基于行为的智能车辆导航控制体系结构

ACTIVITY-BASED CONTROL ARCHITECTURE FOR INTELLIGENT VEHICLE NAVIGATION

在分析典型智能车辆导航控制体系结构优、缺点的基础上,提出一种基于行为的智能车辆导航控制体系结构。该体系结构使用黑板来协调各模块之间的交互,并把智能车辆的行为划分为反射行为和目的行为。反射行为由数据信息驱动,目的行为由黑板选取;通过不断被选取的行为组成的行为序列完成导航任务。在理论分析的基础上进行仿真计算,结果显示智能车辆能有效地完成路面标志线跟踪、障碍躲避、墙面跟踪和路径信息导航各行为,并最终顺利到达目标点。最后,对智能车辆的标志线导航进行实车试验,得到较好的效果。

With the analyses on the advantages and disadvantages of the two typical architectures for intelligent vehicle navigation, an architecture for intelligent vehicle navigation based on activities control is presented. This architecture uses a blackboard to coordinate the interaction between modules, and classifies the activities of the intelligent vehicle into reflecting activity and voluntary activity. The reflecting activity is activated by data and the voluntary activity is selected by the blackboard. The navigation task is accomplished by sequencing these activated activities. Based on the theoretic analyses, simulations are performed and their results show that the vehicle can efficiently follow the guide path, avoid obstacles and follow walls and road edges, and finally reach the destination. The experiment for path-guided navigation is carried out and the results are satisfying.