• Classified by Topic • Classified by Publication Type • Sorted by Date • Sorted by First Author Last Name • Classified by Funding Source •
Rethinking Social Robot Navigation: Leveraging the Best of Two Worlds.
Amir Hossain Raj, Zichao
Hu, Haresh Karnan, Rohan Chandra, Amirreza Payandeh, Luisa Mao, Peter
Stone, Joydeep Biswas, and and Xuesu
Xiao.
In International Conference on Robotics and Automation, May 2024.
Empowering robots to navigate in a socially compliant manner is essential for theacceptance of robots moving in human-inhabited environments. Previously,roboticists have developed geometric navigation systems with decades of empiricalvalidation to achieve safety and efficiency. However, the many complex factors ofsocial compliance make geometric navigation systems hard to adapt to socialsituations, where no amount of tuning enables them to be both safe (people aretoo unpredictable) and efficient (the frozen robot problem). With recent advancesin deep learning approaches, the common reaction has been to entirely discardthese classical navigation systems and start from scratch, building a completelynew learning-based social navigation planner. In this work, we find that thisreaction is unnecessarily extreme: using a large-scale real-world socialnavigation dataset, SCAND, we find that geometric systems can produce trajectoryplans that align with the human demonstrations in a large number of socialsituations. We, therefore, ask if we can rethink the social robot navigationproblem by leveraging the advantages of both geometric and learning-basedmethods. We validate this hybrid paradigm through a proof-of-concept experiment,in which we develop a hybrid planner that switches between geometric andlearning-based planning. Our experiments on both SCAND and two physical robotsshow that the hybrid planner can achieve better social compliance compared tousing either the geometric or learning-based approach alone.
@InProceedings{karnansocial2024,
author = {Amir Hossain Raj and Zichao Hu and Haresh Karnan and Rohan Chandra and Amirreza Payandeh and Luisa Mao and Peter Stone and Joydeep Biswas and and Xuesu Xiao},
title = {Rethinking Social Robot Navigation: Leveraging the Best of Two Worlds},
booktitle = {International Conference on Robotics and Automation},
year = {2024},
month = {May},
location = {Yokohama, Japan},
abstract = {Empowering robots to navigate in a socially compliant manner is essential for the
acceptance of robots moving in human-inhabited environments. Previously,
roboticists have developed geometric navigation systems with decades of empirical
validation to achieve safety and efficiency. However, the many complex factors of
social compliance make geometric navigation systems hard to adapt to social
situations, where no amount of tuning enables them to be both safe (people are
too unpredictable) and efficient (the frozen robot problem). With recent advances
in deep learning approaches, the common reaction has been to entirely discard
these classical navigation systems and start from scratch, building a completely
new learning-based social navigation planner. In this work, we find that this
reaction is unnecessarily extreme: using a large-scale real-world social
navigation dataset, SCAND, we find that geometric systems can produce trajectory
plans that align with the human demonstrations in a large number of social
situations. We, therefore, ask if we can rethink the social robot navigation
problem by leveraging the advantages of both geometric and learning-based
methods. We validate this hybrid paradigm through a proof-of-concept experiment,
in which we develop a hybrid planner that switches between geometric and
learning-based planning. Our experiments on both SCAND and two physical robots
show that the hybrid planner can achieve better social compliance compared to
using either the geometric or learning-based approach alone.
},
}
Generated by bib2html.pl (written by Patrick Riley ) on Wed Jan 15, 2025 08:40:53