Transferability and Data-Efficient Learning for Perimeter Control in Urban Traffic Networks

The Transport and Traffic Engineering Laboratory (LICIT-ECO7) is a Joint Research Unit which is placed under the supervision of University Gustave Eiffel and ENTPE (Post-graduate School of Transport and Civil Engineering).

LICIT-ECO7 is also part of the University de Lyon and Lyon Urban School.

The laboratory is located on both the ENTPE Campus and the University Gustave Eiffel Bron Campus in Lyon.

LICIT-ECO7 is part of the JRT GEST (Joint Research Team on Energy Management and Storage for Transport).

LICIT-ECO7’s main research theme is the Dynamic Modelling, Monitoring and Control of Mobility Networks as well as New Vehicles and New Components Usage, Ageing and Control.

Perimeter Control (PC) is a well-established strategy for managing traffic in urban networks under oversaturated conditions, regulating vehicle flows into and out of a Protected Network (PN) using the Macroscopic Fundamental Diagram (MFD). Many existing studies implement efficient control strategies, but these are typically optimized for specific regions and conditions While deep reinforcement learning (DRL) has demonstrated its efficiency in managing traffic signals and optimizing perimeter control strategies, these data-driven approaches can face significant challenges in real-life applications. These challenges include dealing with unexpected perturbations, such as accidents or sudden traffic increases, and ensuring robustness when deployed in unseen regions where the model has not been trained.

This internship will not focus on optimizing new PC strategies but on the transferability of existing strategies across different urban areas. The core question is: How can we effectively learn from data in one region and successfully apply those learned strategies to another region with different traffic conditions, including disturbances such as accidents or surges in traffic?

The primary objective of this internship is to explore recent deep learning transfer techniques such as transfer learning, few-shot learning, and knowledge distillation to enable efficient transfer of knowledge across different regions with minimal data and retraining.

The ideal candidate should preferably have:

•  Experience with Deep Learning and familiarity with concepts like reinforcement learning, Graph neural networks and attention mechanism.
• Proficiency in Python programming and experience with deep learning libraries such as PyTorch.
• Interest in urban mobility and traffic control systems.