ICRA 2012 Paper Abstract

Close

Paper WeA01.2

Zucker, Matthew (Swarthmore College), Bagnell, James (Carnegie Mellon University)

Reinforcement Planning: RL for Optimal Planners

Scheduled for presentation during the Regular Session "Learning and Adaptation Control of Robotic Systems II" (WeA01), Wednesday, May 16, 2012, 08:45−09:00, Meeting Room 1 (Mini-sota)

2012 IEEE International Conference on Robotics and Automation, May 14-18, 2012, RiverCentre, Saint Paul, Minnesota, USA

This information is tentative and subject to change. Compiled on November 18, 2017

Keywords Learning and Adaptive Systems, Motion and Path Planning, Robust/Adaptive Control of Robotic Systems

Abstract

Search based planners such as A* and Dijkstra’s algorithm are proven methods for guiding today’s robotic systems. Although such planners are typically based upon a coarse approximation of reality, they are nonetheless valuable due to their ability to reason about the future, and to generalize to previously unseen scenarios. However, encoding the desired behavior of a system into the underlying cost function used by the planner can be a tedious and error-prone task. We introduce Reinforcement Planning, which extends gradient based reinforcement learning algorithms to automatically learn useful surrogate cost functions for optimal planners. Reinforcement Planning presents several advantages over other learning approaches to planning in that it is not limited by the expertise of a human demonstrator, and that it acknowledges the domain of the planner is a simplified model of the world. We demonstrate the effectiveness of our method in learning to solve a noisy physical simulation of the well-known “marble maze” toy.

 

 

Technical Content © IEEE Robotics & Automation Society

This site is protected by copyright and trademark laws under US and International law.
All rights reserved. © 2002-2017 PaperCept, Inc.
Page generated 2017-11-18  19:10:53 PST  Terms of use