| Vehicle test is important in automobile industry and intelligent transportation.Appropriate driver modeling can reduce the amount of money and time used by vehicle manufacture.Drving styles are of vital importance in vehicle test and in some specific applications,the speed tracking precision is strictly required.Thus it’s important to build a driver model capable of style-retaining and self-learning for speed control.And based the built driver model and vehicle model,we can automate the calibration of shift schedule by means of biomic optimization algorithm.We propose a unified driver model to model the two kinds of driving maneuvers inherently existing in every driver,i.e.,a commonality model to mimic the common driving skills and a personality model to imitate the unique driving characteristics of each driver.A networked PID controller was proposed for the commonality model.By employing the distal learning control approach,the commonality model was self-tuned for specific applications,whereas the personality model was initially established based on the real-world driving data and was later re-tuned with the collaboration of the commonality model for specific applications.The FTP-72 driving cycle was borrowed as the desired speed profile.Driving cycle test simulation results show that the speed control performance is continuously improved to a satisfactory extent through self-learning,meanwhile the driving style of the original human driver is well retained.We propose to automate the calibration of shift-scheduling using the bionic optimization,i.e.,PSO in this work,to guide the searching process,and to integrate driving styles into the calibration by equipping the robot drivers with personalized driver models.The personalized driver model is established by imitating human driving behavior,and is employed as a robot driver to conduct the driving cycle test,i.e.,FTP-72 or US06,for candidate shifting schedules.The shifting performance is evaluated online via the computed performance index and/or AVL-Driver,regarding both driveability and fuel economy.Guided by PSO,candidate schedules are generated,tried and evaluated until an optimal or near-optimal solution is obtained through iterations.Experiments are presented to verify the feasibility and effectiveness of the proposed scheme. |
