Saved in:
Bibliographic Details
Main Authors: Bachiller-Burgos, Pilar, Bernardet, Ulysses, Calderita, Luis V., Chhetri, Pranup, Francis, Anthony, Hirose, Noriaki, Pérez, Noé, Shah, Dhruv, Singamaneni, Phani T., Xiao, Xuesu, Manso, Luis J.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.01251
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914226351636480
author Bachiller-Burgos, Pilar
Bernardet, Ulysses
Calderita, Luis V.
Chhetri, Pranup
Francis, Anthony
Hirose, Noriaki
Pérez, Noé
Shah, Dhruv
Singamaneni, Phani T.
Xiao, Xuesu
Manso, Luis J.
author_facet Bachiller-Burgos, Pilar
Bernardet, Ulysses
Calderita, Luis V.
Chhetri, Pranup
Francis, Anthony
Hirose, Noriaki
Pérez, Noé
Shah, Dhruv
Singamaneni, Phani T.
Xiao, Xuesu
Manso, Luis J.
contents This paper presents a joint effort towards the development of a data-driven Social Robot Navigation metric to facilitate benchmarking and policy optimization for ground robots. We compiled a dataset with 4427 trajectories -- 182 real and 4245 simulated -- and presented it to human raters, yielding a total of 4402 rated trajectories after data quality assurance. Notably, we provide the first all-encompassing learned social robot navigation metric, along qualitative and quantitative results, including the test loss achieved, a comparison against hand-crafted metrics, and an ablation study. All data, software, and model weights are publicly available.
format Preprint
id arxiv_https___arxiv_org_abs_2509_01251
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Data-Driven Metrics for Social Robot Navigation Benchmarking
Bachiller-Burgos, Pilar
Bernardet, Ulysses
Calderita, Luis V.
Chhetri, Pranup
Francis, Anthony
Hirose, Noriaki
Pérez, Noé
Shah, Dhruv
Singamaneni, Phani T.
Xiao, Xuesu
Manso, Luis J.
Robotics
This paper presents a joint effort towards the development of a data-driven Social Robot Navigation metric to facilitate benchmarking and policy optimization for ground robots. We compiled a dataset with 4427 trajectories -- 182 real and 4245 simulated -- and presented it to human raters, yielding a total of 4402 rated trajectories after data quality assurance. Notably, we provide the first all-encompassing learned social robot navigation metric, along qualitative and quantitative results, including the test loss achieved, a comparison against hand-crafted metrics, and an ablation study. All data, software, and model weights are publicly available.
title Towards Data-Driven Metrics for Social Robot Navigation Benchmarking
topic Robotics
url https://arxiv.org/abs/2509.01251