Back to Classics: Controlling Smart Thermostats with Natural Language… with Personalization
- 8 Downloads
Fuzzy Sets and Fuzzy Logic are classical tools for controlling devices, including thermostats. Fuzzy sets have been also used to describe Natural Language hedges that appear in every day speech. We use both of these classic applications for personalized thermal preferences of occupants with a goal of saving energy. As such, we combine previous knowledge on preferred comfort temperature range of groups of individuals with fuzzy hedges for temperature control setting, create fuzzy sets for energy consumption and saving, and use the intersection of the created sets with the preferred temperatures to optimize natural language interpretations of occupants’ commands on temperature settings of smart thermostats.
KeywordsNatural language Computing with words Fuzzy hedges Smart thermostat
This work has been partially supported by National Science Foundation under grant number 1737591.
- 3.Glöckner, I.: Fuzzy Quantifiers: A Computational Theory. Springer (2006)Google Scholar
- 7.Zadeh, L.A.: From computing with numbers to computing with words. From manipulation of measurements to manipulation of perceptions. IEEE Trans. Circuits Syst. I Fundam. Theory Appl. 46(1) (1999)Google Scholar
- 8.Zadeh, L.A.: Computing with Words: Principal Concepts and Ideas. Springer (2012)Google Scholar
- 9.Mendel, J.M.: Computing with words, when words can mean different things to different people. In: ICSC Congress on Computational Intelligence: Methods and Applications (1999)Google Scholar
- 10.Raskin, V., Taylor, J.: Computing with nouns and verbs. In: IEEE International Conference on Fuzzy Systems (2012)Google Scholar
- 11.Lee, S., Karava, P., Tzempelikos, A., Bilionis, I.: Inference of thermal preference profiles for personalized thermal environments with actual building occupants. Build. Environ. (2018)Google Scholar
- 12.Awalgaonkar, N., Bilionis, I., Liu, X., Karava, P., Tzempelikos, A.: Learning personalized thermal preferences via Bayesian active learning with unimodality constraints. http://arxiv.org/abs/1903.09094 (2019)