Adaptive Coordinated Motion Control: Automated Tuning for Predictive Safety in Electric Vehicles

Integrating subsystem functions in vehicle chassis control can significantly enhance motion performance. However, the coupling characteristics of vehicle dynamics may cause conflicts among subfunctions. Increased control requirements also intensify controller tuning burden and make parameters highly dependent on driving conditions. To address these issues, this article proposes an automated recursive tuning based predictive safety control for electric vehicles. This strategy optimally allocates additional torques to enhance vehicle handling agility, driving stability, traction ability, and longitudinal-lateral motion coordination simultaneously. An unscented Kalman filter based automated recursive tuning system is employed to balance these control requirements automatically and offer real-time adaptability to varying conditions. Distinct real-world experiments are designed for closed-loop tuning and control. First, a tuning experiment involving sequential longitudinal and lateral motions allows tuned parameters to quickly adapt to varying driving conditions and integrate dynamic features. Then, extreme longitudinal and lateral experiments are conducted independently for predictive safety control. Compared with results with controller tuned manually, the proposed strategy enhances handling agility, lateral stability, and antiskidding in the lateral test by 12.16%, 15.16%, and 30.49%. In the longitudinal test, traction ability and accelerating capacity are improved by 51.24% and 9.4%. Besides, vehicle lateral and yaw fluctuations are notably reduced by 49.78% and 42.96%.