论文部分内容阅读
为改善纯电动汽车平路起步的操控性,提出了一种基于驾驶员意图识别的电动汽车平路起步控制策略.将驾驶员意图分为缓慢起步,急速起步,正常起步,利用k-medoids聚类算法对数据进行离散化处理,利用双层隐形马尔科夫模型对驾驶员的起步操作意图进行识别,分别制定了控制策略,使得缓慢起步状态下能将车辆稳定在较低车速,急速起步状态下能够最大利用车辆的动力性.仿真结果表明,该控制策略能正确识别驾驶员的起步操作意图,在缓慢起步状态下能够将车速稳定在5km/h,急速起步状态下能够达到地面能提供的最大加速度,保证了车辆起步时的动力性、安全性、操控性,实现既定的控制要求.
In order to improve the controllability of pure electric vehicles, this paper proposes a control strategy based on the driver’s intention to identify the start-up of electric vehicles. The driver’s intention is divided into slow start, rapid start, normal start, k- Class algorithm to discretize the data, the use of double hidden Markov model of the driver’s intention to identify the starting operation, were developed control strategy, making the slow start state can stabilize the vehicle at a lower speed, rapid start state Under the condition of slow start, the vehicle speed can be stabilized at 5km / h, the ground speed can be provided under the condition of rapid start-up, the control strategy can correctly identify the initial operation intention of the driver, Maximum acceleration, to ensure the start of the vehicle power, safety, handling, to achieve the established control requirements.