CaRLi-V: Camera-RADAR-LiDAR Point-Wise 3D Velocity Estimation

arXiv cs.RO / 4/13/2026

💬 OpinionIdeas & Deep AnalysisTools & Practical UsageModels & Research

共有:

Key Points

The paper introduces CaRLi-V, a fusion pipeline combining RADAR, LiDAR, and camera data to estimate point-wise 3D velocities in dynamic environments involving non-rigid moving agents.
It proposes a RADAR-specific “velocity cube” representation that densely encodes RADAR radial velocities, supporting robust radial velocity extraction.
The method estimates tangential velocity using optical flow and recovers point-wise range from LiDAR via a closed-form solution to produce dense 3D velocity estimates.
CaRLi-V is released as an open-source ROS2 package, and the authors report field testing on a custom dataset with low error metrics that outperform prior scene flow approaches.

Abstract

Accurate point-wise velocity estimation in 3D is crucial for robot interaction with non-rigid dynamic agents, enabling robust performance in path planning, collision avoidance, and object manipulation in dynamic environments. To this end, this paper proposes a novel RADAR, LiDAR, and camera fusion pipeline for point-wise 3D velocity estimation named CaRLi-V. This pipeline leverages raw RADAR measurements to create a novel RADAR representation, the velocity cube, which densely encodes RADAR radial velocities. By combining the velocity cube for radial velocity extraction, optical flow for tangential velocity estimation, and LiDAR for point-wise range measurements through a closed-form solution, our approach can produce 3D velocity estimates for a dense array of points. Developed as an open-source ROS2 package, CaRLi-V has been field-tested on a custom dataset and achieves low velocity error metrics relative to ground truth while outperforming state-of-the-art scene flow methods.