Realistic PointGoal Navigation via Auxiliary Losses and Information Bottleneck

TL;DR

This paper introduces a novel training approach for realistic PointGoal Navigation that uses auxiliary losses and an information bottleneck to improve localization without ground-truth data, outperforming existing methods in noisy environments.

Contribution

The paper presents a new architecture and training paradigm that leverages auxiliary losses and privileged information with an information bottleneck to enhance localization in navigation tasks.

Findings

Outperforms baselines by 18-21% in semi-idealized settings.

Achieves 15-20% improvement in Success and SPL in realistic noisy environments.

Enables better self-localization while maintaining strong navigation performance.

Abstract

We propose a novel architecture and training paradigm for training realistic PointGoal Navigation -- navigating to a target coordinate in an unseen environment under actuation and sensor noise without access to ground-truth localization. Specifically, we find that the primary challenge under this setting is learning localization -- when stripped of idealized localization, agents fail to stop precisely at the goal despite reliably making progress towards it. To address this we introduce a set of auxiliary losses to help the agent learn localization. Further, we explore the idea of treating the precise location of the agent as privileged information -- it is unavailable during test time, however, it is available during training time in simulation. We grant the agent restricted access to ground-truth localization readings during training via an information bottleneck. Under this setting, the agent incurs a penalty for using this privileged information, encouraging the agent to only leverage this information when it is crucial to learning. This enables the agent to first learn navigation and then learn localization instead of conflating these two objectives in training. We evaluate our proposed method both in a semi-idealized (noiseless simulation without Compass+GPS) and realistic (addition of noisy simulation) settings. Specifically, our method outperforms existing baselines on the semi-idealized setting by 18\%/21\% SPL/Success and by 15\%/20\% SPL in the realistic setting. Our improved Success and SPL metrics indicate our agent's improved ability to accurately self-localize while maintaining a strong navigation policy. Our implementation can be found at https://github.com/NicoGrande/habitat-pointnav-via-ib.