首页|高机动目标的改进强跟踪CKF自适应IMM算法

高机动目标的改进强跟踪CKF自适应IMM算法

Improved strong tracking CKF adaptive IMM algorithm for high maneuvering targets

扫码查看
为提升高机动目标跟踪精度,提出了一种改进的强跟踪 CKF 自适应交互多模型跟踪算法.在IMM 算法运动模型集中引入 CS-Jerk 模型,增强对高机动目标的适应能力,采用奇异值分解(SVD)算法解决模型集中因模型扩维而导致CKF算法无法Cholesky分解的问题;提出了一种改进的强跟踪CKF算法,降低强跟踪CKF算法的计算量;利用模型的后验信息对IMM算法模型转移概率进行自适应调整,提高跟踪精度.仿真结果表明,基于所提算法目标的位置均方根误差均值和速度均方根误差均值较IMM-CKF算法分别降低了 22.50%和 16.58%,有效提高了目标跟踪精度.
Aiming to improve the tracking performance of high maneuvering targets,an improved strong tracking CKF adaptive interactive multi model tracking algorithm is proposed.The CS-Jerk model is introduced in the model set of IMM to enhance its adaptability to high maneuvering targets,the singular value decomposition(SVD)algorithm is adopted to solve the problem of CKF algorithm being unable to Cholesky decompose due to model dimension expansion.An improved strong tracking CKF algorithm is proposed to reduce the computational complexity of strong tracking CKF algorithm,the model transition probability of IMM is adjusted adaptively by using the posterior information of the model to improve the tracking accuracy.The simulation results show that the mean RMSE of position and the mean RMSE of velocity based on the proposed algorithm are reduced by 22.50%and 16.58%respectively compared to the IMM-CKF algorithm,effectively improving the accuracy of target tracking.

high maneuvering targettarget trackingadaptive interacting multi-modelstrong tracking CKFsingular value decomposition

成怡、刘铭阳、徐国伟

展开 >

天津工业大学,控制科学与工程学院,天津 300387

天津工业大学,机械工程学院,天津 300387

高机动目标 目标跟踪 自适应交互多模型 强跟踪CKF SVD分解

国家自然科学基金

61973234

2024

中国惯性技术学报
中国惯性技术学会

中国惯性技术学报

CSTPCD北大核心
影响因子:0.792
ISSN:1005-6734
年,卷(期):2024.32(7)