| 基于神经网络方法的集成式驾驶员跟车模型 |
| |
| 引用本文： | 张磊,李升波,王建强,李克强. 基于神经网络方法的集成式驾驶员跟车模型[J]. 清华大学学报(自然科学版), 2008, 48(11) |
| |
| 作者姓名： | 张磊 李升波 王建强 李克强 |
| |
| 作者单位： | 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 |
| |
| 摘 要： | 为了提高驾驶辅助系统的跟车性能,基于神经网络方法建立了一种集成式驾驶员跟车模型.通过真实交通环境下的驾驶员实验获得了稳定跟车状态数据,并利用Kalman滤波器对数据进行了处理和估计.设计了以BP神经网络为核心的集成式模型结构,该模型以前车速度为输入,计算跟车过程中的两个特性参数并输入神经网络以模拟驾驶员控制的自车加速度.利用处理后的数据样本对网络进行了训练,并对该模型进行了仿真验证.仿真结果表明;神经网络模型具有模拟驾驶员跟车行为的能力,模型体现出较为准确的跟踪特性,并对不同的前车工况具有良好的适应性.
|
| 关 键 词： | 驾驶员模型 跟车 神经网络 |
Composite driver car-following model based on neural network approach |
| |
| ZHANG Lei,LI Shengbo,WANG Jianqiang,LI Keqiang. Composite driver car-following model based on neural network approach[J]. Journal of Tsinghua University(Science and Technology), 2008, 48(11) |
| |
| Authors: | ZHANG Lei LI Shengbo WANG Jianqiang LI Keqiang |
| |
| Affiliation: | ZHANG Lei,LI Shengbo,WANG Jianqiang,LI Keqiang(State Key Laboratory of Automotive Safety , Energy,Department of Automotive Engineering,Tsinghua University,Beijing 100084,China) |
| |
| Abstract: | A driver model in car-following scenario was developed based on neural network approach to improve the car-following performance of the driving assistance system.The model has a composite structure and is designed with back-propagation neural network to simulate the driver's control of the vehicle longitudinal acceleration.Experiments were made with an instrumented vehicle test-bed and the data segments of the steady state car-following scenario were extracted to obtain high-quality data in real traffic for... |
| |
| Keywords: | driver model car following neural network |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
