The Finite Time Control For Teleoperation System

With the rapid development of science and technology,teleoperation robot system has been widely used in a variety of environments.In recent years,master-slave teleoperation robot system can be found in many fields from micro to macro,such as space exploration,archaeological exploration,deep mining,nuclear plant detection and remote surgery.For the teleoperation control system,the most important performance requirements are stability and transparency.However,due to the complexity of the internal structure of the manipulator joint and the uncertainty of the working environment,the manipulator is a typical nonlinear,strong coupling,time-varying uncertain system.At the same time,the actuator saturation will be caused when the manipulator is operated quickly.All these problems mentioned above seriously affect the stability and transparency of teleoperation system.Therefore,in order to achieve positive performance,it is particularly important to study the control problems of the teleoperation system under the condition of uncertain model and actuator saturation.In this dissertation,a finite time control method is proposed to solve the problems of communication delay,actuator saturation and model uncertainty in the manipulator teleoperation control system.The main research contents of this paper are as follows1)An adaptive finite time tracking control algorithm is proposed to solve the problem of actuator saturation and unknown disturbance of a single manipulator with unknown parameters,so that the manipulator can track a given trajectory with any small precision.An unknown saturation factor is introduced to describe the actuator saturation,and an adaptive law is designed to identify the saturation factor to compensate the actuator saturation,RBF neural network approximates the inverse dynamic solution and unknown disturbance of the manipulator,and proposes a nonsingular fast terminal sliding mode controller,which enables the manipulator to track a given trajectory in finite time.The effectiveness of the designed control law is verified by simulating the Geomagic Touch manipulator,and a physical experiment platform is established to further verify the designed algorithm on the physical manipulator.2)Considering the information transmission delay of the teleoperation system,the saturation of the actuator and the unknown situation of the slave end environment,the non-singular fast terminal sliding mode controller is used to realize the finite time bilateral control of the teleoperation system.In order to solve the problem of actuator saturation,an unknown saturation factor is introduced to describe the saturation of the actuator,in the meanwhile an adaptive law identification saturation factor is designed to compensate for the saturation of the actuator;in order to solve the information transmission delay,a reference trajectory generation is designed at the slave end using a filter It solves the delay problem from the master to the slave;the environmental force of the slave is modeled by a general RBF neural network,and its parameters are estimated and then transmitted to the environmental torque reconstruction of the master to provide force feedback to the master operator.To ensure the transparency of the system;based on RBF neural network at both ends of the master and slave,the NFTSM controller is designed to realize the position tracking and speed tracking of the master and slave ends,and ensure the bilateral stability of the teleoperation system in the finite time.A simulation example proves the good tracking ability of the designed algorithm.Finally,a teleoperation physical platform is established,and experiments are carried out on the physical platform to realize the teleoperation control of the dual joints of two Gemomagic Touch manipulators.