Smart Lighting for City Services (SLCS) is an activity in the Smart Spaces Action Line of the EIT ICT Labs.
Due to the LEDification trend, many new luminaire manufacturers entered the outdoor lighting market (Samsung, OSRAM) resulting in aggressive pricing. However, the vast majority is providing solutions for functional lighting, a few offer sensor-driven light control while none target the integration of sensors for city monitoring. Currently a lot of different companies are on the market providing sub-systems like environmental monitoring stations (La Crosse, Jacob Jensen) or information for city services through data analytics (Google, TomTom and IBM). However, such deployments are costly and scarce resulting in limited spatial coverage and providing information on a very rough scale. In such, the competition within the smart city ecosystem remains fragmented, without the availability of an end-to-end solution. The EIT ICTLabs activity “Smart Lighting for City Services” will differentiate by providing a complete smart lighting infrastructure, from functional lighting to new digital service propositions for smart cities. Its main goals of the SLCS project are to:
TU/e has been investigating the use of data analysis and machine learning for smart distributed indoor lighting solutions. Within SLCS TU/e makes use of this experience to adapt and apply machine learning algorithms in the context of smart city services. In particular, it focuses on designing, implementing and evaluating a Smart City Sensing Platform for gathering, storing and analyzing large amounts of data streaming from sensors deployed around the city.
The selected use case for this work is predicting the number of people in Stratumseind, a bar district in Eindhoven. Several stake holders are interested in being able to anticipate the number of visitors in that area, including the bar owners, police, hospitals and cleaning services, for plannning their resources and personnel. The TU/e will use several data sources (e.g. cameras counting people entering and leaving the area, weather, special events) to make the predictions.