集中式自适应网格IMMJPDA算法

A centralized multisensor multitarget tracking algorithm based on variable structure multiple-model

现有的集中式交互多模型联合概率数据互联(IMMJPDA)算法在多模型这个意义上都是基于固定结构的,而固定结构多模型算法存在的缺陷这些算法都不可避免的存在。为此,将一种变结构多模型算法——自适应网格交互多模型(AGIMM)算法和联合概率数据互联(JPDA)算法相结合,提出了用于多传感器多目标跟踪的集中式自适应网格IMMJPDA(AGIMMJPDA)算法。该算法通过自适应网格实现模型集合自适应调整来克服固定结构IMMJPDA存在的缺陷。仿真结果显示,该算法可以有效克服固定结构IMMJPDA算法存在的缺陷,并提高IMMJPDA算法的费效比。

Existing centralized interacting multiple model joint probabilistic data association （IMMJPDA） algorithms have a fixed structure in the sense that they use a fixed set of models at all times, and they inevitably have the limitation of fixed structure multiple-model. A variable structure multiple-model algorithm, i. e., adaptive grid interacting multiple model （AGIMM）algorithm is combined with joint probabilistic data association （ JPDA ）, centralized adaptive grid IMMJPDA （AGIMMJPDA） algorithm is proposed using for muhisensor multitarget tracking. The algorithm achieves model-set adaptation by adaptive grid, the limitation of fixed structure IMMJPDA is overcomed. Finally, simulation indicates that this algorithm can overcome the limitation of fixed structure IMMJPDA and effectively improve the cost-effectiveness of IMMJPDA.