The framework of mixed observable Markov decision processes (MOMDP) models many robotic domains in which some state variables are fully observable while others are not. In this work, we identify a significant subclass of MOMDPs defined by how actions influence the fully observable components of the state and how those, in turn, influence the partially observable components and the rewards. This unique property allows for a two-level hierarchical approach we call HIerarchical Reinforcement Learning under Mixed Observability (HILMO), which restricts partial observability to the top level while the bottom level remains fully observable, enabling higher learning efficiency. The top level produces desired goals to be reached by the bottom level until the task is solved. We further develop theoretical guarantees to show that our approach can achieve optimal and quasi-optimal behavior under mild assumptions. Empirical results on long-horizon continuous control tasks demonstrate the efficacy and efficiency of our approach in terms of improved success rate, sample efficiency, and wall-clock training time. We also deploy policies learned in simulation on a real robot.
Hierarchical Reinforcement Learning Under Mixed Observability
Springer Proceedings in Advanced Robotics
International Workshop on the Algorithmic Foundations of Robotics ; 2022 ; , MD, USA June 22, 2022 - June 24, 2022
2022-12-15
17 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
Hierarchical Reinforcement Learning Under Mixed Observability
TIBKAT | 2023
|AUTONOMOUS BEHAVIOR GENERATION WITH HIERARCHICAL REINFORCEMENT LEARNING
Europäisches Patentamt | 2021
|