Group-$k$ Consistent Measurement Set Maximization for Robust Outlier Detection

TL;DR

This paper introduces group-$k$ consistency maximization (G$k$CM), a novel method for robust measurement selection in SLAM that considers group-level consistency rather than pairwise checks, improving outlier detection.

Contribution

The paper proposes G$k$CM, a new approach that estimates the largest internally group-$k$ consistent measurement set by formulating it as a maximum clique problem, advancing robustness in SLAM outlier detection.

Findings

G$k$CM outperforms pairwise consistency methods in simulated tests.

G$k$CM effectively identifies the largest consistent measurement groups.

The approach is adaptable using existing maximum clique algorithms.

Abstract

This paper presents a method for the robust selection of measurements in a simultaneous localization and mapping (SLAM) framework. Existing methods check consistency or compatibility on a pairwise basis, however many measurement types are not sufficiently constrained in a pairwise scenario to determine if either measurement is inconsistent with the other. This paper presents group-$k$ consistency maximization (G$k$CM) that estimates the largest set of measurements that is internally group-$k$ consistent. Solving for the largest set of group-$k$ consistent measurements can be formulated as an instance of the maximum clique problem on generalized graphs and can be solved by adapting current methods. This paper evaluates the performance of G$k$CM using simulated data and compares it to pairwise consistency maximization (PCM) presented in previous work.