Crowds Flow Detection and Indoor Temperature Monitoring for Advanced Air-conditioning Control

In this paper, we propose a novel air-conditioning system that will be effective for wide public space such as a concourse of railway station, exhibition hall or museum. In these spaces, conventional air-conditioners are not enough, since the air-conditioner sends a wind to also empty area and where the temperature is already sufficient. Generally, it is difficult to ventilate effectively to the extensive coverage. On the other hand, our proposed system monitors people-flow and temperature distribution simultaneously, so that the system detects an optimum location to send a cool/warm wind, i.e. where the population density is relatively high and the temperature is insufficient. In order to track a number of pedestrians, we exploit multiple laser range scanners, since they are not only independent from lighting condition, but also can measure a range data directly with wider coverage. Additionally, we attempt to extract a crowds-flow or stream of confluences within lots of trajectories. This approach has seldom researched. While, by using wireless sensor network technology, a temperature distribution in the target space is calculated with IDW spatial interpolation. Finally, by integrating these two types of information, crowds-flows and temperature distribution, the proposed system detects a position where should be fan, and then, it controls the nozzle’s orientation on the real-time basis. The availability is tested by conducting an experiment at railway station in Japan. By applying the proposed system, a ventilator can aim at more persons for longer period to fan each person than conventional air-conditioning system.