基于线性波门改进拍卖算法的航迹关联算法研究

Research on Track Correlation Algorithm Based on Auction Algorithm Improved by Linear Gate

针对分布式探测系统中多源传感器航迹关联问题,常用的基于距离统计构建关联代价矩阵,拍卖算法求解关联结果的流程在面临探测区域中目标数量大,目标航迹分布密集的场景时,会出现关联准确率下降及关联时间长的问题。提出了一种基于线性波门改进的拍卖算法来改进现有关联算法在面临上述场景时的关联性能。考虑传感器的测量偏差对距传感器远的目标航迹数据影响大于距离近的目标,因此设计了与距离有关的线性波门对可能关联的目标航迹筛选。仿真数据结果表明,提出的线性波门能够对传统拍卖算法在大批次目标下提升关联准确率和降低关联时间,证明了线性波门的有效性,对实际多源传感器航迹关联问题具有重要意义。

The problem of multi-source sensor track correlation in distributed detection systems is usually solved in such a way that the commonly used process of constructing a correlation cost matrix based on distance statistics and auction algorithm will occur when facing a scenario where the number of targets in the detection area is large and the target tracks are densely distributed.The problem of decreased correlation accuracy and long correlation time are thus quite common.An improved auction algorithm based on linear wave gate is proposed to improve the correlation performance of existing correlation algorithms in the above scenarios.Considering that the measurement deviation of the sensor has a greater impact on the track data of targets far away from the sensor than targets close to the sensor,a linear wave gate related to distance is designed to filter the target tracks that may be associated.Simulation data results show that the proposed linear wave gate can improve the correlation accuracy and reduce the correlation time of the traditional auction algorithm under large batches of targets,proving the effectiveness of the linear wave gate and having import-ant significance for actual multi-source sensor track correlation problems.