Stubborn: A Strong Baseline for Indoor Object Navigation

TL;DR

This paper introduces 'Stubborn', a robust, learning-free baseline for indoor object navigation that improves exploration, object recognition, and obstacle avoidance, outperforming previous methods on the Habitat Challenge.

Contribution

The paper presents a semantic-agnostic exploration strategy, a temporal information integration method for better object identification, and a multi-scale collision map for obstacle detection, all without learning.

Findings

Outperforms previous methods on Habitat Challenge

Semantic-agnostic exploration is highly effective

Multi-scale collision map reduces trapping in small regions

Abstract

We present a strong baseline that surpasses the performance of previously published methods on the Habitat Challenge task of navigating to a target object in indoor environments. Our method is motivated from primary failure modes of prior state-of-the-art: poor exploration, inaccurate object identification, and agent getting trapped due to imprecise map construction. We make three contributions to mitigate these issues: (i) First, we show that existing map-based methods fail to effectively use semantic clues for exploration. We present a semantic-agnostic exploration strategy (called Stubborn) without any learning that surprisingly outperforms prior work. (ii) We propose a strategy for integrating temporal information to improve object identification. (iii) Lastly, due to inaccurate depth observation the agent often gets trapped in small regions. We develop a multi-scale collision map for obstacle identification that mitigates this issue.