A Behavioral Approach to Visual Navigation with Graph Localization Networks

TL;DR

This paper presents a behavioral visual navigation method using graph neural networks and topological maps, enabling robots to navigate with only visual input, outperforming baselines in simulated environments.

Contribution

It introduces a novel approach combining graph neural networks and primitive behaviors for visual navigation based on topological maps.

Findings

Outperforms relevant baselines in Gibson simulator

Effective in both seen and unseen environments

Utilizes primitive behaviors for navigation

Abstract

Inspired by research in psychology, we introduce a behavioral approach for visual navigation using topological maps. Our goal is to enable a robot to navigate from one location to another, relying only on its visual input and the topological map of the environment. We propose using graph neural networks for localizing the agent in the map, and decompose the action space into primitive behaviors implemented as convolutional or recurrent neural networks. Using the Gibson simulator, we verify that our approach outperforms relevant baselines and is able to navigate in both seen and unseen environments.