Tracking control study of AUV large curvature path based on artificial physics method

To enhance the performance of underwater robot path tracking, an AUV path tracking method based on artificial physics (PTM-AP) is proposed. To ensure smooth AUV movement when tracking large curvature paths and reduce energy consumption, a series of continuous and smooth heading control instructions are parsed for the AUV. The main concept is to create a virtual physical force between the AUV and the intended path, and determine the heading control instructions based on Newton’s law. To improve tracking accuracy and convergence speed when following paths with large curvature, multiple virtual physical forces are added. The kinematics controller is then designed in conjunction with the traditional line of sight method. The algorithm designs an adaptive adjustment function and an anti-saturation link to distribute multiple virtual forces. Through simulation experiments, the most appropriate dynamics controller parameters are determined. The tracking effects of line-of-sight (LOS), artificial physics (AP), and PTM-AP are compared. The actual maritime experiment was conducted in Tuandao Bay, Qingdao, China. The results showed that compared with other algorithms, PTM-AP reduced the path convergence time by at least 26.67% and saved energy consumption by at least 25.49%. The results show that the proposed AUV path tracking algorithm based on artificial physics method is feasible when tracking larger curvature paths.