Validation of Multiple Visitor Tracking with a Laser Rangefinder Using SMC Implementation of PHD Filter
Yuu Ishihara, Takeshi Uchitane, Nobuhiro Ito, Kazunori Iwata · 2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS) · 2022
In this study, the goal is to verify the accuracy of tracking multiple visitors using a laser rangefinder with a Sequential Monte Carlo (SMC) implementation of a Probability Hypothesis Density (PHD) filter. The feature of our method is using only a laser rangefinder. It is difficult to track multiple visitors using only a laser rangefinder for visitor measurement. However, this method is better to the method using cameras in viewpoint of privacy protection. The use of a PHD filter is also discussed to consider the variation of the number of visitors. In the results of the numerical experiment, the estimated number of visitors was compared with the actual number of visitors. From the comparison results, the PHD method suppressed the effect of occlusion compared with detection only. seems However, the errors at each time step slightly increased. The results of the experiment will be used to examine future challenges to improve accuracy.