Cluster aware kalman filtering improves cooperative 3D tracking accuracy

CDKF-Track: Cluster-aware Data-Driven Kalman Filtering for Cooperative 3D Multi-Object Tracking

Robotics

Summary

Tracking multiple objects in 3D space is tricky because sensors can miss things or give confusing information. The authors propose a new method called CDKF-Track that smartly combines data from multiple vehicles to avoid duplicate detections and better understand how objects move. Their method also uses wavelets to smooth out sudden jumps in object positions, making tracking more stable. Tests on real-world data show their approach is significantly more accurate than existing methods.

What this means in practice

  • For autonomous vehicle engineers: Improve 3D tracking accuracy by coordinating multiple vehicles’ LiDAR data and reducing redundant detections for safer navigation.
  • For robot fleet operators: Enhance multi-robot coordination by fusing spatially clustered sensor data and learning real object motion from experience.

Authors

Maria Damanaki, Nikos Piperigkos, Alexandros Gkillas, Aris S. Lalos

Abstract

Multi-Object Tracking (MOT) is essential for EdgeAI perception systems, where accurate object localization and reliable identification enable safe decision-making. Singleagent MOT suffers from occlusions, sensor noise, and partial scene understanding in complex real-world scenarios. While multi-agent systems improve robustness by exploiting shared information, they introduce redundant measurements that lead to false data associations, and still struggle to capture nonlinear object dynamics. To address these challenges, we propose CDKFTrack, a Cluster-aware Data-Driven Kalman Filtering framework for Cooperative 3D MOT. The proposed method first fuses multivehicle 3D LiDAR detections through a Graph Laplacian-based formulation. Then, a cluster-aware redundancy reduction scheme groups spatially related detections and selects representative observations to reduce duplicate inputs to the tracker. The resulting detections are processed by a data-driven Kalman filter that learns object motion dynamics from data, reducing dependence on predefined linear motion assumptions. Furthermore, a wavelet-based temporal refinement module leverages the multiresolution decomposition property of wavelets to attenuate shortterm positional fluctuations and improve trajectory continuity. To the best of our knowledge, CDKF-Track is the first framework to jointly address detection-level fusion redundancy and learnable motion modeling in cooperative 3D MOT. Experimental results on the real-world V2V4Real dataset indicate that CDKF-Track achieves up to 27.99% improvements in tracking accuracy over state-of-the-art multi-agent MOT methods.