Li, Chengmin (Tongji university), Wang, Junhua (Tongji University), Wang, Tao (Chang'an University), Fu, Ting (Tongji university), Qiangqiang, Shangguan (Tongji university), Xue, Jiangtian (Tongji university), Li, Yuhang (Tongji university)

Denoising and Reconstruction of Vehicle Trajectories Using Millimeter Wave Radar Groups

Scheduled for presentation during the Video Session "On-Demand Video Presentations" (VP-VP), Saturday, November 22, 2025, 08:00−18:00, On-Demand Platform

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on April 2, 2026

Abstract

Reconstructing trajectory data becomes significantly more challenging with the involvement of roadside millimeter-wave radars (MMRs), as the combination of noise, outliers, missing data, and inconsistencies in multi-dimensional observations (e.g., speed and position) complicates traditional methods. This paper proposes an optimization-based reconstruction method tailored for wide-area trajectories from MMR groups, handling complex anomalies while maintaining internal consistency across dimensions. The method incorporates a prior-guided mechanism comprising two key components: a masking strategy to address data missing caused by MMR blind zones, and a scaling factor to ensure adaptability and transferability across diverse operational scenarios. Extensive field experiments confirm the method’s robustness and transferability. Ablation studies further highlight the critical roles of the masking strategy and scaling factor in managing anomalies and ensuring generalization. This work paves the way for further research in large-scale trajectory reconstruction and multi-sensor integration. The code for our method can be found at https://github.com/tuqing123/MMRsTraRec.