A Tracking Algorithm for Dense Target Environments

A tracking algorithm operating in a dense target environment is described. This algorithm is designed for a passive optical (angle only) sensor handling a very large number of targets. For the application considered, the sensor has limited angular resolution and the sampling rate is exceedingly low. To further complicate the algorithm design issues, the underlying system is nearly unobservable. The standard extended Kalman filter is not applicable in this case; an iterative maximum likelihood estimator must be used.