A Stable Control Method Designing for Longitudinal Platoon Problems
Ziliang Wang, Xiaohui Yu, Wenbo Li, Shuyou Yu, Hong Chen · 2023
This paper proposes a neural network architecture that guarantees stability and applies it to address the longitudinal vehicle platoon problem. Considering the safety and consistency requirements during the operation of the platoon, traditional neural network algorithms struggle to meet the stability requirements. To address this issue, this paper introduces a control framework, i.e., QRNET, that combines the Linear Quadratic Regulator (LQR) and Neural Network (NN). QRNET ensures asymptotic stability at least near the equilibrium point and improves the problem-solving dimensionality. Simulation results demonstrate that the proposed neural network architecture ensures stability and satisfactory dynamic performance of the longitudinal vehicle platoon.