Docking Control of an Autonomous Underwater Vehicle Using Reinforcement Learning

Abstract

To achieve persistent systems in the future, autonomous underwater vehicles (AUVs) willneed to autonomously dock onto a charging station. Here, reinforcement learning strategies wereapplied for the first time to control the docking of an AUV onto a fixed platform in a simulationenvironment. Two reinforcement learning schemes were investigated: one with continuous stateand action spaces, deep deterministic policy gradient (DDPG), and one with continuous state butdiscrete action spaces, deep Q network (DQN). For DQN, the discrete actions were selected as stepchanges in the control input signals. The performance of the reinforcement learning strategies wascompared with classical and optimal control techniques. The control actions selected by DDPG sufferfrom chattering effects due to a hyperbolic tangent layer in the actor. Conversely, DQN presents thebest compromise between short docking time and low control effort, whilst meeting the dockingrequirements. Whereas the reinforcement learning algorithms present a very high computational costat training time, they are five orders of magnitude faster than optimal control at deployment time,thus enabling an on-line implementation. Therefore, reinforcement learning achieves a performancesimilar to optimal control at a much lower computational cost at deployment, whilst also presentinga more general framework.