摘要
基于分层架构思路,设计了自适应巡航控制系统。决策控制层,结合前车运动状态动态计算可变车间时距,进一步通过期望车间距与实际间距的比较判定巡航控制模式;车距模式下,采用模糊神经网络通过驾驶数据训练自动生成隶属函数和模糊规则的模糊神经网络控制器进行跟车控制。执行控制层,基于车辆动力及制动系统特性,通过驱动或制动力矩控制实现车速与车间距的精确控制。通过构建“ACC控制策略-车辆动力学-交通场景”闭环系统模型,采用三类典型工况进行验证,仿真结果表明车辆响应较快且跟随稳定,满足控制目标和舒适度的要求。
An adaptive cruise control system is designed based on the hierarchical idea. In the decision-making control layer, a variable time gap to the preceding vehicle is calculated dynamically according to the movement of the preceding vehicle, and the cruise control mode is further determined by comparing the expected distance with the actual distance. In the distance control mode, the fuzzy neural network controller is used to automatically generate membership function and fuzzy rules through driving data training for vehicle following control. In the executive control layer, the precise control of vehicle speed and gap distance is realized by driving or braking moment control based on the characteristics of vehicle driving and braking system. The ACC control strategy is verified by three typical scenario through constructing a closed-loop system model of “ACC control strategy - vehicle dynamics - traffic scenario”. The simulation results show that the vehicle responds fast and follows steadily, which meets the control objectives and comfort requirements.
作者
李鹏飞
李占旗
王述勇
LI Peng-fei;LI Zhan-qi;WANG Shu-yong(CATARC(Tianjin)Automotive Engineering Research Institute Co.,Ltd.,Tianjin 300300)
出处
《新型工业化》
2019年第6期68-73,111,共7页
The Journal of New Industrialization
基金
天津市科技计划项目（17YDLJGX00020)
关键词
自适应巡航系统
模糊神经网络
分层控制
驾驶数据训练
Adaptive cruise control
Fuzzy neural network
Hierarchical control
Driving data training
