AUV 3D Trajectory Tracking based on Linear Active Disturbance Rejection Control

Autonomous underwater vehicle (AUV) has the characteristics of nonlinear, time-varying, strong coupling and easy to be disturbed. In order to further improve the AUV trajectory tracking control effect, a five-degree-of-freedom spatial motion model was established and the linear active disturbance rejection control method was introduced into AUV trajectory tracking control. Considering the characteristics of the object studied in this paper, firstly, coordinate transformation of the standard motion equation of the underactuated AUV can obtain a simplified motion model of the object studied in this paper. Variable transformation including all degrees of freedom is introduced in this paper, so that the underwater vehicle has dynamic characteristics in all directions and includes all control inputs. Secondly, for the position control system, a linear extended state observer (LESO) is designed for the three motion channels after coordinate transformation to observe and compensate the internal uncertainties and external disturbances in real time. Finally, the trajectory tracking control based on linear active disturbance rejection control is realized by simulation, and the effectiveness of the proposed method is verified.

I. Introduction

Autonomous underwater vehicle (AUV) has become an important tool of Marine research because of its important role in Marine development research and underwater warfare, and has been paid attention to by Marine and military powers in the world. AUV path tracking control in complex Marine environments is a prerequisite and foundation for successful execution of high-precision underwater operations. However, due to the inherent uncertainty of the model parameters and the existence of complex disturbances in the Marine environment, the trajectory tracking control of AUVs is faced with great challenges. At present, the research on AUV path tracking control has achieved a lot of research results, and the commonly used control algorithms include backward step control, neural network control and active disturbance rejection control. In [1], collaborative path tracking control is realized based on Lyapunov theory and backstep method. In [2], a controller design method based on neural network adaptive control is proposed to solve the problem of cooperative path tracking control of multiple autonomous surface ships. In [3] - [4], an extended state observer was used to estimate the disturbance, and an anti-disturbance path tracking controller was designed on this basis. The environmental disturbance of the system was estimated and compensated by ESO, so that the control error tended to zero after a period of transient time. Based on Mr. Han Jing qing's active disturbance rejection control theory [5], Mr. Gao Zhiqiang [6] proposed linear active disturbance rejection control, and adjusted the controller and observer parameters by using the concept of bandwidth, which greatly reduced the number of parameters and improved the engineering practicability. In [7], the coordinate of the virtual reference point in front of the AUV system is used as the output equation to ensure the linearization of the input and output feedback of the AUV kinematics and dynamics. Literature in [8] presents a nonsingular terminal sliding mode and active disturbance rejection decoupling control scheme. This paper introduces three control input vectors based on the input-output feedback linearization model proposed in [7], which is more concise and has better tracking effect than the five-channel decoupling in [8].