Predicting Topological Maps for Visual Navigation in Unexplored Environments

TL;DR

This paper introduces a probabilistic layout graph-based navigation system enabling robots to explore and navigate unseen environments more efficiently by predicting and utilizing scene graphs.

Contribution

It presents a novel three-stage framework combining perception, prediction, and navigation using probabilistic scene graphs for unexplored environments.

Findings

Improved success rate in navigation tasks in unseen environments

Faster goal achievement compared to prior methods

Effective use of probabilistic scene graphs for exploration

Abstract

We propose a robotic learning system for autonomous exploration and navigation in unexplored environments. We are motivated by the idea that even an unseen environment may be familiar from previous experiences in similar environments. The core of our method, therefore, is a process for building, predicting, and using probabilistic layout graphs for assisting goal-based visual navigation. We describe a navigation system that uses the layout predictions to satisfy high-level goals (e.g. "go to the kitchen") more rapidly and accurately than the prior art. Our proposed navigation framework comprises three stages: (1) Perception and Mapping: building a multi-level 3D scene graph; (2) Prediction: predicting probabilistic 3D scene graph for the unexplored environment; (3) Navigation: assisting navigation with the graphs. We test our framework in Matterport3D and show more success and efficient navigation in unseen environments.