Research On Shift Strategy Of DCT Vehicle Base On Driver Intention Recognition

The shift control of the automatic transmission is one of the main research contents of automatic transmission,and the shift rule directly affects the power and economy of the vehicle.At present,most of the vehicles equipped with DCT on the market use the shift schedule based on the two parameters of throttle opening and vehicle speed.While the traditional shifting rules do not fit the driver’s driving intentions very well during driving,which affects the vehicle’s dynamics,economy,driving comfort and safety.In recent years,the intelligent connected vehicles have become a hotspot,as well as the higher requirements have been put forward for the intelligent control strategy of vehicles.The research on the recognition of the driver’s driving intention is not only the basis for intelligent connected vehicles to achieve personalized autonomous driving,but also promotes advanced the development of driving assistance systems,which is of great significance to the research of intelligent shifting strategies based on driving intentions.Firstly,a complete vehicle system simulation model of a wet dual-clutch automatic transmission is built by Matlab/Simulink software,which includes the engine model,gearbox model,shift logic control module,and vehicle dynamics module provides a simulation and verification basis for the driving intention shift strategy.A semi-physical driving simulation platform is built based on the Logitech G27 steering wheel,equipped with a high-fidelity virtual scene,and the UDP protocol is used to complete the communication between Matlab and Lab VIEW,which realizes the collection between the G27 steering wheel parameters and the vehicle model and date acquisition.The driving simulation platform can be used for the simulation verification of the algorithm.Secondly,this paper uses a fuzzy neural network to identify the driver’s intention including accelerate,brake,and decelerate.The accelerator pedal opening and its change rate,and the brake pedal opening and its change rate are selected as the driver’s intention recognition parameters.Based on driving simulation the platform,the intention recognition parameters are collected,and are used to train the acceleration,deceleration and braking intent recognition models.In order to maintain the advantages of the traditional shift schedule and solve its shortcomings,the study of the driving intention is added to the traditional shift strategy.Based on the basic shift schedule,the control logic of the stateflow module is used to limit deceleration and upshift.Taking into account the economy and power of the vehicle,the algorithm overcome the traditional shortcoming about the fixed shifting points,and solve the problem of high fuel consumption when using the best dynamic shift schedule under general acceleration intentions and insufficient power when using the best economical shift schedule.Finally,based on the built driving simulation platform,the driving intention recognition model is verified by offline and online tests,and the results show that the driver intention recognition model is accurate and reliable.Compared with the traditional shift strategy,the results show that the shift strategy based on the driving intention can limit the deceleration and unexpected upshift,improve the local acceleration performance of the vehicle,and take into account the power and economy of the vehicle.