Date:        March 12, 2010
Time:       12:00 PM
Place:       CCRF Building 1 Room 104 directions
Food:        11:30 AM

In theory, a good particle filter allows to approximate an exact Bayesian filter solution arbitrarily well. This has motivated a strong and successful development of particle filtering approaches towards target tracking. In this presentation we pose the question whether this theory also applies to problems of tracking two targets that maneuver in and out a formation flight, and where the observations may include false measurements, missed detections and limited sensor resolution.

The specific aspects addressed in the presentation are:

- Mathematical formulation of the problem [1]-[2]

- Exact Bayesian recursion of the conditional joint density [1]-[2]

- IMMJPDA and IMMJPDA* filtering results [1]-[2]

- Particle Filtering results [2]-[5]

- Unique decomposition of the conditional joint density [6]

Emphasis of the presentation is on the insight gained. Background information and references to relevant other work is available in the references below.

References:

1.      H.A.P. Blom, E.A. Bloem, Exact Bayesian filter and joint IMM coupled PDA tracking of maneuvering targets from possibly missing and false measurements, Automatica, Vol. 42 (2006), pp. 127-135.

2.      H.A.P. Blom, E.A. Bloem, Joint particle filtering of multiple maneuvering targets from possibly unassociated measurements, Journal of Advances in Information Fusion, Vol. 1, Number 1, July 2006, pp. 15-34. Available at http://www.isif.org/

3.      H.A.P. Blom, E.A. Bloem, Bayesian tracking of two possibly unresolved targets, IEEE Tr. Aerospace and Electronic Systems, Vol. 43 (2007), pp. 612-627.

4.      H.A.P. Blom, E.A. Bloem, Hybrid SIR joint particle filter under limited sensor resolution, Proc. Int. Conf. on Information Fusion 2007, Quebec, Canada, 9-12^th July 2007.

Despite advances in formal and probabilistic verification approaches, fault and event trees are still the dominant techniques used for safety risk analysis in aviation. However, the combination of concurrent, dynamic, and random effects that appear in air traffic cannot properly be captured by these classical techniques. In this lecture, it will be explained how safety risk modeling and analysis can be formulated as a problem of estimating the rare event probability of a large scale stochastic hybrid system. Subsequently it is explained how rare event estimation theory for diffusions can be extended to a large scale stochastic hybrid system (SHS), in which a large number of rare discrete modes may contribute significantly to the rare event estimation.  Essentially, the approach taken is to develop a compositional model of the air traffic operation considered in the form of a large scale SHS; then to introduce a suitable aggregation of the discrete modes of this large scale SHS; and then to develop importance sampling and Rao-Blackwellization relative to these aggregations. The practical use of this approach will be demonstrated for the estimation of the mid-air collision probability for an advanced air traffic application.  

Brief Biography

Henk Blom is a Principal Scientist at National Aerospace Laboratory NLR in Amsterdam, The Netherlands. He received his BSc and MSc degrees from Twente University in 1975 and 1978 respectively. Subsequently he performed research in forward-looking infrared image processing at TNO Physics Laboratory, The Hague. In 1980 he joined NLR to work on research in Air Traffic Management. In 1988 he was a visiting scholar at the University of Connecticut, Storrs.  In 1990 he received a PhD from Delft University of Technology with a thesis entitled “Bayesian estimation for decision-directed stochastic control.”

Dr. Blom has over twenty five years of experience in the theory of stochastic modeling and analysis and its application to signal processing, data fusion and safety risk analysis. Since joining NLR, his leading research motivation has been to develop stochastic hybrid systems theory applicable to Air Traffic Management. He is the scientific leader of innovative developments such as the Interacting Multiple Model (IMM) filter algorithm, Eurocontrol’s Bayesian multi-sensor multi-target tracking system ARTAS (ATM Radar Tracking And Server) and NLR’s safety risk analysis methodology TOPAZ (Traffic Organization and Perturbation AnalyZer) and supporting tool sets. He is the author of over one hundred articles in scientific journals, books and conference proceedings, and of the volume “Stochastic Hybrid Systems, Theory and Safety Critical Systems”, Springer, 2006. He has been the organizer and coordinator of several large European research projects, including the current iFly project with active participation by 10 universities and 8 other project partners (see http://iFly.nlr.nl).  In 2004 he received NLR’s Dr.Ir. B.M. Spee Award. Dr. Blom has been a Fellow of the IEEE since 2007.

Probabilistic Safety Verification of Future Air Traffic Management

Henk Blom bio