A novel interacting multiple model method for nonlinear target tracking

The state estimation of targets is a difficult task, particularly if the target exhibits nonlinear behaviour, which is often the case. Currently, the most popular filters used in target tracking are the Kalman filter (KF) and its various forms, as well as the particle filter (PF). Introduced in 2007, the smooth variable structure filter (SVSF) is a relatively new predictor-corrector method based on sliding mode estimation. In the past, this filter has been used successfully for the state and parameter estimation of mechanical and electrical systems for the purpose of control. This paper introduces a new interacting multiple model (IMM) method that makes use of the SVSF estimation strategy. An air traffic control (ATC) problem is used to compare the common EKF-IMM with the proposed SVSF-IMM in terms of tracking accuracy, robustness, and computational complexity. Furthermore, this paper demonstrates that the SVSF is an effective method for nonlinear target tracking.