论文部分内容阅读
对移动对象的不确定轨迹实时、高效的预测是智能交通系统研究热点之一.针对现有轨迹预测方法在动态环境中预测精度不高、实时效果差的问题,提出一种环境自适应车辆轨迹预测方法(EAVTP),该方法主要步骤包括:首先运用历史数据构造虚拟参考点有效改进环境动态变化通讯信号不稳定情况下车辆位置不准确信息;其次利用高斯混合模型对虚拟参考点数据与历史轨迹数据集训练实现环境自适应功能;最后利用虚拟参考点和历史轨迹数据集对车辆轨迹实时预测.最后对所提方法模拟仿真,结果表明EAVTP方法具有一定的环境自适应性,且预测精度和实时性比现有其它方法有所提高.
Real-time and efficient prediction of uncertain trajectory of moving objects is one of the hot topics in ITS research.Aiming at the problem of low prediction accuracy and poor real-time performance of existing trajectory prediction methods in dynamic environment, an environment-adaptive vehicle trajectory (EAVTP). The main steps of this method include: Firstly, using the historical data to construct the virtual reference point can effectively improve the inaccurate information of the vehicle position under the condition of unstable dynamic communication signal. Secondly, using the Gaussian mixture model to reconstruct the virtual reference point data and historical trajectory Finally, the simulation of the proposed method is carried out. The results show that the EAVTP method has a certain environmental adaptability, and the prediction accuracy and real-time There is some improvement over other existing methods.