面向驾驶员行为研究的基础理论——多源信息融合算法综述

The Basic Theory Facing to Driving Behavior——An Overview on the Multi-information Fusion Algorithm

驾驶员行为的研究是智能运输系统,特别是其先进的车辆控制系统仿真、研究的关键环节。而目前交通运输领域及智能车辆领域中基于多源信息融合的驾驶员行为研究较为罕见。因此应从人机控制观点出发,应用模糊逻辑、非参数统计等理论以及智能协同技术,对驾驶员任务集聚、协同反应过程,特别是多源信息融合、人机协同效能在不同路网环境中的实现等建模关键理论与方法进行研究,建立驾驶员微观协同仿真模型。这是智能化交通控制与管理系统开发和实现的关键技术,对提高交通系统管理效率、缓解拥挤,减少污染和能源消耗,培育未来新产业,具有十分重要的理论价值和现实意义。

The research on driving behavior is the key of intelligent traffic system, especially for the advanced vehicle control system＇s simulation, but it is rarer in the traffic and intelligent vehicle field. So, it is necessary to research the key theo~ and method about the process of driver＇s gather and cooperation, especially for realization muhi-information fusion and the effect of people and vehicle synergistic reaction under different road net condition. Last building driver mico-cooperative simulation model set out from controlling view, using fuzzy logical and non-parameter theory and intelligent cooperative technology. It is the key technology for developing intelligent traffic control and management system, which has important theory value and practical means for improving management efficiency of traffic system, relieving crowd, reducing pollution and energy consumption and fostering new industry in the future.