一种基于自适应去偏转换滤波的机动目标跟踪

针对于机动目标的跟踪问题,提出了一种基于交互式多模型的自适应去偏转换卡尔曼滤波器.该算法利用交互多模型算法来完成不同跟踪模型的相互切换;根据自适应去偏转换测量卡尔曼滤波算法来推导跟踪目标状态,同时自适应因子可以确保不正常测量时的鲁棒性.与传统的去偏转换卡尔曼滤波算法对比,该算法可以很好地改善所获量测信息在雷达被干扰时的目标跟踪精度.仿真结果表明了算法的有效性和可行性,且跟踪精度相对传统的去偏转换卡尔曼滤波算法减少9.38％的位置误差.

Maneuvering target tracking with adaptive debiased converted measurement filter

An adaptive debiased converted measurement Kalman filter (ADCMKF) based on interacting multiple model (IMM-ADCMKF) is proposed to solve the tracking problem of maneuvering target,which applies interacting multiple model (IMM) to carry out the mutual switching of different tracking models.The states of the tracked target are derived based on the ADCMKF,and an adaptive factor is designed to ensure the robustness to abnormal measurement.By embedding the ADCMKF into the IMM framework,the IMM-ADCMKF is capable of tracking a maneuverable target whether the radar works normally or not.Compared with the conventional DCMKF,the proposed adaptive algorithm solves the abnormal measurement when the radar is interfered.Simulation results demonstrate the validness and feasibility of the proposed algorithm.The position estimation error of IMM-ADCMKF is reduced by 9.38％ compared with that of IMM-DCMKF.