Distributed State Observers-based Nonlinear Controller for Vehicle Platoon

This paper proposes an adaptive control algorithm for vehicle platoon based on distributed observers. Firstly, considering that the target trajectory is only accessible to leader vehicle, a distributed state observer is proposed for follower vehicles to estimate the state of leader vehicle. Secondly, based on the observer, an adaptive tracking controller is designed to make follower vehicles track the trajectory of leader vehicle. The controller utilizes a neural network to approximate the unknown dynamics, thereby removing the assumption that vehicle dynamics is fully known. Additionally, this controller utilizes hyperbolic tangent functions to compensate the dynamic coupling between vehicles. Finally, stability of the vehicle platoon is proved via Lyapunov method, and the effectiveness of the controller is verified through numerical simulation and SimulinklPreScan co-simulation.

1. Introduction

In the past decade, the rapid growth in car ownership has increased traffic congestion, a problem that cannot be easily solved by merely expanding infrastructure. Intelligent transportation systems (ITS) consisting of connected and automated vehicles (CAVs), smart highways and cloud computing platforms are expected to solve the traffic congestion problem [1]. Researches on ITS have focused significantly on the platoon control problem for CA V s. [2]. Many countries have carried out related research projects, such as the United States has the PATH initiative [3], and Japan has implemented the ITS for Energy [4].