SLICT: Multi-input Multi-scale Surfel-Based Lidar-Inertial Continuous-Time Odometry and Mapping
About
While feature association to a global map has significant benefits, to keep the computations from growing exponentially, most lidar-based odometry and mapping methods opt to associate features with local maps at one voxel scale. Taking advantage of the fact that surfels (surface elements) at different voxel scales can be organized in a tree-like structure, we propose an octree-based global map of multi-scale surfels that can be updated incrementally. This alleviates the need for recalculating, for example, a k-d tree of the whole map repeatedly. The system can also take input from a single or a number of sensors, reinforcing the robustness in degenerate cases. We also propose a point-to-surfel (PTS) association scheme, continuous-time optimization on PTS and IMU preintegration factors, along with loop closure and bundle adjustment, making a complete framework for Lidar-Inertial continuous-time odometry and mapping. Experiments on public and in-house datasets demonstrate the advantages of our system compared to other state-of-the-art methods. To benefit the community, we release the source code and dataset at https://github.com/brytsknguyen/slict.
Key insights
What problem does SLICT address?
SLICT tracks robot motion by fusing lidar and IMU data. It uses a multi-scale surfel global map updated incrementally, avoiding repeated k-d tree recomputation, and supports multiple lidar sensors for robustness in degenerate cases.
How does SLICT differ from prior lidar odometry methods?
SLICT matches lidar points directly to multi-scale surface elements. On the 18 NTU VIRAL sequences, the paper reports the lowest trajectory error in most comparisons with FAST-LIO2, LIO-SAM and MARS, rather than every sequence.
What are the practical uses of SLICT?
SLICT suits autonomous vehicle and UAV navigation needing accurate lidar-inertial odometry and mapping. The authors release source code and datasets as a research tool, but document no production deployment.
What limitations and future work do the authors note?
SLICT averages 165ms per cycle on a core-i7, exceeding the 100ms lidar period, so real-time performance is not guaranteed. Future work includes adaptive PTS association, semantic surfel integration, and more advanced loop closure.