基于自适应神经网络的工业机器人双臂协同鲁棒控制OA北大核心CSTPCD

To overcome the influence of uncertainties such as mechanical friction,external interference,and model errors on the accuracy of industry robot dual-arm motion trajectory control,a collaborative robust control method for industry robot dual-arm based on adaptive neural network was designed.Firstly,a dynamic model of industry robot dual-arm with various uncertainties was established.Then,a collaborative control law with uncertainty was designed by constructing an obstacle Lyapunov function,and an adaptive neural network was designed to estimate the uncertainty of the system,obtaining a robust collaborative control law for in-dustry robot dual-arm.Finally,the Lyapunov stability theory was used to demonstrate that the designed collaborative robust control law can constrain the trajectory tracking error,velocity tracking error,and uncertainty estimation error of the industry robot dual-arm within an arbitrarily small neighborhood.The simulation results show that the designed adaptive neural network can accurately estimate the uncertainty in the industry robot dual-arm system,and the maximum estimation error is only 0.04 N·m.The proposed collaborative robust control law can stably and accurately track trajectory control instructions,and the maximum trajectory tracking error is only 1.3 mm,verifying the rationality of the designed method.In the fixed coordinate positioning test in 3D space,the proposed collaborative robust control law has higher control accuracy compared with other methods,the average and maximum positioning error are only 1.1 mm and 1.4 mm,respectively,demonstrating stronger robustness and better engineering applicability.