Front-to-rear torque vectoring Model Predictive Control for terminal understeer mitigation

In this paper we propose a constrained optimal control strategy that uses combined velocity, yaw and sideslip regulation to stabilize the vehicle near the limit of lateral acceleration using the front-to-rear axle electric torque vectoring configuration of a fully electric vehicle. A bicycle model is used to find reference steady-state cornering conditions for the controller to follow, while a nonlinear vehicle model with nonlinear and coupled tyre forces is used to design a linear Model Predictive Control (MPC) strategy. Despite the limited capabilities of a front-to-rear torque vectoring configuration when compared to the left-to-right wheel or four wheel independent torque vectoring cases, we then show that such a configuration is still effective in stabilizing the vehicle in the limits of handling by validating the proposed solution in a high fidelity simulation environment.