Abstract: Aiming at the existing problems of synchronizing speed difference and shifting shock in conventional gear-shifting control for hybrid electric vehicle, a new way of gear-shifting control with speed / torque double-synchronization is proposed. During speed synchronization, the target speed is adjusted based on the speed change at driven end, and the output torque of motor is adjusted according to synchronizing speed difference so as to make the speeds at both driving and driven ends synchronize, with same moving tendency, for reducing shifting shock. To improve the performance of torque closed-loop PID control, fuzzy RBF neural network is adopted to tune the parameters of PID control, with the rapidity and accuracy of motor torque tracing enhanced. The results of experiment show that compared with conventional control, the gear-shifting control with speed/torque double-synchronization can not only shorten shifting time but also significantly reduce shifting shock

Key words: hybrid electric vehicles, gear-shifting control, fuzzy RBF neural network, speed/torque double-synchronization, shifting shock