Learning hierarchical relationships for object-goal navigation

TL;DR

This paper introduces MJOLNIR, a hierarchical navigation algorithm that leverages object-object relationships and context to improve object-goal navigation efficiency and generalization across environments.

Contribution

The paper presents MJOLNIR, a novel hierarchical navigation method that incorporates object relationships and context, outperforming existing methods in success rate and convergence speed.

Findings

Achieves 82.9% success rate improvement over state-of-the-art.

Achieves 93.5% SPL improvement over existing methods.

Learns faster and avoids overfitting compared to other algorithms.

Abstract

Direct search for objects as part of navigation poses a challenge for small items. Utilizing context in the form of object-object relationships enable hierarchical search for targets efficiently. Most of the current approaches tend to directly incorporate sensory input into a reward-based learning approach, without learning about object relationships in the natural environment, and thus generalize poorly across domains. We present Memory-utilized Joint hierarchical Object Learning for Navigation in Indoor Rooms (MJOLNIR), a target-driven navigation algorithm, which considers the inherent relationship between target objects, and the more salient contextual objects occurring in its surrounding. Extensive experiments conducted across multiple environment settings show an $82.9\%$ and $93.5\%$ gain over existing state-of-the-art navigation methods in terms of the success rate (SR), and success weighted by path length (SPL), respectively. We also show that our model learns to converge much faster than other algorithms, without suffering from the well-known overfitting problem. Additional details regarding the supplementary material and code are available at https://sites.google.com/eng.ucsd.edu/mjolnir.