基于驾驶员意图识别的液力机械式自动变速器换挡规律优化

Optimization of Shift Schedule for Automatic Transmission Based on Driver Intention Recognition

构建了基于隐马尔可夫模型的驾驶员意图模型,用于辨识不同的驾驶意图。以油门踏板开度以及制动踏板开度等作为输入数据,结合BaumWelch算法训练驾驶员意图模型,以计算所得新数据的极大似然值作为辨识结果,验证了驾驶员意图模型的准确性。根据识别出的驾驶员意图模型参数,基于基本换挡规律,对不同意图下的换挡规律进行优化。仿真结果表明,优化后的换挡规律更符合驾驶员期望的挡位操作,且有助于消除意外换挡的现象,验证了意图模型优化的有效性。

A driver intention model is described based on the hidden Markov model, which is used to identify a driver ~s intentions. The model is trained with Baum - Welch algorithm and data on the degree of opening of accelerator pedal and the brake pedal is inputted; thereafter the maximum likelihood value is calculated as recognition results, which verify the accuracy of the model. On the basis of basic shift schedule and the identified driver intention, the shift schedule was optimized under different intentions. The simulation results show that the optimized shift schedule is more in accordance with the driver＇s intended action, and helps eliminate accidental shift phenomenon, which verifies the effectiveness of the optimization.