A Path-Velocity Decomposition Approach to Collision Avoidance for Autonomous Passenger Ferries in Confined Waters

Abstract

A deliberate collision avoidance approach for autonomous surface vehicles operating in confined waters with high traffic is presented. The approach focuses on dynamic obstacles, by assuming a predefined set of paths that are collision-free with respect to static obstacles. Hence, the collision avoidance problem is reduced to a velocity planning problem, which is solved by first transforming all dynamic obstacles to a path-time space and subsequently constructing a conditioned visibility graph and traversing it with Dijkstra’s algorithm. The performance of this approach is demonstrated through both simulations and full-scale experiments in Trondheim harbor, using the NTNU milliAmpere ferry platform and virtual dynamic obstacles.