Exciting Contact Modes in Differentiable Simulations for Robot Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sathyanarayan, Hrishikesh, Abraham, Ian
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915036195192832
author Sathyanarayan, Hrishikesh
Abraham, Ian
author_facet Sathyanarayan, Hrishikesh
Abraham, Ian
contents In this paper, we explore an approach to actively plan and excite contact modes in differentiable simulators as a means to tighten the sim-to-real gap. We propose an optimal experimental design approach derived from information-theoretic methods to identify and search for information-rich contact modes through the use of contact-implicit optimization. We demonstrate our approach on a robot parameter estimation problem with unknown inertial and kinematic parameters which actively seeks contacts with a nearby surface. We show that our approach improves the identification of unknown parameter estimates over experimental runs by an estimate error reduction of at least $\sim 84\%$ when compared to a random sampling baseline, with significantly higher information gains.
format Preprint
id arxiv_https___arxiv_org_abs_2411_10935
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Exciting Contact Modes in Differentiable Simulations for Robot Learning
Sathyanarayan, Hrishikesh
Abraham, Ian
Robotics
Information Theory
In this paper, we explore an approach to actively plan and excite contact modes in differentiable simulators as a means to tighten the sim-to-real gap. We propose an optimal experimental design approach derived from information-theoretic methods to identify and search for information-rich contact modes through the use of contact-implicit optimization. We demonstrate our approach on a robot parameter estimation problem with unknown inertial and kinematic parameters which actively seeks contacts with a nearby surface. We show that our approach improves the identification of unknown parameter estimates over experimental runs by an estimate error reduction of at least $\sim 84\%$ when compared to a random sampling baseline, with significantly higher information gains.
title Exciting Contact Modes in Differentiable Simulations for Robot Learning
topic Robotics
Information Theory
url https://arxiv.org/abs/2411.10935