Geometry-based Graph Pruning for Lifelong SLAM

TL;DR

This paper introduces a geometry-based graph pruning method for lifelong SLAM that maintains graph sparsity and accuracy over long-term robot operation, significantly improving computational efficiency.

Contribution

It proposes a novel geometric criteria-based pruning algorithm and a robust marginalization approach for lifelong SLAM, enhancing speed and robustness in long-term mapping.

Findings

Achieves 40x speedup in SLAM processing.

Maintains 6cm map error over 25-hour datasets.

Keeps graph sparse while preserving accuracy.

Abstract

Lifelong SLAM considers long-term operation of a robot where already mapped locations are revisited many times in changing environments. As a result, traditional graph-based SLAM approaches eventually become extremely slow due to the continuous growth of the graph and the loss of sparsity. Both problems can be addressed by a graph pruning algorithm. It carefully removes vertices and edges to keep the graph size reasonable while preserving the information needed to provide good SLAM results. We propose a novel method that considers geometric criteria for choosing the vertices to be pruned. It is efficient, easy to implement, and leads to a graph with evenly spread vertices that remain part of the robot trajectory. Furthermore, we present a novel approach of marginalization that is more robust to wrong loop closures than existing methods. The proposed algorithm is evaluated on two publicly available real-world long-term datasets and compared to the unpruned case as well as ground truth. We show that even on a long dataset (25h), our approach manages to keep the graph sparse and the speed high while still providing good accuracy (40 times speed up, 6cm map error compared to unpruned case).