A variable γ H∞ filter for a maneuvering target tracking using acceleration estimation

The Kalman filter and its applications have been employed for tracking systems. A non-maneuvering target can be accurately tracked by using a Kalman filter with a constant velocity model. However, a maneuvering target might be mis-tracked since prediction errors increase significantly. The H_∞ filter is a robust filter and changes its characteristics depending on the parameter γ. In order to track accurately not only non-maneuvering but also highly maneuvering targets, we propose a variable γ H_∞ filter with an acceleration term when a maneuver is detected