PhD Thesis Offer Title: Lagrangian Micro–Macro Control of Collective Dynamics

III. State of the art In the context of multi-agent systems, Lagrangian actuation bears resemblance to pinning control [8], in the sense that a few agents (nodes) are used to influence the macroscopic (network) dynamics through local interactions with their neighborhood. A key distinction is that since it typically entails moving actuator agents through the flow, Lagrangian actuation effectively causes the input distribution (network structure) to change. When it comes to highway traffic, while the singlelane case is fairly straightforward, macroscopically modeling the multi-lane multi-class case, e.g., through finding its mean-field limit [9], is considerably more challenging, but necessary in order to implement true Lagrangian control. One extensively studied approach is to model the actuator vehicles as moving bottlenecks, limiting the overtaking traffic flow at their positions [3]. However, this typically assumes that the actuator vehicle is acting in isolation, and collaborative Lagrangian actuation, where multiple agents act together to exert more control over the macroscopic state than they would be able to do alone, has only been partially addressed in the simplest case of multilane controlled platoons [10]. This project aims to formulate a more generic approach for building macroscopic models that are conducive to Lagrangian control design, and then exploit them for regulating the macroscopic collective dynamics, providing performance guarantees. IV. Required skills The candidate should have a strong background in engineering, applied mathematics, or a related field (Grandes Ecoles or Master), focusing on dynamical systems and control. Good written and oral communication skills in English are required. Experience with research, numerical simulation, and data-driven modeling will be considered an asset. Prior exposure to modeling and simulation of road traffic is welcome but not mandatory. The PhD can be preceded by a research internship at the lab. V. Application To apply, send an email with your CV to Mladen Čičić: mladen.cicic@centralesupelec.fr.