Network Scheduling for Distributed Controls of Electric Vehicles Considering Actuator Dynamic Characteristics

Yang Zhao,Weiwen Deng,Jian Wu,Rui He

DOI: https://doi.org/10.4271/2017-01-0019

2017-01-01

SAE International Journal of Passenger Cars - Electronic and Electrical Systems

Abstract:Electric vehicle (EV) has been regarded as not only an effective solution for environmental issues but also a more controllable and responsible device to driving forces with electric motors and precise torque measurement. For electric vehicle equipped with four in-wheel motors, its tire longitudinal forces can be generated independently and individually with fully utilized tire adhesion at each corner. This type of the electric vehicles has a distributed drive system, and often regarded as an over-actuated system since the number of actuators in general exceeds the control variables. Control allocation (CA) is often considered as an effective means for the control of over-actuated systems. The in-vehicle network technology has been one of the major enablers for the distributed drive systems.The vehicle studied in this research has an electrohydraulic brake system (EHB) on front axle, while an electromechanical brake system (EMB) on rear axle. The focus of this study is on the scheduling of the networked control for the distributed systems. A network control system (NCS) model is first established with a continuous-time based plant model, a discrete controller and a FlexRay communication model, in addition to the electric vehicle model with tire model and driving/braking actuator models.An integrated vehicle control allocation method based on constraints optimization is then proposed. The dynamic characteristics of each actuator are considered on the control allocation with higher-bandwidth actuators being used for faster control commands. It is also considered for the scheduling optimization of FlexRay which minimizes a cost function and subject to a set of constraints.Finally, a hardware-in-the-loop (HIL) system is built, which consists of a driving simulator, four in-wheel motors, EHB and EMB brake systems, all connected via FlexRay communication in order to validate the proposed scheduling methods. The experimental results have demonstrated the improved responses of actuators when considering their dynamic characteristics, improved control performance via properly designed control allocation, and reduced bus efficiency via optimized scheduling methods.