Trajectory Tracking of Industrial Robot Based on PDAG Algorithm

Industrial robot trajectory tracking is a process of using sensors to track the path in three-dimensional space. The most common sensor used for trajectory tracking is the camera installed on the robot. According to the repetitive characteristics of the tasks performed by industrial robots, a PDAG algorithm is proposed. The learning matrix is applied to the error data to establish the mathematical model of PDAG algorithm, and the stability and convergence of PDAG algorithm are proved theoretically. Gravity compensation is used as feedforward control to eliminate the influence of joint gravity in dynamic process and improve the convergence speed and trajectory tracking accuracy of iterative learning control. In this paper, we will discuss the PDAG algorithm, which can be used to track the position and direction of an industrial robot or mobile robot through a camera installed on it. PDAG represents particles in cells and was developed by Dr. David Lowe of the University of California, Berkeley, in 1994. Since then, it has been widely used in various applications, such as robot simulation, control system design, etc. Combining variable structure control with PID+feedforward algorithm, a PID+feedforward control algorithm based on variable structure compensation is proposed to improve the accuracy of trajectory tracking. The fast variable structure control algorithm optimizes the approaching motion to improve the control quality; Based on fast variable structure control, adaptive fuzzy variable structure control uses fuzzy algorithm to establish the relationship between error and reaching law to improve control quality. These two algorithms can improve track tracking accuracy and response time. Finally, through simulation and comparison, the advantages and feasibility of the proposed algorithm in tracking accuracy and response time are theoretically verified.