Enhancing of nominal characteristic trajectory following control for motion systems

The goal of this paper is to enhance a practical nominal characteristic trajectory following(NCTF) controller that is specifically designed for two-mass point-to-point positioning systems. A nominal characteristics trajectory contained in the NCTF controller acts as movement/motion reference and a compensator is utilized to force the object to detect and follow the reference/desired trajectory. The object must follow and track closely and should be as fast as possible. The NCTF controller is designed with two different intelligent based compensator approaches which are fuzzy logic and extended minimal resource allocation network. The proposed controller which is NCTF are compared with the conventional proportional integral compensator. Then the results of simulation are discussed for the positioning performances. The inertia variations due to the effect of the design parameters are also assessed to see the robustness of controllers. The results show that the NCTF control method designed from an intelligent based compensator has a better positioning performance in terms of percentage of overshoot, settling time, and steady state error than the classical based compensator.

Adaptive Learning with Large Variability of Teaching Signals for Neural Networks and Its Application to Motion Control of an Industrial Robot

Recently, various control methods represented by proportional-integral-derivative （PID） control are used for robotic control. To cope with the requirements for high response and precision, advanced feedforward controllers such as gravity compensator, Coriolis/centrifugal force compensator and friction compensators have been built in the controller. Generally, it causes heavy computational load when calculating the compensating value within a short sampling period. In this paper, integrated recurrent neural networks are applied as a feedforward controller for PUMA560 manipulator. The feedforward controller works instead of gravity and Coriolis/centrifugal force compensators. In the learning process of the neural network by using back propagation algorithm, the learning coefficient and gain of sigmoid function are tuned intuitively and empirically according to teaching signals. The tuning is complicated because it is being conducted by trial and error. Especially, when the scale of teaching signal is large, the problem becomes crucial. To cope with the problem which concerns the learning performance, a simple and adaptive learning technique for large scale teaching signals is proposed. The learning techniques and control effectiveness are evaluated through simulations using the dynamic model of PUMA560 manipulator.

Fuzzy Control Method with Application for Functional Neuromuscular Stimulation System

A fuzzy control technique is applied to a functional neuromuscular stimulation (FNS) physical multiarticular muscle control system. The FNS multiarticular muscle control system based on the fuzzy controller was developed with the fuzzy control rule base. Simulation experiments were then conducted for the joint angle trajectories of both the elbow flexion and the wrist flexion using the proposed fuzzy control algorithm and a conventional PID control algorithm with the FNS physical multiarticular muscle control system. The simulation results demonstrated that the proposed fuzzy control method is more suitable for the physiological characteristics than conventional PID control. In particular, both the trajectory following and the stability of the FNS multiarticular muscle control system were greatly improved. Furthermore, the stimulating pulse trains generated by the fuzzy controller were stable and smooth.