| 摘要: |
| 机载雷达在运动平台上探测目标，探测数据描述的是平台和目标的相对位置。目标的机动状态不明，且探测数据和目标真实位置之间是非线性关系，导致机载雷达的航迹滤波难度大，滤波器精度和稳定性差。本方法中先进行数据预处理，量测误差在天线坐标系下获得，目标状态方程和量测方程在固定坐标中心的直角坐标系下建立，量测误差协方差矩阵由量测误差和位置进行无偏转换获得；采用交互多模型滤波器，模型集由匀速直线运动模型、当前统计模型和角速度未知的匀速转弯运动模型组成；非线性模型迭代由容积卡尔曼滤波实现，各模型的初始概率和模型参数基于速度和径向速度进行预估。仿真实验表明该方法提高了机载PD雷达的航迹情报质量，尤其提高了机动目标跟踪的精度和稳定性。 |
| 关键词: 机载PD雷达 交互多模型 数据预处理 机动目标 |
| DOI：10.3969/j.issn.1672-2337.2021.01.016 |
| 分类号:TN953；TP391.9 |
| 基金项目:国防重大专项（No.2017ZX01013201-006) |
|
| An Interacting Multiple Model Algorithm for Airborne PD Radar |
|
MA Juan, FAN Yuzhu, LI Le
|
|
1.The 38th Research Institute of China Electronics Technology Group Corporation, Hefei 230088, China;2.Unit 63629 of PLA, Beijing 100162, China
|
| Abstract: |
| The airborne radar obtains measurement on the moving platform, which describes the relative position of the antenna and the target. The unknown maneuvering state of the target and the nonlinear relationship between the measured data and the real position result in the difficulty of track filtering, which has poor accuracy and stability. In this paper a method is proposed, in which the data is preprocessed first, and the measurement error is calculated in antenna coordinate system. The target state equation and measurement equation are set up in a rectangular coordinate system with a fixed coordinate center, and the measurement error covariance matrix is obtained by unbiased conversion from the measurement error and measurement value. The model set of interactive multi-model filter is composed of CV model, CS model and CT model. The nonlinear model iteration is realized by cubature Kalman filter. The initial probability and model parameters are estimated based on velocity and radial velocity. Simulation experiments show that the proposed method can improve the track quality of airborne PD radar, especially the accuracy and stability of maneuvering target. |
| Key words: airborne PD radar interacting multiple model data preprocessing maneuvering target |
