Exploration Without Global Consistency Using Local Volume Consolidation
Titus Cieslewski, Andreas Ziegler, Davide Scaramuzza

TL;DR
This paper introduces a drift-robust 2D exploration map that uses local volumes connected by relative poses, enabling full environment coverage despite state estimate drift, validated through real-world experiments.
Contribution
It proposes a novel local volume consolidation method for exploration maps that does not require global consistency, improving robustness to drift compared to traditional grid-based methods.
Findings
Allows full coverage despite large drift
Outperforms grid-based maps in drift scenarios
Validated in real-world experiments
Abstract
In exploration, the goal is to build a map of an unknown environment. Most state-of-the-art approaches use map representations that require drift-free state estimates to function properly. Real-world state estimators, however, exhibit drift. In this paper, we present a 2D map representation for exploration that is robust to drift. Rather than a global map, it uses local metric volumes connected by relative pose estimates. This pose-graph does not need to be globally consistent. Overlaps between the volumes are resolved locally, rather than on the faulty estimate of space. We demonstrate our representation with a frontier-based exploration approach, evaluate it under different conditions and compare it with a commonly-used grid-based representation. We show that, at the cost of longer exploration time, using the proposed representation allows full coverage of space even for very large…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsRobotics and Sensor-Based Localization · Optimization and Search Problems · Data Management and Algorithms
