Laura Giarré

Seminar prof. Falcone

October 16, 2018

Avviso di seminario

Mercoledì 17 ottobre 2018, ore 14.30

Sala Riunioni. edificio 27, Dief

Control Invariance Tools for Control Design with Performance Guarantees with applications to highly automated transportation systems

Paolo Falcone

Department of Electrical Engineering, Chalmers

University of Technology, Gothenburg, Sweden

falcone@chalmers.se

Keywords

Vehicle lateral control, Multi-vehicle systems, Networked control systems, Joint communication and control design

Abstract

Control invariance is a property of dynamical systems state sets, which is widely used in control design for constrained systems. If a system state trajectory starts in a control invariant set C, it will be always possible to find an admissible control input keeping the system state trajectory within C. Hence, if the system constraints are defined w.r.t. to the required performance (e.g., maximum accepted bounds on the tracking errors), control invariant sets can be used to design control laws with performance guarantees. In this talk, we will highlight a few computational aspects related to control invariant sets and present recent results from their applications to a vehicle motion control problem and to the control of a multi-vehicle system over a wireless network. In Level-4+ autonomous vehicles, a human driver will not be available to back up the vehicle control system anymore. Hence, the overall autonomous driving system should be designed in order to provide performance and safety guarantees. Such stringent requirements can be translated into constraints in Model Predictive Control (MPC) control schemes. Moreover terminal control invariant set constraints can be used to enforce the persistent feasibility of the underlying optimization problem, at the cost of possible increased computational complexity. In this talk, we will show how low-complexity invariant sets can be derived thanks to newly developed methods and how these can be used as an additional tuning tool to trade off the MPC controller complexity for the size of its feasibility set. The next challenge, beyond high-level autonomous driving, is the cooperation of autonomous vehicles. Nevertheless, the safety and performance issues arising from the tight coupling between communication and control must be accounted for at the design stage. Starting from a multi-vehicle coordination problem at traffic junctions, we will motivate a joint communication and control paradigm, where a central coordinator decides upon control inputs to a set of dynamical systems and their access to the communication channel. We will show a few results from numerical examples and new research directions.

Bio. Paolo Falcone received his M.Sc. (“Laurea” degree) in 2003 from the University of Naples Federico II and his Ph.D. degree in Information Technology in 2007 from the University of Sannio, in Benevento, Italy. He is Bitradande Professor (Professor) at the Department of Electrical Engineering of the Chalmers University of Technology, Sweden. His research focuses on constrained optimal control applied to autonomous mobile systems, cooperative driving and intelligent vehicles. He is involved in several projects, in cooperation with the Swedish automotive and ICT industries, focusing on autonomous driving, cooperative driving and vehicle dynamics control.