Distributed model predictive anti-disturbance control for vehicle platoon under asynchronous sampling and stochastic communication delay

Abstract A vehicle platoon control system that relies on vehicle-to-vehicle communication often faces nonlinear coupled effects caused by stochastic communication delay and dynamics uncertainty. Additionally, jittering and aging of sensor/sampler devices issues make periodic sampling strategies incompatible with actual communication transmission. To address these challenges, this paper proposes a distributed model predictive platoon control method based on asynchronous sampling. First, an asynchronous sampling model that aligns with a stochastic probability distribution is introduced to simulate the real-world phenomenon of random signal sampling. Second, a longitudinal vehicle time-delay dynamic tracking model under the asynchronous sampling strategy is constructed to effectively handle the coupling between stochastic delays and vehicle dynamic characteristics. Based on this, a distributed model predictive controller (DMPC) is designed using the Karush-Kuhn-Tucker (KKT) condition method to obtain the nominal desired acceleration, ensuring feasible solutions. Then, a distributed extended state observer (DESO) is developed to estimate disturbances caused by communication delay and dynamic uncertainties in the platoon control system, enabling compensation control. Finally, the global convergence of both the distributed extended state observer and the distributed model predictive controller is theoretically proven. Extensive simulation results validate the effectiveness and superiority of the proposed strategy.