Behavior-based control for marine vehicles using task composition

Abstract

Recently, there has been a growing expectation for autonomous systems to cope with complex situations as the autonomy of unmanned systems improves. Although research on unmanned surface vessels (USVs) is being conducted for maritime missions, the dynamic nature of the ocean environment and the complexity of the mission make it difficult to respond immediately with traditional optimization methods. This study proposes behavior-based control that operates robustly in complex mission environments by composing systems with combinations of simple behaviors. USV’s behavior is designed taking into account collision regulations and defined based on the defined objective function for each behavior. Through task composition, each behavior is defined as a logical unit element, and a scalable behavior-based control framework is designed by applying logical operations. The proposed algorithm’s performance and effectiveness were demonstrated through simulation in an environment composed of waypoint tracking, obstacle avoidance, and rescue behavior.