A Delay Analysis for Distributed Model Predictive Control on Vehicles Platooning through a Low-Cost Hardware-In-the-Loop Testbed

Vehicle platooning control is a branch of Cooperative-Intelligent Transport Systems that aims at reducing the risk of rear-end collisions and improving traffic fluidity. Platooning technologies can benefit of advanced and distributed control strategies, such as the Model Predictive Control, suitable to deal with constraints deriving from inter-vehicle distances and communications as well as to provide adaptive behaviour against variations of speed limits, safety distance, environmental and meteorological parameters provided by the Smart Road. Since vehicle platooning is based on inter-vehicle communications, the control strategy needs to be robust with respect to communication delays. This paper performs a real-time and low-cost Hardware-In-the-Loop testbed based on Raspberry Pi boards that emulate a vehicle platoon managed by a Distributed Model Predictive Control strategy. Results showed a good robustness of the proposed control strategy with respect to communication delays, considering a variable transmission delay in the range [100, 400] ms. The obtained results pave the way to real-road tests, considering connected vehicles equipped with IoT Gateways based on Raspberry Pi hardware.