Learning to Explore using Active Neural SLAM

TL;DR

This paper introduces Active Neural SLAM, a hierarchical learning-based approach for exploring 3D environments that combines classical planning with learned modules, improving robustness and efficiency in navigation tasks.

Contribution

It presents a modular, hierarchical framework that integrates classical path planning with learned SLAM and policies, enabling flexible, robust exploration in 3D environments.

Findings

Outperforms previous learning and geometry-based methods in simulated environments

Successfully transfers to PointGoal navigation tasks

Won the CVPR 2019 Habitat PointGoal Navigation Challenge

Abstract

This work presents a modular and hierarchical approach to learn policies for exploring 3D environments, called `Active Neural SLAM'. Our approach leverages the strengths of both classical and learning-based methods, by using analytical path planners with learned SLAM module, and global and local policies. The use of learning provides flexibility with respect to input modalities (in the SLAM module), leverages structural regularities of the world (in global policies), and provides robustness to errors in state estimation (in local policies). Such use of learning within each module retains its benefits, while at the same time, hierarchical decomposition and modular training allow us to sidestep the high sample complexities associated with training end-to-end policies. Our experiments in visually and physically realistic simulated 3D environments demonstrate the effectiveness of our approach over past learning and geometry-based approaches. The proposed model can also be easily transferred to the PointGoal task and was the winning entry of the CVPR 2019 Habitat PointGoal Navigation Challenge.