A New Algorithm Based on Generalized Target Maneuver Detection

In the airborne vehicles, complex target tracking with more accurately is highly desirable. Conventional techniques based on augmented or batch algorithms are computationally expensive, which is the major drawback for the real time target parameter estimation. In contrast, other conventional input estimation techniques which overcome this problem, assume constant acceleration and therefore this is not a generalized modeling (T.C. Wang and P.K. Varshney, 1993). The proposed algorithm is developed to overcome these drawbacks by using a new generalized dynamic modeling of acceleration and acceleration rate (jerk) with reduction on the vector state dimension. The proposed modified Kalman filter algorithm is based on a generalized formulation including our earlier works. Results show the significant improvement for high maneuver target tracking problem.