Towards Vehicular Domain Controller: A Multi-agent Approach to Control Coordination Considering Actuator Dynamics

Lu Xiong,Fankai Zhou,Haoran Lv,Yuhua Zhu,Chen Tang
DOI: https://doi.org/10.1109/itsc57777.2023.10421925
2023-01-01
Abstract:The integrated motion control framework revolutionizes vehicle performance by incorporating sub-functional systems into a unified model. However, such integration also means that the framework becomes rigid in its modeling form, making it challenging to keep a balance between universality and heterogeneity. The distributed control framework trades design complexity for flexibility in both structure and computation. It enables various actuator characteristics to be considered in control modelling in a plug-and-play fashion, bringing more possibilities for control synthesis and computational resource allocation. This paper proposes a novel multi-agent coordination control strategy considering actuator dynamics for chassis domain control. It treats the active front steering (AFS), active rear steering (ARS), and direct yaw moment control (DYC) modules as agents, considering them as the fundamental units for precise modeling and scheduling calculations. Under the distributed control framework, a distributed model predictive control algorithm that combines asynchronous triggering and event triggering is developed to balance control performance and computational efficiency. The flexible triggering mechanism ensures prompt responses to disturbances and facilitates coordinated multi-agent control, leading to optimal control performance.
What problem does this paper attempt to address?