Research on Trajectory Tracking Control for SCARA Manipulator of Tea Picking Robot
【摘要】：In this paper, SCARA manipulator is used as the picking mechanism of an intelligent tea picking robot in order to realize the selective and automatic picking of tea leaves with high quality. During the tea picking process, SCARA manipulator is constantly affected by the surrounding environment and its own nonlinear and joint coupling, which will make the manipulator unable to reach the designated tea position accurately and quickly. To solve the problem, this paper presents an intelligent controller based on neural network(NN) adaptive compensation. The controller consists of a feed-forward NN with parameters adjusted adaptively and a linear PD feedback controller. The NN controller uses radial basis function(RBF) NN to approximate the uncertain terms in the dynamic model of SCARA manipulator through online learning. Lyapunov function is defined and the neural network adaptive law of the manipulator is obtained through Lyapunov analysis. Finally, a large number of numerical simulation experiments are performed using MATLAB/Simulink. By comparing with traditional computed torque controller, it is proved that the proposed controller has better performances, which can eliminate the effects of uncertainties and external disturbance effectively. Experimental results show that the desired trajectory can be tracked well and the root-mean-square error(RMSE) is small.