Information fusion in data association applications

There is a limitation to process data fusion by means of traditional deterministic or probabilistic data association algorithm of multisensor data fusion (MSDF). Those methods for data association do not adequately account for quantitative and qualitative information in an automated fashion. Fuzzy logic offers an enabling technology for automated quantitative and qualitative information in the data fusion process. We propose the fusion system architecture, called fuzzy gating approach, coordinating both quantitative and qualitative information which is realized using a fuzzy-based reasoning approach. This approach is composed of two stages in cascade. The first stage implements available quantitative information, namely target range, azimuth, and elevation angle, to form a subset of statistically likely target solutions via fuzzy validation. The second stage of the fuzzy similarity utilizes available qualitative information, namely infrared image area and brightness, to form another subset of returns. Finally the fuzzy gating approach was tested in a dense clutter environment. Results of test show that performance of fuzzy gating approach is superior to JPDA based on a Bayesian approach. Moreover, adding qualitative information to the tracking algorithm can improve tracker performance effectively.