Reinforcement Learning of Depth Stabilization with a Micro Diving Agent 

ICRA 2018 Spotlight Video
Interactive Session Thu AM Pod P.4
Authors: Brinkmann, Gerrit; Duecker, Daniel Andre; Kreuzer, Edwin; Solowjow, Eugen; Bessa, Wallace M.
Title: Reinforcement Learning of Depth Stabilization with a Micro Diving Agent
Abstract:
Reinforcement learning (RL) allows robots to solve control tasks through interaction with their environment. In this paper we study a value-function- and model-based RL framework, which is suitable for computationaly limited robots. We develop a diving agent, which uses the RL algorithm for underwater depth stabilization. Simulations and experiments with the micro diving agent demonstrate its ability to learn the depth stabilization task.