一种基于模型概率单调性变化的自适应IMM-UKF改进算法  

Improved Adaptive IMM-UKF Algorithm Based on Monotonous Transformation of Model Probability

作　　者：王平波[1] 陈强 卫红凯[1] 贾耀君 沙浩然 WANG Pingbo;CHEN Qiang;WEI Hongkai;JIA Yaojun;SHA Haoran(College of Electronic Engineering,Naval University of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学电子工程学院,武汉430033

出　　处：《电子与信息学报》2024年第1期41-48,共8页Journal of Electronics & Information Technology

摘　　要：针对现有交互式多模型(IMM)算法模型间切换迟滞和转换速率慢的缺点,提出一种基于模型概率单调性变化的自适应交互式多模型无迹卡尔曼滤波改进算法(mIMM-UKF)。该算法利用后验信息模型概率的单调性,对马尔可夫转移概率矩阵及模型估计概率进行二次修正,加快了匹配模型的切换速度及转换速率。仿真结果表明,与现有算法相比,该算法通过快速切换匹配模型,有效提高了水下目标跟踪精度。Considering the hysteresis of model switching and the slow conversion rate of existing adaptive interacting multiple models,an improved algorithm of adaptive interacting multiple models with an unscented Kalman filter based on monotone transformation of model probability(mIMM-UKF)is proposed.In this algorithm,the monotonicity of the model probability in the posterior information is used,and this algorithm makes a secondary modification to the Markov probability transition matrix and model estimation probability is introduced.Consequently,an accelerated switching speed and conversion rate of the matching model are obtained.The simulation results show that compared to existing algorithms,this algorithm significantly improves the accuracy of target tracking by enabling swift switching of matching models.

关 键 词：水下目标跟踪 IMM-UKF算法 自适应 转移概率矩阵 单调性 

分 类 号：TN929.3[电子电信—通信与信息系统] TN953[电子电信—信息与通信工程]