Robust Adaptive Control Of Robots Based On Finite-time Stability Theor

Robots can completely revolutionize the production mode,war mode,and lifestyle of human society.As an important branch of robots,manipulators are widely used in agriculture,vehicles,aerospace systems,and other fields.However,the actual operating environment of industrial robotic manipulators is a nonlinear constraint of multifactor coupling.Existing methods based on backstepping control have limitations such as insufficient control accuracy and “explosion of complexity”.Therefore,under the influence of multiple factors,how to ensure a more rapid and accurate response-ability,and robustness,when parameters change,has become the main challenge of robotic manipulator control,which has important research value in realizing robust control of the robotic manipulator.By analyzing the current research status in the field of controlling robotic manipulators at home and abroad,this paper summarizes the limitations of commonly used methods and several challenges in the field of robust adaptive control of robotic arms.Based on the above research status and the main challenge,this paper comprehensively considers nonlinear factors such as state constraints,parameter perturbations,actuator saturation failures,and time delays in the operation of the robotic manipulator.Based on the framework established by backstepping control and finite-time stability theory,this paper discusses the finite-time prescribed performance control for the full-state-constrained manipulators,finite-time faulttolerant control for the robotic manipulators based on prescribed performance,and finitetime fault-tolerant control for the robotic manipulators with asymmetric output constraints.The main research contents are as follows:(1)To establish the space model of the manipulator,the pose change and coordinate transformation of the manipulator in space are introduced.Based on the Lagrange dynamic analysis method,the dynamic model of the multi-joint manipulator in space is established,and its dynamic characteristics are analyzed.Basic theoretical knowledge is introduced such as the backstepping control,radial basis function neural network,related assumptions,and lemmas to lay a theoretical foundation for subsequent controller design.(2)For the tracking control issue of the manipulator with full-state constraints and multiple unknown nonlinearities,a finite-time modified prescribed performance control scheme is devised based on the backstepping control method.With the aid of the secondorder finite-time command filter and finite-time stability principle,the control accuracy of the manipulator and the convergence speed of errors is improved.Eventually,the advantages and disadvantages of the scheme devised in this chapter are explained from theoretical and experimental perspectives.(3)In the operation of the robotic manipulator,in addition to being constrained by the states,it is also influenced by other factors.Therefore,a novel prescribed performance finitetime fault-tolerant control strategy is proposed to realize high-precision control for the robotic manipulator with full state constraints,actuator input saturation faults,time delays,and various nonlinear terms.The proposed controller ensures that tracking errors converge to the preset range at a faster speed within a limited time.Then,the effectiveness of the designed scheme is demonstrated through a comparison of schemes.(4)The actual state constraints on a robotic manipulator are typically asymmetric.Hence,a finite-time fault-tolerant control strategy is designed for the tracking control problem of the robotic manipulator with asymmetric output constraints and actuator input saturation faults.By combining the arctangent function,hyperbolic tangent function,and piecewise function,fast convergent time-varying boundaries are designed to improve the control accuracy and robustness of the robotic manipulator.Then,from an experimental perspective,demonstrate the superiority of the designed theory.