| KEYWORD |
Design and numerical validation of distributed model reference control algorithms for vehicle platooning via network-based control solutions
Thesis in external company 
Thesis abroad
keywords DISTRIBUTED CONTROL, PLATOONING, TRAFFIC CONTROL, VEHICLE
Reference persons ALESSANDRO RIZZO
External reference persons Prof. Umberto Montanaro and Prof. Aldo Sorniotti, University of Surrey, UK
Thesis type SIMULATION AND DESIGN
Description A vehicle platoon is a cooperative driving scenario where a group of consecutive Connected and Autonomous Vehicles (CAVs) share information over a communication network to travel along a highway in the same lane with a short inter-vehicle distance and at the same velocity. There are several benefits for organising the road vehicles in platoons, e.g., (i) increasing traffic flow, (ii) reducing fuel consumption and pollutant emissions, (iii) improving road safety, and (iv) increasing passengers’ comfort. From a system and control viewpoint, a platoon of CAVs can be represented as a set of dynamical systems interacting over a network structure and the control aim can be formulated as the synchronisation of the states of each vehicle (i.e., node of the network) to the state of the leader vehicle (i.e., synchronisation/pinning control of a network of dynamical systems). However, to achieve these benefits the motion of the CAVs must be precisely synchronised despite uncertainties on the vehicle parameters and external disturbances. In this framework, the target of the thesis is (1) to design distributed model reference adaptive control (MRAC) solutions that can systematically tackle vehicle parameter variations, unmodelled vehicle nonlinearities and disturbances. MRAC is an effective solution to impose desired dynamics to uncertain systems but only recently extended to networks of dynamic systems; (2) implement the solutions in a Matlab/Simulink network-based platoon control simulator available at the University of surrey. This simulator allows operating both vehicle platoons composed either by CAVs with the same characteristics (homogenous platoon) or with different characteristics (heterogeneous platoons), as well as linear and nonlinear vehicle models and automatically scales to with the number of CAVs composing the platoon.
References:
1) S.E. Li et al, An Overview of Vehicular Platoon Control under the Four-Component Framework. In Proceedings of the IEEE Intelligent Vehicles Symposium (IV), 2015.
2) S. Baldi and P. Frasca, Adaptive synchronization of unknown heterogeneous agents: An adaptive virtual model reference approach. J Franklin Inst, 2019
Notes The thesis activity will likely require a visiting period at the University of Surrey, UK.
Deadline 17/09/2021
PROPONI LA TUA CANDIDATURA