Research On Adaptive Control Methods For Telerobotic System With Time-varying Delay

Telerobot can replace operators to perform tasks in dangerous and complex remote environment.It has a wide application prospect in aerospace exploration,telemedicine,rescue and disaster relief,military,hazardous material disposal and other fields.The communication delay and nonlinear,time-varying dynamics of the telerobot system give great challenges in stabilization and high-performance control performance.In the past decades,a lot of research work has been done on the control of time-delay teleoperation systems.Wave variable,proportional damping control,robust control,adaptive control,and other control methods are studied and used in the control of teleoperation system.However,these methods still have some limitations in dealing with time-varying time delay,dynamic uncertainty,external force interference,actuator saturation,and other conditions.Therefore,it is necessary to study the control method of teleoperation system to deal with the above complex factors and improve the control performance.Therefore,in this thesis,some interactive control methods and control structures of telerobot system are discussed and expounded.Based on the adaptive control and finite time control methods,three types of control structure and five controllers are designed and proposed for teleoperation systems.In additional,the position and position-force synchronous control of teleoperation robot are realized respectively.The specific research contents of this thesis are listed as follows:(1)Aiming at the position tracking control of master and slave robots in teleoperation system with above control conditions,the position tracking error integral term is introduced,and a new nonsingular terminal sliding mode structure is constructed.Based on this mode,the corresponding master-slave finite time controller is designed.RBF neural network,adaptive control method and Lyapunov theory are used to solve the uncertainties of dynamics and ensure the stability of the system.(2)Aiming at the position tracking control problem of master-slave robot in teleoperation system under above control conditions and actuator saturation,a new auxiliary variable structure is designed based on the error integral term.The RBF neural network saturation compensation method based on switching gain strategy and adaptive saturation compensation method with second-order auxiliary variables are respectively proposed.Based on these two compensation methods,two different finite time controllers are designed respectively.RBF neural network,adaptive control and Lyapunov method are introduced to solve the problem of dynamic uncertainties and communication delays.The adaptive term of communication delay is additionally designed to realize the compensation control of time-varying delays.These controllers do not need acceleration signal,and can get good position tracking control accuracy and fast error convergence speed under the active external force.(3)Aiming at the position and force tracking control of master and slave teleoperation robots with above control conditions,adaptive control structure with position-force fusion error at master side and finite time adaptive control structure with the position force hybrid error are designed.In the first control scheme,an adaptive controller is designed to realize the position and force tracking control by constructing the auxiliary variables with position error,error integral term and force error filter at the master side.In the second control scheme,a hybrid error system is proposed based on the output of position error and force error filter,and the auxiliary variables of finite time structure are constructed by using hybrid error system and velocity feedback filter.The finite time performance of position and force tracking is firstly achieved.As the work in(2),RBF neural network,adaptive control,delay adaptive compensation term and Lyapunov method are introduced to solve the problems of dynamic uncertainties and time-varying communication delay.The proposed two methods can realize the position and force tracking control for the teleoperation system.The finite time control structure does not need acceleration signal,has faster control accuracy and convergence speed than other position and force control methods.