Quiet Route Planning for Pedestrians in Traffic Noise Polluted Environments

Noise pollution is one of the main stressors in urban environments, having negative impacts on people’s quality of life and health. For some groups of citizens, such as school children, patients, and elders, there is a need to support them in finding pedestrian routes in noise polluted areas of cities. In this paper, we focus on the estimation of traffic noise, and present an approach to provide quiet routing services, taking into account the estimated noise levels of roads. By combining Volunteered Geographic Information, official socio-economic data, and open-access GPS trajectory data, we develop a set of traffic related variables, and apply machine learning methods to perform traffic volume estimations. Given the estimated traffic information, an existing traffic noise model is then employed to derive the noise polluted areas. For generation of quiet routes, a new routing algorithm is proposed based on the Dijkstra algorithm. It minimizes the exposure of pedestrians to traffic noise pollution while taking into account the route distance constraint. We apply our quiet routing approach to the city of Heidelberg (Germany). The application results demonstrate the efficacy of our algorithms in the generation of quiet routes customized to pedestrian preferences.