Multi Sensor Multi Target Data Fusion Tracking And Identification Techniques For Guidance And Control Applications PDF Download

The aim of this AGARDograph is to provide a quick overview of practical advances in multi-sensor, multi-target tracking technology and applications. This will provide the general summary of the MS/MTT techniques and technology with emphasis towards practical implementation. Many examples of sensor fusion involve the methodology of merging various track files taken from different sensors. This allows for more consistent, accurate, and reliable tracks than might be possible with any of the individual systems acting alone. Section 1 relates to the important use of sensor fusion prior to establishing a firm track file. By combining raw sensor information, greater discrimination of targets from background may be possible from the augmented body of available information. Tracking and fusion with multiple sensors deals with integration and correlation of data from diverse sources in order to arrive at the best possible situational assessment. In Section II, we present the tutorial on representative data association and filtering techniques, and also address some of the key initiation issues, approaches and track management methodology that simplify and enhance the practical implementation. Section III presents different types of classification algorithms, Bayesian Belief Networks, and Neural Networks covering the complete Automatic Target Recognition process, including fusion, segmentation and classification, that are very promising for real time, or quasi real time systems applications. Section IV covers the handling of Automatic Target Recognition (ATR) test data, deals with an effective tool to support the development of precision guided munitions, and presents a study of target acquisition and sensor cueing in air to air environment, The last section presents several practical examples of MS/MTT applications.