基于认知结构的高速列车对标停车控制算法  

Cognitive structure-based brake control algorithm of high-speed train

作　　者：郭北苑[1] 孙玉龙 GUO Beiyuan;SUN Yulong(State Key Laboratory of Rail Traffic Control and Safety,Beijing Jiaotong University,Beijing 100044,China;School of Electronic and Information Engineering,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学轨道交通控制与安全国家重点实验室,北京100044 [2]北京交通大学电子信息工程学院,北京100044

出　　处：《北京交通大学学报》2021年第5期8-15,共8页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金(1734210)。

摘　　要：高速列车进站停车是列车自动驾驶系统(Automatic Train Operation,ATO)的一个重要功能,该功能需要在保证乘客舒适性的同时实现精确的对标停车.针对高速列车自动驾驶系统的精确进站停车问题,首先分析了人类高速列车司机在停车过程中的认知处理过程;结合认知结构理论对这一过程进行建模,构建了高速列车停车对标任务认知结构模型,基于该模型提出了一种基于认知结构的对标停车控制算法.最后对设计的算法进行了评估与分析.结果表明:该算法与人类高速列车司机具有相似的驾驶特征,与PID控制算法相比在调整次数、冲击率等性能指标上更加优异,对不同的初始条件也有良好的适应性.High-speed train braking in station is an important function of the Automatic Train Operation(ATO) system, and this function needs to ensure the comfort of passengers while achieve accurate brake control. To achieve accurate high-speed train in-station brake control, this paper focuses on analyzing the cognitive processing process of high-speed train human drivers during the brake control process, and combines the cognitive structure theory to model the cognitive process. The cognitive structure model of high-speed train brake control is then established. Based on this model, a brake control algorithm for high-speed train is designed. Finally, the algorithm is analyzed and evaluated. The results show that the algorithm has driving characteristics similar to that of high-speed train human drivers. Compared with the PID control algorithm, the proposed algorithm shows better performance indicators, such as adjustment frequency and impact rate, and it also has good adaptability to different initial conditions.

关 键 词：交通信息工程及控制 列车自动驾驶 停车对标 认知结构 ACT-R 

分 类 号：U284.48[交通运输工程—交通信息工程及控制]