基于决策图的轨迹运动趋势提取

Overall trajectory trend extraction based on decision graph

在军事应用中,为了提取群体目标的整体运动趋势,提出了一种基于决策图的轨迹聚类来提取轨迹运动趋势的方法。该方法不需要预设参数,且聚类中心的个数既可以通过决策图人工确定,又可以通过数值检测策略自动确定,由此减轻了算法对领域知识的依赖,增强了算法的适用性。仿真实验表明:该方法能正确确定轨迹聚类簇,且对轨迹噪声有一定的抑制作用。

In military application,the trajectories movement trend of the group target reflects its strategic intention to some degree.A trajectory clustering method based on decision graph is proposed to extract the overall motion trend of trajectories.This method does not demand the preset parameters,and the number of clustering centers can be determined manually by decision graph,or automatically determined by a numerical detection strategy,which reduces the dependence on domain knowledge and enhances the applicability of the algorithm.The simulation results show that the proposed method can correctly determine the trajectory clustering clusters and has some inhibitory effect on the trajectory noise.