A LANE OCCUPATION MODEL FOR A MOTORWAY SECTION USING ARTIFICIAL NEURAL NETWORKS
Lorenzo Mussone, G Reitani, Savino Rinelli · Virtual Community of Pathological Anatomy (University of Castilla La Mancha) · 1970
The paper analyses lane occupation in a three lane motorway. Using data collected on a Northern Italian motorway over a period of about nine months and treating them in appropriately the distribution of flow over the three lanes is studied. This distribution depends on the three fundamental variables, speed, flow and density, but also on composition of flow (percentage of heavy vehicles) and environment (brightness). Through the neural network approach it is possible to reconstruct the function which relates these variables to lane occupation.