IPPON: Common Sense Guided Informative Path Planning for Object Goal Navigation

TL;DR

IPPON introduces a novel semantic mapping and path planning method guided by common sense priors, significantly improving object goal navigation efficiency in unexplored environments for robots.

Contribution

The paper presents a new informative path planning approach that integrates semantic probability mapping with common sense priors, achieving state-of-the-art results in object goal navigation.

Findings

Achieved over 20% improvement in SPL and Soft SPL metrics.

Outperformed existing methods in the Habitat ObjectNav Challenge 2023.

Validated effectiveness on real robots.

Abstract

Navigating efficiently to an object in an unexplored environment is a critical skill for general-purpose intelligent robots. Recent approaches to this object goal navigation problem have embraced a modular strategy, integrating classical exploration algorithms-notably frontier exploration-with a learned semantic mapping/exploration module. This paper introduces a novel informative path planning and 3D object probability mapping approach. The mapping module computes the probability of the object of interest through semantic segmentation and a Bayes filter. Additionally, it stores probabilities for common objects, which semantically guides the exploration based on common sense priors from a large language model. The planner terminates when the current viewpoint captures enough voxels identified with high confidence as the object of interest. Although our planner follows a zero-shot approach, it achieves state-of-the-art performance as measured by the Success weighted by Path Length (SPL) and Soft SPL in the Habitat ObjectNav Challenge 2023, outperforming other works by more than 20%. Furthermore, we validate its effectiveness on real robots. Project webpage: https://ippon-paper.github.io/