多关节机械臂轨迹跟踪自适应神经网络滑模控制

Self-Adaptive Neural Network Sliding Mode Control for Trajectory Tracking of Multi-Joint Manipulator

针对不确定干扰和建模误差对多关节机械臂轨迹跟踪控制系统造成的不良影响,对基于滑模控制的自适应神经网络滑模控制算法进行了研究。通过神经网络,对多关节机械臂轨迹跟踪控制系统的不确定干扰和建模误差进行逼近。添加自适应项,补偿神经网络滑模控制中神经网络模型对系统中的不确定干扰和建模误差的逼近误差。设计了具有指数趋近律的滑模面,以提升多关节机械臂轨迹跟踪控制系统的鲁棒性和响应速度。使用李雅普诺夫稳定性理论,证明了多关节机械臂轨迹跟踪控制系统的半全局稳定性,并通过MATLAB对理论结果进行了仿真验证。仿真结果表明,对具有不确定干扰和建模误差的多关节机械臂轨迹跟踪控制系统,采用该算法进行轨迹跟踪时,具有较好的稳定性与鲁棒性。该控制算法能合理应用到此类轨迹跟踪控制系统中。

To deal with the influence of uncertain interferences and modeling errors on the trajectory tracking control of the multi-joint manipulator,the self-adaptive neural network sliding mode control algorithm based on sliding mode control is researched.The uncertain interferences and modeling errors are approximated through neural network,and the self-adaptive term is added to compensate the approximation error of the neural network model and modeling errors.The sliding surface with exponential approach law is designed to improve the robustness and response speed of multi-joint manipulator trajectory tracking control system.The Semi-global stability of the system is proved by using Lyapunov stability theory,and the theoretical results are verified by MATLAB.The simulation results verify that the control algorithm can achieve stable tracking of the desired trajectory in the multi-joint manipulator trajectory tracking control system with uncertain interference and modeling error,which has better stability and robustness.The control algorithm can be applied in similar tracking control systems.