Research Front 1 aims at designing interactive and cooperative learning algorithms, for home inhabitants to confront to the IHEMAS in order to yield knowledge about occupant activities and costs/comforts preferred compromise. This project extends the promising concepts opened up by the ANR INVOLVED project regarding interactions by developing cooperative solutions to learn a global human-system learnt representation. The aim is to identify the practices as well as the activities of the occupants by reconciling the perceptions of the IHEMAS and its more or less numerous sensors with the perceptions of the inhabitants. These perceptions will be translated into activity labels, intentions and preferences, taking into account the volatility of the inhabitants’ memory and their limited consent to interact with an IHEMAS. It induces learning methods with ad hoc notifications {tooltip}1{end-texte}M. Amayri, A. Arora, S. Ploix, S. Bandhyopadyay, Q.-D. Ngo, and V. R. Badarla. Estimating occupancy in heterogeneous sensor environment. Energy and Buildings, 129:5 6–58, 2016.{end-tooltip} {tooltip}2{end-texte}M. Amayri, S. Ploix, N. Bouguila, and F. Wurtz. Estimating occupancy using interactive learning with a sensor environment: Real- time experiments. IEEE Access, 7:53932–53944, 2019.{end-tooltip} but also mechanisms for matching the inhabitants and IHEMAS perceptions. The number of sensors and its related generated features is more or less important (for instance, in some experiences in the EXPESIGNO Rhône-Alpes region project, only one Linky power-meter per household has been used but load recognition/forecasting strategies can be used {tooltip}3{end-texte}B. Farsi, M. Amayri, N. Bouguila, U. Eicker. On Short-Term Load Forecasting Using Machine Learning Techniques and a Novel Parallel Deep LSTM-CNN Approach. IEEE Access, IEEE, 2021, 9, pp.31191 – 31212.{end-tooltip}). It may induce confusions between the IHEMAS and inhabitants perceptions that must be resolved when the system is not able to discriminate labels provided by the inhabitants. These confusions must be resolved by automatically adapting for instance the generated features. Combining sensor data with labels from occupants yield a model thanks to learning algorithms. Interactive learning is a complementary method to discover the energy behavior of a site. Contrary to the INVOLVED approach, explanations and advice are generated without an a priori physical model, but by exploiting similar encountered situations. The aim is to conceive an exploratory approach guiding the inhabitants in the discovery of the effects of actions in similar situations. Inhabitants will thus be put in situation of experimenters of their environment and the IHEMAS will have the role of recording the experiments and guiding inhabitants towards new exploratory. This research front will lead to different prototypes delivered to the research front 2 for the design of a relevant Human System Interface (HSI).
1.. M. Amayri, A. Arora, S. Ploix, S. Bandhyopadyay, Q.-D. Ngo, and V. R. Badarla. Estimating occupancy in heterogeneous sensor environment. Energy and Buildings, 129:5 6–58, 2016.↩
2.. M. Amayri, S. Ploix, N. Bouguila, and F. Wurtz. Estimating occupancy using interactive learning with a sensor environment: Real- time experiments. IEEE Access, 7:53932–53944, 2019. ↩
3.. B. Farsi, M. Amayri, N. Bouguila, U. Eicker. On Short-Term Load Forecasting Using Machine Learning Techniques and a Novel Parallel Deep LSTM-CNN Approach. IEEE Access, IEEE, 2021, 9, pp.31191 – 31212.↩