Hierarchies of Planning and Reinforcement Learning for Robot Navigation

Open Access
Authors
Publication date 2021
Book title 2021 IEEE International Conference on Robotics and Automation (ICRA 2021)
Book subtitle May 31-June 4, 2021, Xi'an, China
ISBN
  • 9781728190785
ISBN (electronic)
  • 9781728190778
Event 2021 IEEE International Conference on Robotics and Automation
Pages (from-to) 10682-10688
Number of pages 7
Publisher Piscataway, NJ: IEEE
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract
Solving robotic navigation tasks via reinforcement learning (RL) is challenging due to their sparse reward and long decision horizon nature. However, in many navigation tasks, high-level (HL) task representations, like a rough floor plan, are available. Previous work has demonstrated efficient learning by hierarchal approaches consisting of path planning in the HL representation and using sub-goals derived from the plan to guide the RL policy in the source task. However, these approaches usually neglect the complex dynamics and sub-optimal sub-goal-reaching capabilities of the robot during planning. This work overcomes these limitations by proposing a novel hierarchical framework that utilizes a trainable planning policy for the HL representation. Thereby robot capabilities and environment conditions can be learned utilizing collected rollout data. We specifically introduce a planning policy based on value iteration with a learned transition model (VI-RL). In simulated robotic navigation tasks, VI-RL results in consistent strong improvement over vanilla RL, is on par with vanilla hierarchal RL on single layouts but more broadly applicable to multiple layouts, and is on par with trainable HL path planning baselines except for a parking task with difficult non-holonomic dynamics where it shows marked improvements.
Document type Conference contribution
Language English
Published at https://doi.org/10.48550/arXiv.2109.11178 https://doi.org/10.1109/ICRA48506.2021.9561151
Other links https://www.proceedings.com/60494.html
Downloads
2109.11178 (Accepted author manuscript)
Permalink to this page
Back