Fuzzy-Based Evidence Accrual for Target Maneuver Detection
Stephen Craig Stubberud, Kathleen Ann Kramer, Allen Roger Stubberud · 2018
One of the primary techniques for target tracking is the Kalman filter or one of its many variants. While the Kalman filter has been the staple of the tracking community, it has been shown to have drawbacks. When a target performs a maneuver, the tracking solution of the filter can experience deleterious issues. The most common issue is a lag in the position of the target track compared to the true target position as the target performs its maneuver. A more problematic issue occurs when the filter covariance collapses which requires the filter to be reinitialized. While techniques exist that compensate for maneuvers, they rely on detecting the error in the estimated trajectory and the measured target position to generate their response. In this effort, a maneuver detection routine is developed that can be used in conjunction with the standard maneuver compensation approaches. This routine validates the existence of a maneuver more quickly than using inherent detection methods of the other methods. The maneuver detection is performed by an evidence accrual system that uses a fuzzy Kalman filter to incorporate new information and provide a level of evidence that maneuver is occurring. The input data uses behavior characteristics of the Kalman gain vector from the tracking algorithm.