高超声速机动目标的强跟踪UKF自适应交互多模型算法  

Strong tracking UKF adaptive interacting multiple-model algorithm based on maneuvering hypersonic-target tracking

作　　者：戴洪德 方君 唐亮 王希彬 DAI HongdeI;FANG Jun;TANG Liang;WANG Xibin(School of Basic Sciences for Aviation,Naval Aviation University,Yantai 264001,China;Aviation Military Representatives Office of Navy in Hanzhong,Hanzhong 723213,China;Unit 91213,Yantai 264007,China)

机构地区：[1]海军航空大学,航空基础学院,烟台264001 [2]海军驻汉中地区航空军事代表室,汉中723213 [3]91213部队,烟台264007

出　　处：《中国惯性技术学报》2018年第3期338-345,共8页Journal of Chinese Inertial Technology

基　　金：山东自然科学基金(ZR2017MF036); 国防科技项目基金(F062102009); 航空基金(20160884004)

摘　　要：针对高超声速强机动目标的运动具有复杂性、突变性和强非线性等特点,单模型跟踪算法难以实现精确跟踪的问题,提出了一种基于多重渐消因子的强跟踪UKF自适应交互多模型算法。考虑现有强跟踪UKF算法引入单渐消因子的不足,根据正交性原理推导得到了引入多渐消因子的强跟踪UKF算法,完成了对非线性目标状态的滤波估计;在交互多模型算法的子模型中选用改进的CS-Jerk模型;对交互多模型中各子模型间的转移概率进行在线自适应调整,并与改进CS-Jerk模型结合克服了单模型算法跟踪强机动目标的不足,实现了模型与目标运动模式的实时最优匹配。仿真结果表明,与单模型算法和经典多模型算法相比,提出的算法使得不同条件下位置和速度的跟踪误差至少降低11.89%,有效提高了高超声速强机动目标跟踪精度。Due to the complexity, mutability and strong non-linearity of the movement of hypersonic strong maneuvering targets, the single-model tracking algorithm is difficult to achieve accurate tracking. To solve this problem, an adaptive interactive multi-model （IMM） algorithm based on strong tracking UKF is proposed. Considering the shortage of the single fading factor, the algorithm introduces a multiple fading factor into the strong tracking UKF algorithm based on orthogonality principle to complete the estimation of nonlinear target states. An improved CS-Jerk model is used in the sub-model of the IMM algorithm, and the model transition probability is updated adaptively in the IMM algorithm and combined with modified CS-Jerk model to overcome the shortage of single model algorithm for high-maneuvering target tracking and realize the optimal matching between models and target motion in real time. Simulation results indicate that, compared with the existing single model algorithms and normal IMM algorithm, the proposed modified IMM algorithm significantly improves the tracking accuracy of the maneuvering hypersonic target and causes the tracking errors of position and velocity be reduced at least 11.89% under various conditions.

关 键 词：高机动目标 目标跟踪 强跟踪UKF CS-Jerk模型 自适应交互多模型 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]