Toward coupling sensor data and volunteered geographic information (VGI) with agent-based transport simulation in the context of smart cities

Abstract

Smart transportation planning is one of the fundamental components of Smart Cities. Agent-based transport simulations are one efficient way to support the development of smart transportation planning systems. However, to be reliable, transportation simulations must integrate information on time-varying phenomena that can constitute obstacles to transportation, such as the impact of snow storms. Sensor networks constitute an efficient solution for gathering data on such time-varying obstacles and feeding the agent-based transportation simulation. We present Sensor-Enabled OSM-Matsim, a multi-agent transportation simulation system which integrates sensor data from the Geospatial Cyberinfrastructure for Environment Sensing (GeoCENS) platform and uses volunteered geographic information (VGI) from Open Street Map (OSM) as input data. The Sensor-Enabled OSM-Matsim is based on MatSim, which provides a toolbox to implement large-scale agent-based transportation simulations. The sensor-enabled OSM-MatSim simulation system identifies obstacles from sensor data gathered through the GeoCENS platform and integrates these obstacles into the transportation simulation for a more realistic simulation. The proposed simulation system innovates by proposing solutions to overcome issues related to the interoperability of sensor services with simulations. This simulation is applied in the context of a snow storm, where thickness of snow monitored by sensors from GeoCENS is integrated into the simulation to facilitate transportation planning in the region of Calgary. An application scenario is presented showing how the simulation can improve smart transportation in the context of smart cities.