cannot be realized but is approximated by the realizable stator current vectors. The approximation is done using the concept of time average for the composition of space vectors. Although the second field current cannot be measured, the observer theory assures rapid convergence of the simulated value to the real one. Since the detailed online simulation takes considerable computational time, the experiments are done for relatively low rotational speeds.<<ETX>>

Impulse Response Identification of Continuous Systems using Generalized Radial Basis Function Networks

Abstract In this paper, the authors propose a new identification method for the continuous-time impulse response model (CTIRM) of a linear system from sampled data in the ill-posed cases of band-limited input, fast sampling rate and measurement noise of high level which may be colored. To achieve smooth estimate, the CTIR which is viewed as a continuous function of time, is approximated by a generalized radial basis function network (GRBFN) in which the gaussian basis functions are equidistantly located in the time-domain and have the same width. Then the parameters of the GRBFN are estimated by the least-squares (LS) method. The design techniques of the center distance and the basis function width axe also shown using frequency-domain analysis. Simulation results are included to verify the performance of the proposed method.