交通流特征深度认知的车队运行参数优化方法

为提升车队对周围交通流环境的认知能力，获取车队周围多车的运行模式，同时通过改变车队运行参数实现车队群体对周围多车群体运行模式的诱导变化，为提升车队及交通流整体运行效率提供优化策略，提出了基于狄利克雷混和高斯过程的车队周围多车运行模式获取算法，将车队周围多车车辆的复杂运行模式视为混和高斯过程，利用狄利克雷分布作为高斯混和权重的先验分布，建立车队周围多车运行模式速度场，从而获取车队周围多车运行模式并分类；通过比较不同多车运行模式下车队的运行效率，提升车队对所处运行环境的认知能力。研究结果表明：利用非参数贝叶斯算法将复杂的多车运行状态进行分类，获取的车队周围多车运行模式可表现为对车队运行效率产生不同影响的速度场；通过将车队周围的多车运行仿真数据分成多个运行模式，可获取不同多车运行模式下车队及交通流整体运行特性；通过更改车队运行参数，观察不同多车运行模式的占比变化，可获取车队运行参数对所处运行环境改变的影响趋势；车队周围多车运行模式的获取，不仅可以提升车队对周围运行环境的认知能力，使得车队能够选取有利路径行驶，同时能够为车队运行策略的优化提供有效的信息。

Platoon Operating-parameter Optimization Method Based on Deep Cognition of Traffic-flow Features

To improve the platoon's cognitive ability of the driving environment, the multi-vehicle motion patterns must be obtained, changes in the surrounding multi-vehicle motion pattern must be induced by changing the platoon operating parameters, and an optimization strategy must be provided. This paper proposes a multi-vehicle motion-pattern-acquisition algorithm based on the Dirichlet process-mixed Gaussian process. This method establishes the multi-vehicle motion-pattern velocity field around the platoon by considering the multi-vehicle motion pattern around the platoon as a mixed Gaussian process and considering the Dirichlet process as the prior distribution of the Gaussian mixture weights. The platoon's perception of the driving environment was realized by comparing the operating efficiency of vehicles in different multi-vehicle motion patterns. The research results show that the non-parametric Bayesian algorithm classifies the multi-vehicle operating state and obtains the multi-vehicle motion pattern around the platoon as different velocity fields in a fixed area. In addition, different velocity fields have different effects on the operation efficiency of the platoon. The overall operating characteristics of both the platoon and traffic flow corresponding to different multi-vehicle motion patterns were obtained by dividing the multi-vehicle operation simulation data into multiple motion patterns. This study changed the operating parameters of the platoon to research the influence on the changes in the operating environment by observing the percentage variation of the motion patterns. The acquisition of the multi-vehicle motion patterns around the platoon improve the platoon's cognitive ability in the driving environment, and allows the platoon to choose an optimized path and provided information for the optimization of the platoon's operational strategy.