[Open-Source] RESPLE: Recursive Spline Estimation for LiDAR-Based Odometry скачать в хорошем качестве

[Open-Source] RESPLE: Recursive Spline Estimation for LiDAR-Based Odometry

GitHub: https://github.com/ASIG-X/RESPLE Authors: Ziyu Cao, William Talbot, and Kailai Li arXiv: https://arxiv.org/abs/2504.11580 We present RESPLE, the first B-spline-based recursive state estimation framework for estimating 6-DoF dynamic motions. RESPLE enables a unified suite of direct LiDAR, LiDAR-inertial, multi-LiDAR, and multi-LiDAR-inertial odometry systems. These four variants have been tested in real-world datasets and our own experiments, covering aerial, wheeled, legged, and wearable platforms operating in indoor, urban, wild environments with diverse LiDAR types. RESPLE achieves superior accuracy and robustness over corresponding state-of-the-art systems with real-time efficiency. We release the code and own experimental datasets to the public.