A Novel Gaussian Particle PHD Filter for Multi-target Tracking

摘要：

The Gaussian particle probability hypothesis density filter(GPPHDF) needs conventional Monte-Carlo (MC) sampling in predict step and update step, which decreases the accuracy and real-time performance of the algorithm. This paper employs Quasi-Monte-Carlo (QMC) sampling to replace MC sampling, and QMC integration method is introduced to approximating the prediction and update distributions of target states. Hence a tracking algorithm based on the QMC method is proposed, which reduces the computational complexity and improves the accuracy and stability of the tracking algorithm.

关键词： multiple target tracking random sets Quasi-Monte-Carlo (QMC) Gaussian particle probability hypothesis density filter

作者: Zengjian Huang Guixi Liu Pengju Chang

作者单位: School of Electronic & Mechanical Engineering, Xidian University, Xian, 710071, China The Northwest Machine CO., LTD. Xian, Shanxi, 710119, China

会议类型: 国际会议

会议名称: 2011 3nd International Conference on Mechanical and Electronics Engineering(2011年第三届机械与电子工程国际会议 ICMEE2011)

会议地点: 合肥

会议语种:英文

页码: 3143-3147

在线出版日期: 2011-09-23（万方平台首次上网日期，不代表论文的发表时间）