Research On Adaptive Fuzzy Control Of Nonlinear Robot System

In recent years,with the increasing variety of robots and the increasing complexity of control devices,the robot system is difficult to be represented by nonlinear mathematical model.For some robot systems,even if the nonlinear mathematical model can be obtained,there are inevitable unknown or unpredictable input uncertainties.Therefore,how to solve the uncertainty problem in robot systems has become a cutting-edge topic highly valued by scientists and engineers.In this thesis,the control of rigid robot is studied.1.Aiming at the unmeasurable state and jitter in nonlinear robot system,an adaptive output feedback fuzzy control is designed.The first is observer based feedback adaptive fuzzy control.In this control algorithm,Mamdani fuzzy rules and adaptive law are used to approximate the unknown function of the system.On this basis,a state observer is designed according to the input and output of the system to estimate the state value of the system and feed back to the fuzzy controller,so that the control system can obtain ideal state parameters,Realize high-quality trajectory tracking control;The other is adaptive fuzzy high gain feedback control.Firstly,the unknown nonlinear function is constructed with the help of fuzzy logic system,and the adaptive law is introduced to adjust the fuzzy system parameters online.Finally,the high gain control term is introduced for the fuzzy approximation error to effectively solve the problem of system jitter.2.Aiming at the nonlinear interference factors such as time,friction and load change in the dead zone,an adaptive algorithm is proposed,Firstly,the control algorithm based on fuzzy compensation is designed,the adaptive fuzzy compensator is used to compensate the nonlinear terms such as friction and load change,and the control algorithm is introduced to jointly design the input of the fuzzy system,which overcomes the influence of the above nonlinear factors and improves the accuracy of the system.Finally,according to the characteristics of each interference term,it is distinguished and compensated respectively,which shortens the error convergence time and ensures the high efficiency of the system.3.For the robot system with uncertain dynamic parameters,fuzzy adaptive control and two different robust controls are combined respectively to improve the robustness of the system.The first control algorithm combines fuzzy adaptive algorithm and sliding mode algorithm.The algorithm uses Mamdani fuzzy rules and adaptive law to approximate the nonlinear function of the system.The sliding mode algorithm is used to construct the sliding mode plane so that the error function is bounded and tends to zero.The second control algorithm combines the fuzzy adaptive algorithm with the high-frequency algorithm.The fuzzy adaptive algorithm is used to linearize the fuzzy modeling of the unknown nonlinear function,and the high-frequency control algorithm is used to eliminate the uncertain factors in the system.Simulation results show that the proposed algorithm has stronger robustness.Simulation results using MATLAB data demonstrate that the adaptive fuzzy control algorithm proposed in this thesi can effectively suppress the influence of uncertainties and its advanced nature.