Path tracking of AUV based on improved line-of-sight method

Abstract

With the development of marine resources, more attention has been paid to marine exploration technology. Autonomous underwater vehicle (AUV) is the necessary equipment for marine exploration. The effect of path tracking has an important impact on the safety and efficiency of AUV when performing tasks. However, due to the disturbance of ocean current, the uncertainty of model and the underactuated of AUV, the tracking effect is greatly affected. And the widely used line of sight (LOS) algorithm for path tracking is not effective because of its fixed look-ahead distance. In order to solve this problem, scholars have put forward various methods to adjust the look-ahead distance. The look-ahead distance of proportional line-of-sight (PLOS) algorithm can change with time in the tracking process, and the integral line-of-sight guidance law can eliminate the steady-state tracking error by integrating the tracking error. In addition, there are some methods to adjust the look-ahead distance by reinforcement learning, but these methods are difficult to migrate due to the limitation of equipment. Therefore, this paper proposes an improved line of sight method which considers AUV velocity, tracking error and path curvature. Meanwhile, the results of the control law are fed back to the tracking law to adjust the parameters of the line of sight method, and simulation experiments and sea trials are carried out.