A steering occasion strategy for local collision avoidance of an Unmanned Surface Vehicle in a complex environment

Abstract

Collision avoidance (CA) for an Unmanned Surface Vehicle (USV) is a key technology for autonomy especially in a complex environment. To decide a suitable steering occasion to maneuver for the CA of an USV may be very helpful in some applications such as the long-term monitoring task in the environment with various obstacles. In this paper, a steering occasion strategy is designed. Firstly, a feasible steering angle set model for the CA of an USV is acquired which is mainly classified into six categories. Then a steering occasion strategy is designed by acquiring the least steering angle rate with respect to the relative distance between the USV and the obstacle based on the boundary of the feasible steering angle set. The simulation results of the steering occasion show the usability for CA, and it will be more consistent with the subjective consciousness of people in the CA of an USV with considering the steering occasion. This work will be an important part of our future work about the local path planning in some specific applications.