English

Learning Skills to Patch Plans Based on Inaccurate Models

Robotics 2021-05-18 v1

Abstract

Planners using accurate models can be effective for accomplishing manipulation tasks in the real world, but are typically highly specialized and require significant fine-tuning to be reliable. Meanwhile, learning is useful for adaptation, but can require a substantial amount of data collection. In this paper, we propose a method that improves the efficiency of sub-optimal planners with approximate but simple and fast models by switching to a model-free policy when unexpected transitions are observed. Unlike previous work, our method specifically addresses when the planner fails due to transition model error by patching with a local policy only where needed. First, we use a sub-optimal model-based planner to perform a task until model failure is detected. Next, we learn a local model-free policy from expert demonstrations to complete the task in regions where the model failed. To show the efficacy of our method, we perform experiments with a shape insertion puzzle and compare our results to both pure planning and imitation learning approaches. We then apply our method to a door opening task. Our experiments demonstrate that our patch-enhanced planner performs more reliably than pure planning and with lower overall sample complexity than pure imitation learning.

Keywords

Cite

@article{arxiv.2009.13732,
  title  = {Learning Skills to Patch Plans Based on Inaccurate Models},
  author = {Alex LaGrassa and Steven Lee and Oliver Kroemer},
  journal= {arXiv preprint arXiv:2009.13732},
  year   = {2021}
}

Comments

8 pages, 10 figures, accepted to Intelligent Robots and Systems (IROS) 2020

R2 v1 2026-06-23T18:51:57.732Z