Global/Local Performance Analysis of Driving Force Based Hierarchical Decentralized Motion Control System for Multi-Motor Vehicles

Although driving force control (DFC) has been proven to be one of the most promising strategies for traction control, its integration with other motion control functions in electric vehicle (EV) remains an open research challenge. This is because DFC has traditionally been developed utilizing single-wheel-model without consideration for other motion control objectives. To address this limitation, this paper presents a preliminary investigation into a hierarchical decentralized motion control system for multi-motor EVs, based on the DFC. In this architecture, the upper layer is responsible for vehicle speed control, while the lower layer comprises multiple DFC units-treating speed and driving force control are as global and local objectives, respectively. Additionally, a middle layer is to allocate the driving forces to attain supplementary goals. To facilitate a systematic design methodology, a role-sharing mechanism between upper and lower layers is established via a concept of "global/local shared model set."This concept enables both layers to be designed using standard robust control techniques. Notably, the design complexity remains independent of the number of actuators. Using a real in-wheel-motor vehicle, this paper demonstrates-both theoretically and numerically-that the trade-off between global and local objectives can be analyzed through the size of the shared model set.