Reduce building energy consumption by 20-40% with self-learning predictive control

viboo offers self-learning predictive control on the basis of physics-informed Machine Learning as a cloud service to smart-thermostat and building automation companies.

20-40% less energy and improved thermal comfort

compared to state of industry controllers.

1-2 weeks of training time

compared to multiple months with pure Machine Learning methods.

Ready for demand response

due to computationally efficient predictive control algorithms.

Self-learning control...

We use physics-informed Machine Learning to generate computationally efficient models for predictive control in buildings.  Room temperature and valve position data is measured for 1-2 weeks and stored in a database. Based on this data, a prediction model of the thermal behavior of the building is generated. This model is used in predictive control to optimize the energy input under the consideration of user preferences and the weather forecast.

This approach improves comfort and saves energy. We are also ready for demand response and variable energy pricing. a cloud service

Your company offers smart thermostats for residential buildings or building automation solutions for light commercial buildings. We provide our control algorithms to you via a cloud service through an intuitive API: Predictive Control as a Service (PCaaS).

Together, we bring self-learning predictive control to buildings that cannot benefit from efficient control solutions so far.

About Us

viboo AG is a spin-off company by researchers of the Urban Energy Systems Lab at Empa and the Automatic Control Lab at ETH Zürich, Switzerland. We are experienced entrepreneurs in the building energy domain and have worked in world-leading labs in the areas of building energy systems, automatic control, and machine learning in Aachen, Berkeley and Zürich.

Our partners