Tracking of a highly maneuvering target using a radar multisensor and an angle sensor

Tracking of highly maneuvering targets in a cluttered environment is a difficult problem. It has been previously reported that the target motion model defined as superposition of multiple constant acceleration vectors on a constant velocity straight line motion model using an automatic optimization of the target correlation gate in response to the target maneuvering based on the Multiple Maneuver Model Probabilistic Data Association (M3PDA) proved to be an effective method for severely cluttered conditions. On the other hand, due to the recent developments in the area of sensor technologies it became possible to use highly accurate angle sensors (such as infrared sensors) for target tracking. It is expected that the use of such sensors will greatly improve the tracking performance of radar systems. In this study, we propose an algorithm based on an expanded M3PDA formed by the six-dimensional state vector consisting of the target position and velocity components using three-dimensional observation data from the radar and two-dimensional observation data from angle sensors. We also demonstrate the advantages of the proposed method compared to conventional ones using a simulation of tracking of one maneuvering target. © 2000 Scripta Technica, Electron Comm Jpn Pt 1, 84(2): 99–109, 2001