标题：Central Difference Information Filter with Interacting Multiple Model for Robust Maneuvering Object Tracking
作者：Liu, Guoliang; Tian, Guohui
作者机构：[Liu, Guoliang; Tian, Guohui] Shandong Univ, Sch Control Sci & Engn, Jinan, Shandong, Peoples R China.
会议名称：IEEE International Conference on Information and Automation 2015
会议日期：AUG 08-10, 2015
来源：2015 IEEE INTERNATIONAL CONFERENCE ON INFORMATION AND AUTOMATION
关键词：Maneuvering object tracking; central difference information filter;; interacting multiple model; nonlinear system; information filter
摘要：In this paper, we introduce a new framework to combine the central difference information filter (CDIF) with the interacting multiple model (IMM) method for maneuvering object tracking. The CDIF has been recently introduced for solving object tracking problem using multiple sensors. The CDIF uses Stirling's interpolation to generate a number of sigma points for approximating the distribution of Gaussian random variables and does not require the calculation of Jacobians. However, the general CDIF method has difficulties to handle maneuvering objects, due to the changing of system model. In the literature, the IMM method is a natural way to estimate the discontinuities of object motion, by running a bank of filters in parallel with multiple models. Here, our contribution is to use the CDIF in the IMM framework (IMM-CDIF), which has better capabilities to handle maneuvering object tracking problem. In the end, a bearing only tracking experiment is demonstrated, and shows that the new IMM-CDIF method has lower mean square error (MSE) comparing with the original CDIF method.