Bandwidth-Aware Event-Triggered Adaptive DMPC for Vehicle Platoon Control

Self-driving vehicle platoons offer promising improvements in traffic efficiency and safety. However, their deployment is hindered by limited communication bandwidth and computational constraints in distributed control systems. This paper presents a bandwidth‑aware event‑triggered adaptive distributed model predictive control (DMPC) method. Firstly, to mitigate resource waste from frequent communications under bandwidth-constrained conditions in conventional DMPC, a bandwidth-aware event-triggered mechanism is designed based on a sigmoid threshold function of vehicle state errors. This mechanism adjusts the triggering threshold according to bandwidth availability, thereby suppressing unnecessary transmissions while maintaining a balance between communication efficiency and control performance. Secondly, a dual-input fuzzy adaptive module is introduced to reduce the computational burden in vehicle platoon control. This module takes maximum position and velocity errors as inputs to tune the prediction horizon dynamically. Finally, numerical simulation results show that under high bandwidth usage, the proposed method reduces communication resource consumption to 57.3%, while decreasing the communication frequency by up to 15.7% compared with the existing method. Meanwhile, it decreases the controller’s computational overhead by approximately 19.5% relative to fixed-horizon DMPC. The proposed approach enhances both communication and computational efficiency, making it applicable for resource‑constrained platoon control scenarios.