RBFNN-Based Consensus Control of Vehicle Platoon With an Adaptive Event-Triggered Mechanism Under DoS Attacks

Vehicle platoon is an essential subject in research on intelligent transportation and autonomous driving. Current control systems frequently rely on accurate feedback linearization, which overlooks denial-of-service (DoS) attacks and assumes full parameter information, assumptions that are rarely realized in environments that are noisy and prone to disturbances. The communication topology is described as a directed graph with a spanning tree in this work, which develops an adaptive distributed control framework for uncertain nonlinear vehicle platoons under DoS attacks. The unknown nonlinear dynamics are modeled using radial basis function neural networks (RBFNNs), and coordinated platoon behavior is achieved by developing an adaptive backstepping control approach. To reduce unnecessary control updates, two Zeno-free event-triggered techniques are also presented. Practical consensus is ensured by Lyapunov-based analysis, and simulation results show that string stability is preserved under two-predecessor-following (TPF) and bidirectional (BD) topologies-even when the DoS attack rate hits 70%.