基于k近邻平均距离的异步航迹关联算法

针对异步不等速率下局部节点航迹关联复杂问题, 提出了基于k近邻平均距离的异步航迹直接关联算法。首先, 给出不等长航迹序列间的k近邻平均距离计算规则, 进而计算得到不等长航迹序列间的灰色关联度, 再利用经典分配法进行航迹关联判定。算法无需时间同步, 避免估值误差传播积累。仿真数据表明, 算法正确关联率高、耗时较短、局部节点采样周期和开机时机不一致等异步因素对算法影响不明显, 并且算法不受噪声分布形式和目标数目变化的影响。

Asynchronous track-to-track association algorithm based on k means distance of nearest neighbors

Aiming at the complex problem of local node track-to-track association under asynchronous and unequal rates, an asynchronous track-to-track direct association algorithm based on the k means distance of the nearest neighbors is proposed. First, the k means distance calculation rule of the nearest neighbor between different track sequences is given. Then the gray relational degree among different track sequences are calculated, the classical allocation method is used to determine the track-to-track association. The algorithm does not need time domain registration, and it avoids the accumulation of estimation error propagation. The simulation data show that the algorithm has high correct correlation rate and low time consumption. Asynchronous factors such as inconsistency of sampling period and startup time of local nodes have no obvious influence on the algorithm, and the algorithm is not affected by the change of noise distribution form and targets number.