Research Seminars

Multi-Object Filtering Techniques for Multi-Target Tracking

Daniel Clark
Signal Processing Lab, University of Cambridge

Tuesday 7 August 2007, 2:00pm, Room 105

Abstract

The applications of multi-target detection and tracking are of crucial importance in defence applications and robotics. Recent advances in multiple-target filtering provide a general systematic treatment of multi-object filtering using random sets. These have led to robust multiple-target tracking algorithms for estimating both the correct number of targets and their positions in complex environments with large numbers of false detections and where the true targets are not always detected. These can be exploited on a range of sensor systems without complex target detection algorithms. Some practical applications of these techniques have been developed for tracking in sonar for autonomous vehicle navigation, concealed weapons detection in millimetre wave imaging, people tracking in video, and audio signal processing for music transcription. This talk will present recent developments in this field and describe some of the current research challenges being investigated.

Speaker’s bio:

Daniel Clark was awarded his PhD entitled "Multiple Target Tracking with the Probability Hypothesis Density Filter" at Heriot-Watt University in Scotland in 2006. During his PhD he made a number of key contributions to the field of multiple-object filtering and tracking. He established convergence properties of the first numerically tractable multi-object filters, known as PHD filters, providing a theoretical justification for application of these techniques for engineering applications. He developed techniques to enable these filters to be used for practical multi-target tracking for the first time and has demonstrated these techniques on real sonar data from an underwater vehicle, millimetre wave imaging and on musical audio data. He has implemented these techniques for oil pipeline tracking for deployment on an autonomous underwater vehicle which achieved a world record in autonomous navigation. The algorithm successfully tracked 22km of pipeline continuously over 5-6 hours, which was more than double what had previously been achieved. He is currently working as a Research Associate in the Signal Processing Lab at the University of Cambridge before taking up a lectureship position at Heriot-Watt.