Improved Divided Difference Filter based on Newton-Raphson Method for Target Tracking

In this paper, improved divided difference filter, which will be called IDDF for brevity, is proposed for tat-get tracking with nonlinear observation models. The new algorithm is derived from the Newton-Raphson method (or Newtons method) to approximate maximum a posterior (MAP) estimation. We demonstrate the direct and intuitive relationship between the iterated extended Kalman filter and Newton-Raphson method and can extend the divided difference filter so that iteration is possible. Simulation results show that the proposed filter provides better performance in tracking accuracy when compared to standard DDF, iterated extended Kalman filter (IEKF) and extended Kalman filter (EKF) in presence of severe nonlinearity.