TRACKING OF MOVING TARGET USING KAL.MAN FILTER A SIMULATION STUDY

Abstract

The problem of tracking a moving target accurately has many military and civilian applications. Targets which follows straight line constant velocity or constant acceleration generate simple tracking problems. These problems are solved by means of a Standard Kalman Filter. The results of the simulation shows that the Kalman ' filter tracks the target. states well for a non maneuvering, target. Modeling a system accurately is one of the most important_ problems when using the Kalman Filter for target tracking. The system model of a target moving with constant velocity in 'a straight line is different from that of a target moving with maneuver. Development of an accurate system model requries maneuver. detection and estimation of . the - magnitude of maneuver. A time delay in the detection of maneuver resutls„ in erroneous target states.. A quck detection scheme is simulated to reduce the errors. After maneuver detection, the system model is modified by adding a maneuver term. A recursive algorithm is presented to estimate the maneuver magnitude, which - is then fed _ to .-a -modified Kalman filter_ for target tracking. Simulation results demonstrate the satisfactory performance of this tracking algorithm for maneuvering targets.