| 摘要: |
| 多传感器融合是提高态势感知能力的重要手段。为提高探测能力，将雷达和红外传感器组网，各传感器独立工作，在统一调度下，完成探测、跟踪、识别任务。研究该系统的雷达/红外数据融合算法，针对传感器异步探测特点，采用观测驱动的融合跟踪方法：雷达探测到目标时，采用基于状态预测的改进BLUE（Best Linear Unbiased Estimation）滤波，通过方位预测的辅助，减小测角误差非线性效应，提高跟踪性能；红外探测到目标时，基于方位预测和斜距观测，构造新的转换量测模型，实现基于不完备观测的修正BLUE滤波器。理论分析和仿真证明：所述雷达/红外数据融合方法，在不同传感器布局下都具备更优的综合性能，其设计思想可解决其他类似的多传感器融合问题，有较好的应用推广潜力。 |
| 关键词: 多传感器数据融合 非线性滤波 最佳线性无偏估计 卡尔曼滤波 |
| DOI：DOI:10.3969/j.issn.1672-2337.2023.05.015 |
| 分类号:TN953 |
| 基金项目:安徽省自然科学基金（No.1708085MF153） |
|
| Asynchronous Fusion Algorithm Research Based on BLUE for Netted System of Radar/IR Sensors |
|
SHENG Hu， WANG Haibing
|
|
1. The 38th Research Institute of China Electronics Technology Group Corporation, Hefei 230088, China;2. College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China
|
| Abstract: |
| Multi?sensor fusion plays an important role in situation awareness improvement. To improve detection ability, a radar/infrared netted system is proposed. All sensors work independently, and accomplish detection, tracking, and recognition tasks under unified schedule. For radar target tracking, an improved BLUE （Best Linear Unbiased Estimation） filter is presented, which enhances tracking ability by decreasing azimuth estimation error. For infrared target tracking, a modified BLUE filter is presented, which solves the bearing?only target tracking problem. Theoretic analysis and numeric results verify that the proposed approach exhibits improved tracking performance and computational advantage over others in different sensor geometries. The architecture design can be used to solve other multi?sensor fusion problems, so this scheme is worth further development and promotion. |
| Key words: multi⁃sensor data fusion nonlinear filtering best linear unbiased estimation Kalman filter |
