Our newest paper titled “Reinforcement Learning-Based Dynamic Zone Positions for Mixed Traffic Flow Variable Speed Limit Control with Congestion Detection” has been accepted for publication and published in the international scientific journal Machines, Special Issue “Optimization and AI of Autonomous Multi-Agents”. The paper is freely available in Open Access with DOI: https://doi.org/10.3390/machines11121058.
This paper was written in the co-authorship of prof. Edouard Ivanjko Ph.D. and assist. prof. Martin Gregurić Ph.D., and Ph.D. students Filip Vrbanić mag. ing. traff. and Mladen Miletić mag. ing. traff..
The paper presents the results of applying Q-learning for dynamic variable speed limit zone placement and variable speed limit control using the congestion detection on urban motorways in improving macroscopic traffic measures of effectiveness with a focus on mixed traffic flows that include human-driven vehicles and connected autonomous vehicles.