As robots become increasingly capable of manipulation and long-term autonomy, long-horizon task and motion planning problems are becoming increasingly important. A key challenge in such problems is that early actions in the plan may make future actions infeasible. When reaching a dead-end in the search, most existing planners use backtracking, which exhaustively reevaluates motionlevel actions, often resulting in inefficient planning, especially when the search depth is large. In this paper, we propose to learn backjumping heuristics which identify the culprit action directly using supervised learning models to guide the task-level search. Based on evaluations on two different tasks, we find that our method significantly improves planning efficiency compared to backtracking and also generalizes to problems with novel numbers of objects.

PDF

In Proceedings of the 6th Conference on Robot Learning (CoRL 2022), Auckland, New Zealand, December 2022.

@inproceedings{CoRL2022-Sung, title={Learning to Correct Mistakes: Backjumping in Long-Horizon Task and Motion Planning}, author={Yoonchang Sung and Zizhao Wang and Peter Stone}, booktitle={Proceedings of the 6th Conference on Robot Learning (CoRL 2022)}, month={December}, address={Auckland, New Zealand}, url="http://www.cs.utexas.edu/users/ai-lab?CoRL2022-Sung", year={2022} }