• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
STERLING: Self-Supervised Terrain Representation Learning from Unconstrained Robot Experience.
Haresh
Karnan, Elvin Yang, Daniel Farkash, Garrett
Warnell, Peter Stone, and Joydeep Biswas.
In
The Conference on Robot Learning (CoRL), November 2023.
Poster,
Video, Project Website
Terrain awareness, i.e., the ability to identify and distinguish different types of terrain, is a critical ability that robots must have to succeed at autonomous off-road navigation. Current approaches that provide robots with this awareness either rely on labeled data which is expensive to collect, engineered features and cost functions that may not generalize, or expert human demonstrations which may not be available. Towards endowing robots with terrain awareness without these limitations, we introduce Self-supervised TErrain Representation LearnING (STERLING), a novel approach for learning terrain representations that relies solely on easy-to-collect, unconstrained (e.g., non-expert), and unlabeled robot experience, with no additional constraints on data collection. STERLING employs a novel multi-modal self-supervision objective through non-contrastive representation learning to learn relevant terrain representations for terrain-aware navigation. Through physical robot experiments in off-road environments, we evaluate STERLING features on the task of preference-aligned visual navigation and find that STERLING features perform on par with fully-supervised approaches and outperform other state-of-the-art methods with respect to preference alignment. Additionally, we perform a large-scale experiment of semi-autonomously hiking a 3-mile long trail which STERLING completes successfully with only two manual interventions, demonstrating robustness to real-world off-road conditions. Robot experiment videos and more details can be found in the appendix and the project website https://hareshkarnan.github.io/sterling/
@InProceedings{CORL23-karnan,
author = {Haresh Karnan and Elvin Yang and Daniel Farkash and Garrett Warnell and Peter Stone and Joydeep Biswas},
title = {STERLING: Self-Supervised Terrain Representation Learning from Unconstrained Robot Experience},
booktitle = {The Conference on Robot Learning (CoRL)},
location = {Atlanta, USA},
month = {November},
year = {2023},
abstract = {Terrain awareness, i.e., the ability to identify and distinguish different types of terrain, is a critical ability that robots must have to succeed at autonomous off-road navigation. Current approaches that provide robots with this awareness either rely on labeled data which is expensive to collect, engineered features and cost functions that may not generalize, or expert human demonstrations which may not be available. Towards endowing robots with terrain awareness without these limitations, we introduce Self-supervised TErrain Representation LearnING (STERLING), a novel approach for learning terrain representations that relies solely on easy-to-collect, unconstrained (e.g., non-expert), and unlabeled robot experience, with no additional constraints on data collection. STERLING employs a novel multi-modal self-supervision objective through non-contrastive representation learning to learn relevant terrain representations for terrain-aware navigation. Through physical robot experiments in off-road environments, we evaluate STERLING features on the task of preference-aligned visual navigation and find that STERLING features perform on par with fully-supervised approaches and outperform other state-of-the-art methods with respect to preference alignment. Additionally, we perform a large-scale experiment of semi-autonomously hiking a 3-mile long trail which STERLING completes successfully with only two manual interventions, demonstrating robustness to real-world off-road conditions. Robot experiment videos and more details can be found in the appendix and the project website https://hareshkarnan.github.io/sterling/},
wwwnote={<a href="https://drive.google.com/file/d/1QHF8YUgIHwvK6pvGwa0zyNKLQKlCLLxA/view?usp=sharing">Poster</a>, <a href="https://youtu.be/7WI41DfJQ2k">Video</a>, <a href="https://hareshkarnan.github.io/sterling/">Project Website</a>}
}
Generated by bib2html.pl (written by Patrick Riley ) on Wed Apr 17, 2024 18:42:50