Estimación de la densidad vehicular en una intersección para semáforos inteligentes con análisis de firma sonora

KRUJOSKI, MATÍAS GABRIEL; MOYA, SERGIO EDUARDO; J. E. KOLODZIEJ

San Miguel de Tucumán

Congreso; ARGENCON 2018; 2018

In this work is developed and evaluated in a simulation environment an algorithm for estimating the amount of cars awaiting the green light of a traffic light. The goal of the proposed system is to control the green light duration in an intelligent way to optimize the vehicular flow.For this purpose, it is evaluated a detection technique through an array of microphones distributed at the intersections of the streets, based on the acoustic signature of the combustion engines. There are exposed some alternatives for optimal spatial configuration of the microphones and filtering techniques of undesirable sounds present in the context of the final implementation. The system includes first an adaptive filter that is able to differentiate the engines in steady state (idle) from those in dynamic regime, and then proceeds to estimate the number of vehicles waiting for the green light. The simulation results presented indicates that with correct selection of the parameters of adaptive filter and the estimation algorithm, is possible to achieve an estimation with acceptable error levels for the intended application, even in the context of environmental noise in the public road.