IPADL: Indoor positioning for activity of daily living

Population ageing is causing health related cost explosion (United Nations 2010). A solution is to enable aging in place. But, to do so, we need to be able to provide punctual assistance at home to the inhabitant (Rashidi & Cook 2009). One of the way to do this, is to rely on technology and implement assistance from a computer system in what we call a smart home (Han & al. 2014). To do so, we have to be able to recognize the ongoing activity of daily living (ADL). The main difficulty in such a system is the lack of granularity in recognized activities (Chen & al. 2012). This is sometime due to imprecision in sensors or sometime due to a lack of available information. The main hypothesis of this project is that these difficulties could be addressed with a system based on objects tracking.

In the first months of this project, we created a prototype of indoor positioning system (IPS) using radio frequency identification (RFID) antennas and passive tags in the DOMUS laboratory, at Université de Sherbrooke. For this system, we divided the lab smart home into zones of varying dimensions and we trained several classifiers to recognise in what zone a tag was using only raw received signal strength indication (RSSI). The preliminary results are very encouraging and thus we plan on pushing the design a step further. We will carefully design a protocol for the installation and choice of RFID tags. Then, we aim at exploring the main filters used in signal processing in order to adapt and try them for our IPS. The system will in turn be tested extensively in a realistic context.

In a second experiment, we built a prototype of indoor tracking system (ITS) using this IPS, with a random forest as the classifier. Our goal was to test rapidly if our main hypothesis was holding within a simple setting. To evaluate the accuracy of the ITS, we designed 24 realistic paths that we tracked 10 times each. At this early stage, the ITS has an accuracy were 75% of crossed zone are identified, with about 50% of them in the right sequential order. The main issue seems to be the “teleportation” of the tracked objects. We have many ideas for filters that we will definitely try in the next months. The hypothesis is that a filter that could completely prevent “teleportation” would greatly increase accuracy.

In the next years, we plan on improving both the IPS and the ITS and then use them for activity recognition. We will put tags on almost everything we can use when performing activities of daily living. The first approach we want to try is to use the proximity of objects to infer activity steps. This technique has been shown to work with a trilateration based ITS (Fortin-Simard & al. 2012), but never with relative positioning (zones). To make our system independent from the smart home configuration, we will try to abstract the zones to use name places, such as kitchen sink, bathroom counter and so on. Some research on the best way to divide a room into section like this would be needed to see, for example, how the number of counter impacts the accuracy.

We believe the proximity based method will allow us to recognize many simple activities. The second step will be to aggregate those simple activities together to recognize complex activities. One challenge we will face will be to decide how to segment readings into activities, especially when the same object is used for many steps of the same complex activity. Given the size of our zones, the logical section of a room will cover more than one. This means we will have to deal with the case where an object does not leave a section for an activity, but does not always stay in the same underlying zone.

The full system, once completed, could be used by any monitoring system. In a cooking assistant, for example, it could keep track of the cooking steps to allow the assistant to provide contextualised and actual help to the user. In a cleaning assistant, it could keep track of clean zone. Even if it was not designed for that task, the ITS could also be used to monitor a robot inside a smart home.


  1. United Nations, World population ageing 2009: United Nations, Dept. of Economic and Social Affairs, Population Division, 2010.
  2. P. Rashidi and D. J. Cook. Keeping the resident in the loop: Adapting the smart home to the user. Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions on, 39(5):949–959, 2009.
  3. J. Han, C.-S. Choi, W.-K. Park, I. Lee, and S.-H. Kim, "Smart home energy management system including renewable energy based on ZigBee and PLC," Consumer Electronics, IEEE Transactions on, vol. 60, pp. 198-202, 2014.
  4. L. Chen, J. Hoey, C. D. Nugent, D. J. Cook, and Z. Yu. Sensor-based activity recognition. Systems, Man, and Cybernetics, Part C: Applications and Reviews, IEEE Transactions on, 42(6):790–808, 2012.
  5. D. Fortin-Simard, K. Bouchard, S. Gaboury, B. Bouchard, and A. Bouzouane, "Accurate Passive RFID Localization System for Smart Homes," presented at the 3th IEEE International Conference on Networked Embedded Systems for Every Application, Liverpool, UK, 2012

Members of the team

Principal investigators

Sylvain Giroux, Université de Sherbrooke

Sébastien Gaboury, Université du Québec à Chicoutimi

Kevin Bouchard, Université du Québec à Chicoutimi

MSc student

Frédéric Bergeron


Book chapters
  1. Bouchard K., Bergeron F., Giroux S.: Applying Data Mining in Smart Home, Assistive Technologies in Smart Environments, CRC press, Taylor & Francis, 2015.
Paper presented at a conference
  1. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Indoor Positioning System for Smart Homes based on Decision Trees and Passive RFID, Pacific Asia Knowledge Discovery and Data Mining Conference (PAKDD), 2016. (acceptance rate 17%)
  2. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Qualitative Tracking of Objects in a Smart Home, PErvasive Technologies Related to Assistive Environments (PETRA) conference, 2016.
Paper presented at a workshop
  1. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Simple Objects Tracking System for Smart Homes, IEEE International Conference on Bioinformatics and Biomedicine, 2015.
  2. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Qualitative RFID Tracking for ADL Recognition, AAAI-16 Workshop on artificial intelligence applied to assistive technologies and smart environments, 2016.
Poster presentation at a conference
  1. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Real-Time Constraints for Activities of Daily Living Recognition, PErvasive Technologies Related to Assistive Environments (PETRA) conference, 2016.
Oral presentation (non peer reviewed)
  1. Bergeron F, Bouchard K, Gaboury S, Giroux S, Bouchard B. Positionnement intérieur pour la reconnaissance d'activité de la vie quotidienne, Colloque interfacultaire des cycles supérieurs, 2016.