© The Institution of Engineering and Technology
Many applications, such as smart buildings, crowd flow, action recognition, and assisted living, rely on occupancy information. Although the use of smart cameras and computer vision can assist with these tasks and provide accurate occupancy information, it can be cost prohibitive, invasive, and difficult to scale or generalise to different environments. An alternative solution should bring similar accuracy while minimising the listed problems. This work demonstrates that a scalable wireless sensor network with CO2-based estimation is a viable alternative. To support many applications, a solution must be transferable and must handle not knowing the physical system model; instead, it must learn to model CO2 dynamics. This work presents a viable prototype and uses the captured data to train machine learning-based occupancy estimation systems. Models are trained under varying conditions to assess the consequences of design decisions on performance. Four different learning models were compared: gradient boosting, k-nearest neighbours (KNN), linear discriminant analysis, and random forests. With sufficient labelled data, the KNN model produced peak results with a root-mean-square error value of 1.021.
References
-
-
1)
-
15. Ebadat, A., Bottegal, G., Varagnolo, D., et al: ‘Regularized deconvolution-based approaches for estimating room occupancies’, IEEE Trans. Autom. Sci. Eng., 2015, 12, pp. 1157–1168.
-
2)
-
6. Fahad, L., Rajarajan, M.: ‘Anomalies detection in smart-home activities’. IEEE 14th Int. Conf. on Machine Learning and Applications (ICMLA), Miami, Florida, December 2015, pp. 419–422.
-
3)
-
11. Gadyszewska-Fiedoruk, K.: ‘Correlations of air humidity and carbon dioxide concentration in the kindergarten’, Energy Build., 2013, 62, pp. 45–50.
-
4)
-
16. Jiang, C., Masood, M., Soh, Y., et al: ‘Indoor occupancy estimation from carbon dioxide concentration’, Energy Build., 2016, 131, pp. 132–141.
-
5)
-
17. Han, H., Jang, K., Han, C., et al: ‘Occupancy estimation based on CO2 concentration using dynamic neural network model’. Proc. of the 34th AIVC–3rd TightVent–2nd Cool Roofs’–1st venticool Conf., Athens, Greece, September 2013.
-
6)
-
22. Nathanson, T.: , 1993.
-
7)
-
3. Biton, R., Katz, G., Shabtai, A.: ‘Sensor-based approach for predicting departure time of smartphone users’. 2nd ACM Int. Conf. on Mobile Software Engineering and Systems, Florence, Italy, May 2015, pp. 146–147.
-
8)
-
10. Spachos, P., Hatzinakos, D.: ‘Real-Time indoor carbon dioxide monitoring through cognitive wireless sensor networks’, IEEE Sens. J., 2016, 16, pp. 506–514.
-
9)
-
12. Imanishi, T., Tennekoon, R., Palensky, P., et al: ‘Enhanced building thermal model by using CO2 based occupancy data’. IECON 2015–41st Annual Conf. of the IEEE Industrial Electronics Society, Yokohama, Japan, November 2015, pp. 003116–003121.
-
10)
-
2. Zhu, Q., Chen, Z., Soh, Y.: ‘Smartphone-based human activity recognition in buildings using locality-constrained linear coding’. IEEE 10th Conf. on Industrial Electronics and Applications (ICIEA), Auckland, New Zealand, June 2015, pp. 214–219.
-
11)
-
12)
-
26. Sangogboye, F., Kjaergaard, M.: ‘PROMT: predicting occupancy presence in multiple resolution with time-shift agnostic classification’, Comput. Sci. Res. Dev., 2017, 33, pp. 105–115.
-
13)
-
14)
-
25. Beltran, A., Erickson, V., Cerpa, A.: ‘Thermosense: occupancy thermal based sensing for HVAC control’. Proc. of the 5th ACM Workshop on Embedded Systems For Energy-Efficient Buildings, Roma, Italy, November 2013, pp. 1–8.
-
15)
-
19. Zuraimi, M., Pantazaras, A., Chaturvedi, K., et al: ‘Predicting occupancy counts using physical and statistical CO2-based modelling methodologies’, Build. Environ., 2017, 123, pp. 517–528.
-
16)
-
4. Manley, E., Deogun, J.: ‘Location learning for smart homes’. 21st Int. Conf. on Advanced Information Networking and Applications Workshops AINAW ‘07, Niagara Falls, Canada, May 2007, pp. 787–792.
-
17)
-
7. Schweizer, D., Zehnder, M., Wache, H., et al: ‘Using consumer behaviour data to reduce energy consumption in smart homes: applying machine learning to save energy without lowering comfort of inhabitants’. IEEE 14th Int. Conf. on Machine Learning and Applications (ICMLA), Miami, Florida, December 2015, pp. 1123–1129.
-
18)
-
9. Candanedo, L., Feldheim, V.: ‘Accurate occupancy detection of an office room from light, temperature, humidity and CO2 measurements using statistical learning models’, Energy Build., 2016, 112, pp. 28–39.
-
19)
-
18. Alam, A.G., Rahman, H., Kim, J.K., et al: ‘Uncertainties in neural network model based on carbon dioxide concentration for occupancy estimation’, J. Mech. Sci. Technol., 2017, 31, pp. 2573–2580.
-
20)
-
8. Lu, J., Sookoor, T., Srinivasan, V., et al: ‘The smart thermostat: using occupancy sensors to save energy in homes’. Proc. of the 8th ACM Conf. on Embedded Networked Sensor Systems (SenSys ‘10), Zurich, Switzerland, November 2010, pp. 211–224.
-
21)
-
13. Oldewurtel, F., Sturzenegger, D., Morari, M.: ‘Importance of occupancy information for building climate control’, Appl. Energy, 2013, 101, pp. 521–532.
-
22)
-
23)
-
1. Zanella, A., Bui, N., Castellani, A., et al: ‘Internet of things for smart cities’, IEEE Internet Things J., 2014, 1, pp. 22–32.
-
24)
-
5. Chen, L., Nugent, C., Wang, H.: ‘A knowledge-driven approach to activity recognition in smart homes’, IEEE Trans. Knowl. Data Eng., 2012, 24, pp. 961–974.
-
25)
-
14. Weekly, K., Bekiaris-Liberis, N., Jin, M., et al: ‘Modelling and estimation of the humans’ effect on the CO2 dynamics inside a conference room’, IEEE Trans. Control Syst. Technol., 2014, 23, pp. 1770–1781.
-
26)
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-wss.2018.5027
Related content
content/journals/10.1049/iet-wss.2018.5027
pub_keyword,iet_inspecKeyword,pub_concept
6
6