Sensor Event and Activity Prediction using Binary Sensors in Real Homes with Older Adults
Metadata
Show metadataAppears in the following Collection
- Institutt for informatikk [4747]
Abstract
A fair amount of research on smart home functions has aimed at assisting older adults in their everyday life. The implementation of these functions relies on the performance of activity recognition and prediction algorithms. The research literature contains a number of well-performing algorithms for activity recognition and prediction. However, most of this work uses data collected in controlled environments (e.g. lab environments) using scripted activities. This thesis applied, evaluated, and compared the performance of state-of-the-art prediction algorithms in real-world scenarios. The project conducted a field trial including eight smart homes with one resident each, located in a care facility in Oslo for older adults over 65 years old. The homes have about 15 sensors, including motion, magnetic, and power sensors. Probabilistic methods and neural networks were applied to predict the next sensor event and activity of daily living in the home, as well as its time of occurrence. The work shows that it is possible to achieve acceptable prediction accuracy with few sensors and about three weeks of collected data. Transfer learning between apartments has also shown to be well performing. In addition, the study can be useful for deciding which prediction methods to use depending on the case, and in accordance with project constraints (e.g. the number of sensors, user privacy).List of papers
Paper I: Casagrande, F.D. and Zouganeli, E. ‘Occupancy and Daily Activity Event Modelling in Smart Homes for Older Adults with Mild Cognitive Impairment or Dementia’. In: Proceedings of The 59th Conference on Simulation and Modelling (SIMS 59) 153 (2018), pp. 236–242. DOI: 10.3384/ecp18153236. The article is included in the thesis. Also available at: https://doi.org/10.3384/ecp18153236 |
Paper II: Casagrande, F.D. , Tørresen, J. and Zouganeli, E. ‘Sensor Event Prediction using Recurrent Neural Network in Smart Homes for Older Adults’. In: 2018 IEEE International Conference on Intelligent Systems (IS) (2019), pp. 662–668. DOI: 10.1109/IS.2018.8710467. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1109/IS.2018.8710467 |
Paper III: Casagrande, F.D. , Tørresen, J. and Zouganeli, E. ‘Comparison of Probabilistic Models and Neural Networks on Prediction of Home Sensor Events’. Published at Proceedings of the 2019 IEEE International Joint Conference on Neural Networks (IJCNN). DOI: 10.1109/IJCNN.2019.8851746. The article is included in the thesis. Also available in DUO: http://urn.nb.no/URN:NBN:no-76722 |
Paper IV: Casagrande, F.D. , Tørresen, J. and Zouganeli, E. ‘Prediction of the Next Sensor Event and its Time of Occurrence in Smart Homes’. In: Proceedings of 28th International Conference on Artificial Neural Networks – Springer-Verlag Lecture Notes in Computer Science (LNCS), (2019), pp. 462–242. DOI: 10.1007/978-3-030-30490-4_37. The article is not available in DUO due to publisher restrictions. The published version is available at: https://doi.org/10.1007/978-3-030-30490-4_37 |
Paper V: Casagrande, F.D. and Zouganeli, E. ‘Prediction of Next Sensor Event and its Time of Occurrence using Transfer Learning across Homes’. Published at Proceedings of the ACS/IEEE International Conference on Computer Systems and Applications (AICCSA), 2019. DOI: 10.1109/AICCSA47632.2019.9035327. The article is included in the thesis. Also available at: https://doi.org/10.1109/AICCSA47632.2019.9035327 |
Paper VI: Casagrande, F.D. , Tørresen, J. and Zouganeli, E. ‘Predicting Sensor Events, Activities, and Time of Occurrence Using Binary Sensor Data from Homes with Older Adults’. In: IEEE Access Journal (2019), pp. 111012–111029. DOI: 10.1109/ACCESS.2019.2933994. The article is included in the thesis. Also available in DUO: http://urn.nb.no/URN:NBN:no-78248 |