TRLO: An Efficient LiDAR Odometry with 3D Dynamic Object Tracking and Removal

TL;DR

TRLO introduces a dynamic LiDAR odometry method that effectively detects and removes moving objects using deep learning and Kalman filtering, resulting in improved localization accuracy and cleaner maps in urban environments.

Contribution

The paper presents a novel LiDAR odometry approach that combines deep learning-based dynamic object detection, a UKF tracker, and a hash-based keyframe database for efficient and accurate state estimation in dynamic scenes.

Findings

Achieves higher localization accuracy on KITTI and UrbanLoco datasets.

Produces cleaner point cloud maps by removing dynamic objects.

Demonstrates robustness in urban environments with moving objects.

Abstract

Simultaneous state estimation and mapping is an essential capability for mobile robots working in dynamic urban environment. The majority of existing SLAM solutions heavily rely on a primarily static assumption. However, due to the presence of moving vehicles and pedestrians, this assumption does not always hold, leading to localization accuracy decreased and maps distorted. To address this challenge, we propose TRLO, a dynamic LiDAR odometry that efficiently improves the accuracy of state estimation and generates a cleaner point cloud map. To efficiently detect dynamic objects in the surrounding environment, a deep learning-based method is applied, generating detection bounding boxes. We then design a 3D multi-object tracker based on Unscented Kalman Filter (UKF) and nearest neighbor (NN) strategy to reliably identify and remove dynamic objects. Subsequently, a fast two-stage iterative nearest point solver is employed to solve the state estimation using cleaned static point cloud. Note that a novel hash-based keyframe database management is proposed for fast access to search keyframes. Furthermore, all the detected object bounding boxes are leveraged to impose posture consistency constraint to further refine the final state estimation. Extensive evaluations and ablation studies conducted on the KITTI and UrbanLoco datasets demonstrate that our approach not only achieves more accurate state estimation but also generates cleaner maps, compared with baselines.