To support traffic authorities in the assessment of traffic signal strategies via simulation, we propose an approach that leverages on the strengths of automated planning knowledge models to generate accurate traffic simulators. By exploiting the sensors' readings of adaptive traffic control systems in operation in a region of interest, and the conciseness of planning knowledge models, the proposed approach can effectively simulate the impact that traffic signal strategies will have on the considered urban region. Our experimental analysis, performed using real-world historical data, shows that the accuracy of our simulated traffic conditions is within 10% of what was actually recorded by deployed sensors.
Leveraging Artificial Intelligence for Simulating Traffic Signal Strategies
Scala E.;Vallati M.
2022-01-01
Abstract
To support traffic authorities in the assessment of traffic signal strategies via simulation, we propose an approach that leverages on the strengths of automated planning knowledge models to generate accurate traffic simulators. By exploiting the sensors' readings of adaptive traffic control systems in operation in a region of interest, and the conciseness of planning knowledge models, the proposed approach can effectively simulate the impact that traffic signal strategies will have on the considered urban region. Our experimental analysis, performed using real-world historical data, shows that the accuracy of our simulated traffic conditions is within 10% of what was actually recorded by deployed sensors.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
