Exploiting Proximity-Aware Tasks for Embodied Social Navigation

TL;DR

This paper introduces an end-to-end social navigation system for embodied agents that uses proximity-aware tasks to incorporate common-sense social behaviors, validated through a new evaluation protocol in simulated environments.

Contribution

It proposes a novel architecture leveraging proximity-aware tasks to improve social navigation and introduces a specialized evaluation protocol for detailed analysis.

Findings

Effective navigation among humans in simulated environments

Incorporation of common-sense social behaviors improves performance

New evaluation protocol captures fine-grained human-robot interactions

Abstract

Learning how to navigate among humans in an occluded and spatially constrained indoor environment, is a key ability required to embodied agent to be integrated into our society. In this paper, we propose an end-to-end architecture that exploits Proximity-Aware Tasks (referred as to Risk and Proximity Compass) to inject into a reinforcement learning navigation policy the ability to infer common-sense social behaviors. To this end, our tasks exploit the notion of immediate and future dangers of collision. Furthermore, we propose an evaluation protocol specifically designed for the Social Navigation Task in simulated environments. This is done to capture fine-grained features and characteristics of the policy by analyzing the minimal unit of human-robot spatial interaction, called Encounter. We validate our approach on Gibson4+ and Habitat-Matterport3D datasets.