Energy-Efficiency Optimization of Torque Vectoring Control for Battery Electric Vehicles

Abstract

We propose a novel torque vectoring concept for battery electric vehicles propelled by wheelindividual electric machines. Under consideration of vehicle dynamics and electric machine efficiency, this paper focuses on taking advantage of the provided degrees of freedom to reduce energy consumption. The proposed approach improves energy efficiency by introducing two functional elements: Firstly, an optimization problem based on a detailed nonlinear four-wheel vehicle model is solved to create reference sideslip angle and yaw rate that minimize the energy consumption due to wheel slip. A subsequent Linear Quadratic Controller calculates a yaw torque around the vehicle's vertical axis to follow the desired reference. To generate both, the acceleration demanded by the driver and the yaw torque requested by the LQR, a torque allocation algorithm is required. The proposed torque allocation algorithm takes into account the energy efficiency characteristics of the electric machines to provide the required propulsion force and the yaw torque with minimized loss power. Based on experimental results, the proposed torque vectoring control system leads to an energy consumption reduction of around 10% for many typical driving situations containing both, high and low lateral acceleration scenarios.