Aiming at the path following control problem of underactuated unmanned ship, an intelligent control method of path following of unmanned ship based on adaptive line-of-sight(LOS) algorithm is proposed. In terms of guidance algorithm, considering the ship motion characteristics, an adaptive LOS(ALOS) algorithm that automatically adjusts the target point switching mechanism according to the angle between adjacent paths is proposed to improve the LOS guidance performance. On this basis, a path following controller of unmanned ship based on Deep Q-Network(DQN) algorithm is designed. The controller can learn the path following control strategy through interaction with the environment without any prior knowledge and experience. The simulation results show that ALOS method can obtain better path following performance, and the proposed control method has good self-learning ability and control performance.
Path following optimization of unmanned ships based on adaptive line-of-sight guidance and Deep Q-Network
01.08.2022
308627 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Path following of underactuated USV based on modified integral line-of-sight guidance strategies
British Library Online Contents | 2018
|LINE-OF-SIGHT GUIDANCE DEVICE AND LINE-OF-SIGHT GUIDANCE METHOD
Europäisches Patentamt | 2019
|