LT-mapper: A Modular Framework for LiDAR-based Lifelong Mapping

TL;DR

This paper introduces LT-mapper, an open-source modular framework for LiDAR-based lifelong mapping that effectively manages long-term 3D maps in dynamic urban environments through multi-session SLAM and change detection.

Contribution

It presents a novel modular framework that combines multi-session SLAM with dynamic change detection and management, handling large-scale, long-term urban mapping without requiring precise initial alignment.

Findings

Effective change detection across multiple temporal gaps

Automatic object segregation in large-scale point clouds

Reliable long-term mapping in real-world urban environments

Abstract

Long-term 3D map management is a fundamental capability required by a robot to reliably navigate in the non-stationary real-world. This paper develops open-source, modular, and readily available LiDAR-based lifelong mapping for urban sites. This is achieved by dividing the problem into successive subproblems: multi-session SLAM (MSS), high/low dynamic change detection, and positive/negative change management. The proposed method leverages MSS and handles potential trajectory error; thus, good initial alignment is not required for change detection. Our change management scheme preserves efficacy in both memory and computation costs, providing automatic object segregation from a large-scale point cloud map. We verify the framework's reliability and applicability even under permanent year-level variation, through extensive real-world experiments with multiple temporal gaps (from day to year).