LT-mapper: A Modular Framework for LiDAR-based Lifelong Mapping
About
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).
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Multi-robot map merging | Outdoor A01-H01 | RMSE (m)4.328 | 6 | |
| Multi-robot map merging | NTU 01-02 | RMSE (m)3.679 | 6 | |
| Multi-robot map merging | NTU 01-10 | RMSE (m)3.34 | 6 | |
| Multi-robot map merging | NTU Avg. | RMSE (m)3.51 | 6 | |
| Multi-robot map merging | Outdoor A01-A02 | RMSE (m)121.2 | 6 | |
| Multi-robot map merging | Outdoor Avg. | RMSE (m)122 | 6 | |
| Multi-robot map merging | Outdoor A01-S01 | RMSE Error270.8 | 6 | |
| Multi-robot map merging | Outdoor A01-S02 | RMSE (m)86.612 | 6 | |
| Change Detection | LT-ParkingLot seq. 03 → 04 | PD Precision9.4 | 4 | |
| Multi-robot map merging | Roundabout 01-02 | e_rmse (m)7.787 | 3 |