Multi-vehicle tracking (MVT) system is a prerequisite for intelligent vehicles. Due to the high angular resolution and high range accuracy, LiDAR has become the most commonly used sensor in MVT system. However, the inter-target occlusion (ITO) which is often observed in MVT system using LiDAR, causes estimated track break and vehicle target loss. To address this issue, this paper proposed a multi-vehicle corner tracking algorithm based on the labeled multi-Bernoulli (LMB) filter with detection probability optimization model (DPOM). The DPOM takes the inter-target occlusion probability of tracked vehicle corners into consideration and achieves better modeling of the likelihood of gaining vehicle-generated corner measurements during the occlusion stage. We designed MVT simulation and experiment with challenging occlusions to verify the effectiveness of our proposed method. Both simulation and experiment results show that our proposed method can greatly improve the ability of the MVT algorithm to deal with the ITO problem and outperform conventional algorithms.
Inter-target Occlusion Handling in Multi-vehicle Corner Tracking Based on Labeled Multi-Bernoulli Filter with Detection Probability Optimization Model
Lect. Notes Electrical Eng.
International Conference on Intelligent Transportation Engineering ; 2021 ; Beijing, China October 29, 2021 - October 31, 2021
2021 6th International Conference on Intelligent Transportation Engineering (ICITE 2021) ; Chapter : 16 ; 178-189
2022-06-01
12 pages
Article/Chapter (Book)
Electronic Resource
English
British Library Conference Proceedings | 2022
|Context-based multi-target tracking with occlusion handling
British Library Online Contents | 2016
|