考虑舒适性的电动汽车制动意图分类与识别方法  

Braking intention classification and identification considering braking comfort for electric vehicles

作　　者：潘宁[1] 于良耀[1] 宋健[1] PAN Ning YU Liangyao SONG Jian(State Key Laboratory of Automotive Safety and Energy, Department of Automotive Engineering, Tsinghua University, Beijing 100084, China)

机构地区：[1]清华大学汽车工程系,汽车安全与节能国家重点实验室,北京100084

出　　处：《清华大学学报（自然科学版）》2016年第10期1097-1103,共7页Journal of Tsinghua University(Science and Technology)

基　　金：国家“八六三”高技术项目(2012AA110903);国家科技支撑计划项目(2015BAG17B05)

摘　　要：液压执行机构(HCU)在电动汽车上被广泛用作电液复合制动系统的液压力精确调节机构。为改善制动舒适性,需要采用合适的制动意图分类与识别方法。提出一种以提高舒适性为目的的制动意图分类方法,将制动意图分为常规减速、紧急制动和压力跟随,并根据分类结果控制液压执行机构;提出一种制动意图在线识别方法,用于在制动过程中在线识别制动意图的类别。该方法利用多传感器数据融合,使用神经网络对制动意图进行识别。仿真及试验结果表明,采用所提出的制动意图分类与识别方法后制动舒适性及安全性得以改善。Hydraulic control units （HCUs） are widely used as precise pressure regulators for the composite brakes in electric vehicles. The braking comfort can be improved by appropriate braking intention classification and identification. A braking intention classification method is developed to improve braking comfort that classifies the braking intention as normal deceleration, emergency braking and a pressure following pattern. The pressure control method is then based on the classification results. The on-line braking intention identification method uses multiple sensors and a neural network. Simulations and tests show that the braking comfort and safety are improved by this method.

关 键 词：制动意图识别 电动汽车 液压执行机构 制动舒适性 主动控制 

分 类 号：U463.5[机械工程—车辆工程]