A Koopman-based approach for torque vectoring in electric vehicles

Torque Vectoring plays a pivotal role in enhancing vehicle dynamics and performance. In electric vehicles with multiple motors, the torque at each wheel may be controlled independently, offering significant opportunity to enhance safety and stability. This study explores the application of Koopman-based Model Predictive Control (MPC) in torque vectoring systems. By considering an electric vehicle equipped with four in-wheel motors, a linear Koopman model is identified from data. The model is then used to formulate a linear MPC. The performance of the scheme is assessed through numerical simulations.