AT线控换挡执行机构挡位位置自学习算法研究

对于全自动泊车系统搭载的线控换挡执行机构,由于制造、装配误差会导致线控换挡执行机构装配于变速器后的R、N、D位置可能会与实际情况有偏差,从而影响变速器性能,所以从系统的动力学特性出发,提出了一种线控换挡执行机构下线学习算法,利用Matlab/Simulink建立系统动力学模型和算法模型,并开发了线控换挡执行机构控制器软硬件。通过仿真分析与实际台架测试表明,该下线挡位位置自学习算法能够在很短的时间内学习到精确的R、N、D挡位位置,不仅满足了车辆下线对时间的严格要求,还满足了变速器换挡的精度要求。

Study on Self-learning Algorithms for Gear Shift Positions of AT Shift-by-Wire Actuator

The manufacturing and assembly errors may lead to the deviations of the R,N and D shift lever positions after the shift-by-wire actuator is equipped in the auto parking assist,which may affect the transmission performance.based on the dynamic characteristics of the system,this paper proposed a selflearning algorithm for the shift-by-wire actuator in end-of-line testing,established a system dynamics model by using Matlab/simulink,and also developed the hardware and software for the actuator.the simulation analysis and actual bench test show that the self-learning algorithm can learn the accurate R,N,D gear shift positions in a very short time,so it meets the strict time requirements of vehicle end-of-line testing,and also meets the accuracy requirements of transmission shift.