Paper Number: 5
Date: Tuesday, 11 July 2006
Time:
Duration: 20 minutes
Session: Nonlinear Filtering I
Location: Bargello
Combined Unscented Kalman and Particle Filtering for Tracking Closely Spaced Objects
Robert Pawlak
Abstract: Tracking closely spaced objects with resolution limited sensors is a difficult problem. One way to address this issue is to track these targets individually,
and employ relatively complex data association approaches as a means of pairing detections and tracks. The algorithm outlined in this paper takes a different approach, and instead estimates the group velocity using an unscented Kalman Filter (UKF). The UKF state estimate is then employed within a particle filter, which
estimates the distribution of objects within the group. It is shown that this approach can be very effective, especially for groups of irregularly spaced objects.
Presenter Biography: Robert Pawlak received his Ph.D. in 1992, and has been working in the areas of radar and data fusion since 1989. He has extensive background in the areas of data fusion, target tracking, pattern recognition, statistics and statistical signal processing. He is the author of numerous trade studies concerning the effectiveness of array and signal processing enhancements on radar suite integration. More recently, he has been focused on the areas of sensor and weapons systems for small boat detection, tracking and engagement. His papers have been published in the IEE Proceedings on Radar, Sonar and Navigation, the Proceedings of the American Society of Naval Engineers, Proceedings of the SPIE (the international Society for Optical Engineering), and he has numerous symposia papers to his credit.