Trajectory Tracking Control Of Cushion Robot Based On Neural Network Algorithm

Since the cushion robot serves the elderly with mobility difficulties and patients in paralysis of the lower limbs,the robot must accurately track the pre-planned path while ensuring the safety of users.Due to the factors of the sitting position and mass inertia of the robot user,the center of gravity of the robot will shift during use.In addition,in the process of robot movement,it is inevitable that it will be disturbed by various external uncertainties.These factors will cause the cushion robot to deviate from the predetermined trajectory in the process of movement.Considering the safety of users,it is necessary to solve the problems of mobile robot center of gravity deviation and external interference,and to complete the accurate tracking of the predetermined trajectory.Based on the approximation ability of neural networks to nonlinear functions,an accurate trajectory tracking controller is designed in this paper.The main contents are as follows:Firstly,the basic functions of the cushion robot studied in this thesis are introduced,the problems about the motion control of the mobile service robot are expounded,and the kinematics model and dynamics model of the cushion robot are constructed.Secondly,aiming at the problem that the center of gravity of the cushion robot deviates in use,a neural network PD adaptive controller is designed.Based on Lyapunov stability theory,the stability of the system is analyzed.MATLAB experimental simulation proves the effectiveness of the method and eliminates the adverse impact of the center of gravity deviation on the trajectory tracking process of the cushion robot.Finally,a robust neural network controller is designed to solve the gravity center offset problem and the external dry uncertain disturbance of the cushion robot system.the neural network is used to approximate the uncertainty of the robot system,and the approximation error is treated as the uncertain disturbance and eliminated by the robust controller.The global asymptotic stability of the robot system is guaranteed.