PhD studentship in Multi-Object TrackingBackground| Project Description
Joint Research Institute of Signal and Image Processing, Edinburgh Research Partnership.
Send Expressions of Interest to: Dr. Daniel Clark
There is substantial interest in the development of autonomous systems to enable remote surveillance of environments where it is too dangerous or too costly for humans to go. For example, scientific investigation of Mars has been made possible by the development of NASAs Mars Exploration Rover and the EUs Beagle missions. The DARPA Grand Challenge was created to accelerate research and development in autonomous ground vehicles. Unmanned Aerial Vehicles (UAVs) are remotely operated or self-piloted aircraft that can carry cameras, sensors, communications equipment or other payloads and have been used in reconnaissance since the 1950s. Advances in sensing technologies enable these to conduct more challenging roles. Therefore, the principal aim of this project is the development of a framework for the detection, identification, and tracking of targets to enable the autonomous surveillance of complex multi-target multi-sensor environments.
Target tracking is a necessary part of systems that perform functions such as crime prevention, defence and anti-terrorism. Tracking algorithms take their input measurements from sensors which provide the signals such as radar, sonar or video. The measurements are taken at regular intervals and the task is to estimate the state of a target at each point in time, such as its position, velocity or other attribute. Successive estimates provide the tracks which describe the trajectory of a target.
Multi-object filtering is a mathematical theory in which the number of objects and their individual states must be jointly estimated from a sequence of multi-object observations without knowledge about which object generated which observation. This problem, also known as multi-target tracking, can be traced back to the early 1970s and was driven primarily by aerospace applications such as radar, sonar, guidance, navigation, and air traffic control. Finite set statistics (FISST) provides a general systematic treatment of multi-object filtering using random finite sets (RFSs). In the FISST approach, multi-object states and multi-object observations are modelled as RFSs. This allows probabilistic descriptions of the multi-target dynamics and observation process, which resulted in novel multi-object filters such as the multi-object Bayes filter and the Probability Hypothesis Density (PHD) filter.
Advances in sensing technologies have led to an enormous growth in applications which are based on the estimation and control of multi-object systems. With the recent advances in autonomous systems, such an Autonomous Underwater Vehicles (AUVs), Unmanned Aerial Vehicles (UAVs) and Unmanned Ground Vehicles (UGVs), the need for a mathematically coherent framework for data fusion from different vehicles is necessary if the best understanding of the environment is to be achieved. The fundamental aim of this project is the development of a framework for the detection, identification, and tracking of targets to enable the surveillance of environments from multiple sensor platforms. The research objectives of this project are:
1. To develop novel multi-sensor multi-target data fusion methodology.
2. To develop algorithms for modelling and tracking co-ordinated multi-target motion.
3. To develop algorithms for tracking multiple extended objects, where objects produce multiple observations.
4. To formulate techniques for multi-target tracking with unthresholded sensor data.
A better understanding of multi-object systems will lead to principled engineering approximations that result in efficient and reliable algorithms to support rapidly expanding multi-sensor multi-object systems based applications.
Demonstrations of multiple target tracking algorithms are given in the PHD filter section.
Project led by: Dr. Daniel Clark (Heriot-Watt University)
Candidates should have a strong mathematical skills, with a background in mathematics, statistics, electrical engineering or computer science. Experience in software development would be beneficial. Preference will be given to EU students.
© Heriot-Watt University and the University of Melbourne 2008