Ensemble Kalman Filter Variants for Multi-Object Tracking with False and Missing Measurements
In this chapter, we present an approach to Multi-Object Tracking (MOT) that is based on the Ensemble Kalman Filter (EnKF). The EnKF is a standard algorithm for data assimilation in high-dimensional state spaces that is mainly used in geosciences, but has so far only attracted little attention for object tracking problems. In our approach, the Optimal Subpattern Assignment (OSPA) distance is used for coping with unlabeled noisy measurements and a robust covariance estimation is done using FastMCD to deal with possible outliers due to false detections. A simple gating technique allows handling of missing detections. Additionally, a recently proposed JPDA variant of the EnKF is discussed. The filters are evaluated in two different scenarios with false detections, where a nearest neighbour Kalman Filter (NN-KF) serves as a baseline.
KeywordsEnsemble Kalman Filter OSPA-metric False detections Missing detections Multi-Object Tracking
This work was supported by the Simulation Science Center Clausthal-Göttingen.
- 1.Bar-Shalom, Y., Willett, P.K., Tian, X.: Tracking and Data Fusion: A Handbook of Algorithms. YBS Publishing, Storrs (2011)Google Scholar
- 7.Crouse, D.F.: Advances in displaying uncertain estimates of multiple targets. In: SPIE – Signal Processing, Sensor Fusion, and Target Recognition XXII, vol. 8745, pp. 874,504–874,504–31 (2013). https://doi.org/10.1117/12.2015147
- 9.Guerriero, M., Svensson, L., Svensson, D., Willett, P.: Shooting two birds with two bullets: how to find minimum mean OSPA estimates. In: Proceedings of the 13th International Conference on Information Fusion (Fusion 2010) (2010)Google Scholar
- 10.Hanebeck, U.D., Baum, M.: Association-free direct filtering of multi-target random finite sets with set distance measures. In: Proceedings of the 18th International Conference on Information Fusion (Fusion 2015), Washington, USA (2015)Google Scholar
- 13.Romeo, K., Crouse, D.F., Bar-Shalom, Y., Willett, P.: The JPDAF in practical systems: approximations. In: Proceedings of SPIE 7698, Signal and Data Processing of Small Targets 2010, vol. 7698, pp. 76,981I–76,981I–10 (2010). https://doi.org/10.1117/12.862932
- 17.Sigges, F., Baum, M.: A nearest neighbour Ensemble Kalman Filter for multi-object tracking. In: 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI), Daegu, South Korea, pp. 227–232 (2017). https://doi.org/10.1109/MFI.2017.8170433