【作者】 李剑辉； 郑玲； 李以农； 乔旭强； 张紫薇；

【Author】 Li Jianhui;Zheng Ling;Li Yinong;Qiao Xuqiang;Zhang Ziwei;College of Mechanical and Vehicle Engineering,Chong Qing University;

【机构】 重庆大学机械与运载工程学院；

【摘要】 为改善自适应巡航(adaptive cruise control,ACC)系统的跟踪性能和舒适性,本文提出了一种考虑驾驶人特性和前车加速度扰动的ACC控制算法。首先,基于驾驶数据,建立驾驶人跟车模型;其次,考虑前车加速度的扰动,提出基于高斯过程(gaussian process,GP)的前车加速度在线预测方法;最后设计了基于模型预测控制(model predictive control,MPC)的ACC算法,以实现基于安全、舒适、节能等多目标,并考虑驾驶人驾驶偏好的拟人化自适应巡航(ACC)控制。仿真结果表明,与传统的LQR算法相比,本文提出的拟人化ACC算法能够在车辆跟随过程中,充分体现了驾驶人的驾驶习惯,改善了ACC系统的舒适性和宜人性。更多还原

【Abstract】 An ACC algorithm that fully considers the characteristics of the driver and disturbance of the preceding vehicle is proposed in order to improve tracking performance and comfort of the Adaptive Cruise Control(ACC) system in this paper.Firstly,the driver car-following model is established based on the driving data;Afterwards,aiming at disturbance of preceding vehicle,an online prediction method of the acceleration of the leading vehicle based on Gaussian Process(GP) is proposed.Finally,an MPC-based ACC algorithm is designed to realize the human-like adaptive cruise control(ACC) based on safety,comfort,energy-saving and other objectives,and considering the driver’s driving preferences.The simulations show that,compared with the traditional LQR algorithm,the human-like ACC algorithm in this paper can demonstrate the driving habits for drivers,which will help to improve the agreeableness and the comfort of the ACC system.更多还原