IROS 2015 Paper Abstract

Close

Paper ThAP.11

Mouret, Jean-Baptiste (Inria), Clune, Jeff (University of Wyoming)

Late-Breaking Abstract: Illuminating Search Spaces in Robotics

Scheduled for presentation during the Poster session "Late Breaking Posters" (ThAP), Thursday, October 1, 2015, 09:45−10:00, Saal G1

2015 IEEE/RSJ International Conference on Intelligent Robots and Systems, Sept 28 - Oct 03, 2015, Congress Center Hamburg, Hamburg, Germany

This information is tentative and subject to change. Compiled on July 19, 2019

Keywords Evolutionary Robotics, Robot Learning, Soft-bodied Robots 1

Abstract

Nearly all science and engineering fields use search algorithm}, which automatically explore a search space to find high-performing solutions: chemists search through the space of molecules to discover new drugs; engineers search for stronger, cheaper, safer designs; scientists search for models that best explain data; and roboticists typically search for the best parameters for control laws or the best kinematic design.

The goal of search algorithms has traditionally been to return the single highest-performing solution in a search space. Here we describe a new, fundamentally different type of algorithm that is more useful because it provides a holistic view of how high-performing solutions are distributed throughout a search space. It creates a map of high-performing solutions at each point in a space defined by dimensions of variation that a user gets to choose (Fig. ref{fig:concept}). Typical dimensions are a subset of the parameters to tune, or quantifiable measures about the robot, like its overall weight or how dynamic its gait is. We call these high-performing solutions the elites of the search space, that is, each of them is the best achievable solution for each possible coordinate of the feature space.

Hence, this Multi-dimensional Archive of Phenotypic Elites (MAP-Elites) algorithm illuminates search spaces by searching for the elites, allowing researchers to understand how interesting attributes of solutions combine to affect performance, either positively or, equally of interest, negatively. This new algorithm can be an effective alternative to the traditional grid search approaches, which do not scale to high-dimensional search spaces, and to black-box optimization algorithms, which provide only a limited view of the search space.

 

 

Technical Content © IEEE Robotics & Automation Society


This site is protected by copyright and trademark laws under US and International law.
All rights reserved. © 2002-2019 PaperCept, Inc.
Page generated 2019-07-19  14:07:06 PST  Terms of use