Saved in:
Bibliographic Details
Main Authors: Malladi, Rahath, Harsh, Amol, Sangwan, Arshia, Chauhan, Sunita, Manjanna, Sandeep
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2411.19497
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912137950003200
author Malladi, Rahath
Harsh, Amol
Sangwan, Arshia
Chauhan, Sunita
Manjanna, Sandeep
author_facet Malladi, Rahath
Harsh, Amol
Sangwan, Arshia
Chauhan, Sunita
Manjanna, Sandeep
contents This paper introduces SANGO (Socially Aware Navigation through Grouped Obstacles), a novel method that ensures socially appropriate behavior by dynamically grouping obstacles and adhering to social norms. Using deep reinforcement learning, SANGO trains agents to navigate complex environments leveraging the DBSCAN algorithm for obstacle clustering and Proximal Policy Optimization (PPO) for path planning. The proposed approach improves safety and social compliance by maintaining appropriate distances and reducing collision rates. Extensive experiments conducted in custom simulation environments demonstrate SANGO's superior performance in significantly reducing discomfort (by up to 83.5%), reducing collision rates (by up to 29.4%) and achieving higher successful navigation in dynamic and crowded scenarios. These findings highlight the potential of SANGO for real-world applications, paving the way for advanced socially adept robotic navigation systems.
format Preprint
id arxiv_https___arxiv_org_abs_2411_19497
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SANGO: Socially Aware Navigation through Grouped Obstacles
Malladi, Rahath
Harsh, Amol
Sangwan, Arshia
Chauhan, Sunita
Manjanna, Sandeep
Robotics
Machine Learning
This paper introduces SANGO (Socially Aware Navigation through Grouped Obstacles), a novel method that ensures socially appropriate behavior by dynamically grouping obstacles and adhering to social norms. Using deep reinforcement learning, SANGO trains agents to navigate complex environments leveraging the DBSCAN algorithm for obstacle clustering and Proximal Policy Optimization (PPO) for path planning. The proposed approach improves safety and social compliance by maintaining appropriate distances and reducing collision rates. Extensive experiments conducted in custom simulation environments demonstrate SANGO's superior performance in significantly reducing discomfort (by up to 83.5%), reducing collision rates (by up to 29.4%) and achieving higher successful navigation in dynamic and crowded scenarios. These findings highlight the potential of SANGO for real-world applications, paving the way for advanced socially adept robotic navigation systems.
title SANGO: Socially Aware Navigation through Grouped Obstacles
topic Robotics
Machine Learning
url https://arxiv.org/abs/2411.19497