I’ve been working on
air, ground, and under water target tracking with different sensing and data
processing techniques. I am interested in efficient data association techniques
to solve the measurement-to-track or track-to-target correspondence problem
under difficult scenes, .e.g., finite radar resolution (which often results in
merged/unresolved measurement), crossing, spawning or closely spaced targets. I
would also like to see how target recognition and tracking can be handled
jointly with various kinds of uncertainties, i.e., target motion, target type,
measurement origin. I have recently worked on the target maneuver detection
problem by appropriately using the range rate (Doppler) information from the
radar. Another interesting project is to incorporate the terrain information in
target acquisition through coordinated UAV path planning under variable
environment.
I am taking a
leading role to build a testbed consisting of
different surveillance cameras and 5-8 mobile vehicles with sensors, GPS
receivers and embedded computing and communication capability for the use of
ground target surveillance including tracking people and moving vehicles. The
system will allow testing new approaches for command and control across the
dynamic varying sensor network. The major research focus will be on the
development of integrated sensing and data fusion techniques to enhance the
tracking and discrimination capability in a dense target environment. The testbed will also allow research and education in
developing distributed command and control algorithms, dynamic reconfigurable
network topologies, data compression in a multi-network and lossy
environment, sensor data fusion for feature aided tracking which are relevant
to future DoD surveillance
systems. A unique feature of the testbed is the
integration of sensing and data processing in a dynamic network environment
with multiple moving vehicles performing cooperative tasks. The testbed will be used for several ongoing research programs
in real-time target inference, joint decision, estimation and model fusion, and
distributed control and computing. My research has been supported in part by
ARO,
-------------------------------------------------------------------------------------------------------------------------------------------
People working in
Target Tracking area having homepages (excluding ISL@UNO)
Yaakov Bar-Shalom at UConn, Eric A Wan at OGI K P Murphy at MIT, G Bishop at UNC, Lang Hong at
Wright
Rob Evans at