Learning to Avoid Local Minima in Planning for Static Environments - Robotics Institute Carnegie Mellon University

Learning to Avoid Local Minima in Planning for Static Environments

Conference Paper, Proceedings of 27th International Conference on Automated Planning and Scheduling (ICAPS '17), pp. 572 - 576, June, 2017

Abstract

In many robot motion planning problems such as manipulation planning for a personal robot in a kitchen or an industrial manipulator in a warehouse, all motion planning queries are in an environment that is largely static. Consequently, one should be able to improve the performance of a planning algorithm by training on this static environment ahead of operation time. In this work, we propose a method to improve the performance of heuristic search-based motion planners in such environments. The first, learning, phase of our proposed method analyzes search performance on multiple planning episodes to infer local minima zones, i.e, regions where the existing heuristic(s) are weakly correlated with the true cost-to-go. Then, in the planning phase of the method, the learnt local minima are used to modify the original search graph in a way that improves search performance. We prove that our method preserves guarantees on completeness and bounded suboptimality with respect to the original search graph. Experimentally, we observe significant improvements in success rate and planning time for challenging 11 degree-of-freedom mobile manipulation problems.

BibTeX

@conference{Vats-2017-18155,
author = {Shivam Vats and Venkatraman Narayanan and Maxim Likhachev},
title = {Learning to Avoid Local Minima in Planning for Static Environments},
booktitle = {Proceedings of 27th International Conference on Automated Planning and Scheduling (ICAPS '17)},
year = {2017},
month = {June},
pages = {572 - 576},
keywords = {robot motion planning, learning local minima, planning in static environments},
}