Coupling Multi-agent and Macroscopic Simulators of Traffic

Abstract

Traffic simulations exist in multiple scales and each of these scales presents some advantages and is useful in certain contexts. Usually, multi-agent simulations use more detailed models and give more precise results than macroscopic models but their high calculation cost does not allow them to simulate very big areas such as an entire region. To overcome these limitations, multi-scale simulations emerged with the coupling of two or more simulations of different scales. This paper presents a generic solution to combine a macroscopic simulator working on a large area which contains a smaller area simulated by a multi-agent simulator. The main challenge is to assure the coherence between both simulators on the smallest area since it is simulated by both simulators at the same time. We first highlight the issues to tackle and the problems to solve when coupling two existing simulators then we propose solutions for the coupling, and finally evaluate them on an example scenario.

Strong coupling between scales in a multi-scalar model of urban dynamics

“…Multi-scale models have also been used for the simulation of crowd dynamics [Crociani et al, 2016]. The study of traffic is also made more accurate by coupling macroscopic and microscopic models [Boulet et al, 2020]. The management of ecosystems requires integrating across actors and scales [Belem and Müller, 2013].…”