基于神经网络的机器人二阶滑模控制

Second Order Sliding Mode Trajectory Tracking Control of Robot Based on Neural Network

本文提出了一种基于神经网络与二阶滑模控制融合的控制策略用于非线性机器人控制,设计了一种新颖简易的二阶滑模控制方法,有效地避免了常规变结构控制的抖震问题,并采用神经网络辨识未知的机器人的非线性模型,通过Lyapunov直接法设计网络的权值更新率,确保了系统闭环全局渐近稳定性。最后,通过仿真验证了算法的有效性。

This paper proposes a synergetic controls algorithm by adaptive neural network and Second order sliding mode control. Design a second order sliding mode control method with novelty and facility, and the chattering problem is avoided effectively, Neutral network is used to adaptive learn and compensate the unknown nonlinear model. The learning algorithm for the free neutral network parameters are presented by Lyapunov direct method. The global asymptotic stability is guaranteed. Finally, the control performance of the proposed controller is verified with simulation studies.