Selected Topics: Multi-Object Filtering Theory
Reader, School of Engineering and Physical Sciences
Heriot-Watt University, United Kingdom
ASEN 5519, 3 semester hours, Section 200, Class No. 18004
Session B: July 8–August 8, 2014
In the last decade, the field of sensor fusion has witnessed a paradigm shift in the way that methods for detecting, estimating and tracking multiple targets in signal processing are developed. Heuristic approaches for estimating multiple dynamical objects have been developed since the 1970s, yet these methods suffer from systematic failure due to the heuristics introduced for track management. A radically different approach that considers the problem in a unified way based on point process theory that enables operators to estimate the number of targets and state vectors in challenging environments where there may be many false alarms and the targets are not always observed. This led to principled low computational cost approximate solutions that could be deployed on real-time systems known as Probability Hypothesis Density (PHD) filters.
There is an increasing interest in developing accurate estimation techniques for problems in multi-sensor multi-target data fusion, yet there is often a knowledge gap between the background of signal processing engineers and the mathematical skills required to derive algorithms with multi-object estimation techniques.
The objectives of this course are to fill a gap in knowledge for signal processing engineering research students and faculty and to facilitate the process of taking new theoretical developments into practical engineering applications. The course will cover concepts in probability theory, stochastic filtering, variational calculus, point process theory, multi-object estimation, and practical implementations with sequential Monte Carlo and Gaussian mixture techniques. To ensure that participants are equipped to derive new algorithms for their applications, the course will be delivered through traditional lectures and mathematics tutorials on set exercises.
Professor Clark’s research interests are in the development of the theory and applications of multi-object estimation algorithms for sensor fusion problems. He is widely published in the field and has collaborated closely with government and industry internationally on projects spanning theoretical algorithm development to practical deployment.
Special Topics: Geothermal Energy: Prospecting, Production, and Utilization
Associate Professor, Petroleum Engineering
University of Miskolc, Hungary
CVEN 4838, 3 semester hours, Section 200, Class No. 18218
CVEN 5838, 3 semester hours, Section 200, Class No. 18219
Session B: July 8–August 8, 2014
This is an introductory course covering the natural conditions, production, utilization, and environmental impact of geothermal energy. The course will provide students with a broad understanding of these topics and their history. Information in the class can be used when prospecting for geothermal sites, applying the appropriate geothermal production technology, and development of geothermal surface facilities.
Aniko Toth is currently leading a European Union project focused on the development of a graduate-level track in the field of geothermal energy. She has extensive experience in geothermal heat recovery, most of which is used in direct-use applications in Hungary. She is active in international research in the field of geothermal energy.