Towards sustainable traffic control: evaluation of learning traffic controllers

Traffic control is a crucial component of urban transportation systems. In recent years, machine learning-based traffic controllers have been widely used in urban traffic control. This paper presents a comprehensive evaluation of the performance of traffic control systems on urban motorways and signalized intersection networks. By integrating sustainability criteria into the evaluation, this research contributes to the development of efficient and sustainable urban traffic control solutions. The proposed methodological evaluation framework is comprehensive and can be used as a roadmap for the implementation and testing of various ITS-based urban road transport solutions. In addition, future research directions are outlined that guide the evolution of sustainable and efficient traffic management solutions based on machine learning.