Disturbance and uncertainty compensation control for heterogeneous platoons under network delays

• Published in: Computers & electrical engineering Vol. 123; p. 110066
• Main Authors: Silva, Emerson A., Mozelli, Leonardo A., Alves, Armando, Souza, Fernando O.
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.04.2025
• Subjects: Communication time delay; Distributed networked control system; Disturbance rejection; Heterogeneous vehicular platoon; Integral control; Vehicle-to-Vehicle communication; Heterogeneous vehicular platoon; Integral control; Distributed networked control system; Disturbance rejection; Vehicle-to-Vehicle communication; Communication time delay
• ISSN: 0045-7906
• DOI: 10.1016/j.compeleceng.2025.110066

Vehicle platooning, i.e., enforcing a longitudinal formation of vehicles that follow each other closely, is a promising solution to stringent problems that affect traffic worldwide. Better results ensue using autonomous driving support systems and connectivity technology enabled by the Internet of Vehicles, typically in a Vehicle-to-Vehicle topology through a wireless network. We propose a new protocol for stabilizing vehicle platoons to address the real-world challenges of this distributed networked control system. We assume that the connected and automated vehicles drive in dedicated lanes, operating under a constant spacing policy, but subject to heterogeneous dynamics, connected through look-ahead topologies, with heterogeneous delays in the transmission channels, contending with distributed disturbances from external or internal sources. The modified controller incorporates an integral term and delay compensation, thus enabling the formation to counteract disturbances and zero the spacing errors in steady-state. Unlike previous approaches considering disturbances originating only from the leader, our proposal handles distributed disturbances, such as road grade, wind gusts, and parametric uncertainty. We formally demonstrate the necessary and sufficient conditions for internal stability and disturbance rejection. Experiments in the Car Learning to Act (CARLA) simulator illustrate the effectiveness and endurance under severe disturbances. We provide numerical comparisons against recent methods, and the analysis reveals that the proposed method improves the performance when real-life conditions are present. •Vehicle platooning depends on reliable IoT communication subjected to time-delays•In real-world scenarios vehicles face disturbances such as road grade and wind gusts•Model considers look-ahead topologies, heterogeneous time-delays and dynamics•A quadratic cost design ensures internal stability and null steady-state errors•Numerical simulations using CARLA and a full nonlinear model validate the proposal