SCORE: A Second-Order Conic Initialization for Range-Aided SLAM

TL;DR

This paper introduces SCORE, a convex second-order cone relaxation method for range-aided SLAM that provides reliable initializations, enabling efficient and high-quality solutions for large-scale multi-robot localization and mapping problems.

Contribution

The paper presents the first convex relaxation approach for RA-SLAM using SOCP, improving scalability and initialization quality for complex multi-robot SLAM tasks.

Findings

SCORE enables efficient recovery of high-quality solutions in large-scale RA-SLAM.

The method is scalable and effective in real-world and simulated multi-robot scenarios.

SCORE outperforms traditional non-convex initialization methods in challenging environments.

Abstract

We present a novel initialization technique for the range-aided simultaneous localization and mapping (RA-SLAM) problem. In RA-SLAM we consider measurements of point-to-point distances in addition to measurements of rigid transformations to landmark or pose variables. Standard formulations of RA-SLAM approach the problem as non-convex optimization, which requires a good initialization to obtain quality results. The initialization technique proposed here relaxes the RA-SLAM problem to a convex problem which is then solved to determine an initialization for the original, non-convex problem. The relaxation is a second-order cone program (SOCP), which is derived from a quadratically constrained quadratic program (QCQP) formulation of the RA-SLAM problem. As a SOCP, the method is highly scalable. We name this relaxation Second-order COnic RElaxation for RA-SLAM (SCORE). To our knowledge, this work represents the first convex relaxation for RA-SLAM. We present real-world and simulated experiments which show SCORE initialization permits the efficient recovery of quality solutions for a variety of challenging single- and multi-robot RA-SLAM problems with thousands of poses and range measurements.