Learning from Experience in Manipulation Planning: Setting the Right Goals - Robotics Institute Carnegie Mellon University

Learning from Experience in Manipulation Planning: Setting the Right Goals

Anca Dragan, Geoffrey Gordon, and Siddhartha Srinivasa
Conference Paper, Proceedings of International Symposium on Robotics Research (ISRR '11), pp. 309 -​ 326, July, 2011

Abstract

In this paper, we describe a method of improving trajectory optimization based on predicting good initial guesses from previous experiences. In order to generalize to new situations, we propose a paradigm shift: predicting qualitative attributes of the trajectory that place the initial guess in the basin of attraction of a low-cost solution. We start with a key such attribute, the choice of a goal within a goal set that describes the task, and show the generalization capabilities of our method in extensive experiments on a personal robotics platform.

BibTeX

@conference{Dragan-2011-7320,
author = {Anca Dragan and Geoffrey Gordon and Siddhartha Srinivasa},
title = {Learning from Experience in Manipulation Planning: Setting the Right Goals},
booktitle = {Proceedings of International Symposium on Robotics Research (ISRR '11)},
year = {2011},
month = {July},
pages = {309 -​ 326},
keywords = {learning from experience, trajectory prediction, manipulation planning, trajectory optimization, goal sets},
}