Adaptive Path Following Control System for Unmanned Surface Vehicles

Abstract

The accurate and reliable path following control under environmental disturbances is key and difficult for the autonomous navigation of unmanned surface vehicles(USV). An adaptive path following control system with three levels, i. e., algorithm design, system realization and experimental verification is studied. Specifically, on the algorithm level, an adaptive line-of-sight(LOS) navigation algorithm and an LEM(Line-of-sight & extended state observer & model predictive control) based path following control method is proposed. On the system realization level, an architecture of the proposed adaptive path following control system is designed, and the fast solving problem of MPC and the evaluation problem of unmeasured motion states are solved. On the experimental verification level, a path following control simulation platform based on a model ship is built in the outdoor pool. On the basis of this platform, the comparison experiments on the path following performance between MPC and PID(Proportional-integral-derivative) method, and between LEM and TLM(Traditional LOS & MPC) method are conducted, respectively. The experiment results show that the built platform functions well, and the proposed LEM based path following methods have higher accuracy and reliability than PID based methods.