Driving behavior oriented torque demand regulation for electric vehicles with single pedal driving

Abstract

Driving behaviors, induced by psychological activities and environment stimulation, impose the dominant impact on vehicle driving performance. To exhaustively improve the performance of electric vehicles (EVs), information unscrambled from various driving behaviors is recommended to be incorporated into the controlling process. In this context, a novel method is presented to regulate the torque demand of EVs with single pedal driving (SPD) by efficiently interpreting intention from different driving behaviors for eco driving. Specifically, a brand-new driving behavior identifier (DBI) is constructed by integrally employing the binary dragonfly algorithm (BDA) and adaptive neuro-fuzzy inference system with particle swarm optimization (ANFIS-PSO). Simultaneously, the whale optimization algorithm (WOA) generates the torque demand look-up tables (TDLTs) offline under different driving behaviors for SPD by referring to the constraints from drivability and energy efficiency. In the instant implementation, the driving behaviors are identified instantaneously by the DBI, and the homologous TDLTs are assigned to vehicle controller, thereby attaining efficient control of vehicle powertrain. A case study about the vehicle traction control is performed to validate the prospective optimal performance of the proposed method and further evaluate the impact on vehicle performance from driving behaviors.