Situation- and Context-aware Activity Recognition System (SACAAR)
This project proposes, develops and validates a context driven activity theory (CDAT) and a reasoning approach for recognition of complex activities of daily living (ADL). Our proposed Situation- and Context-aware Activity Recognition System (SACAAR) recognizes complex activities which are concurrent and interleaved. SACAAR and CDAT are validated using extensive experimention on real users. The novel complex activity recognition (CAR) algorithm recognizes complex activities in different scenarios and achieves an accuracy of 88.5% for concurrent and interleaved activities. The inferencing of complex activities is performed online and mapped onto situations in near real-time mode.
The architecture of our system is shown below:
More details can be found in this paper:
Saguna, A. Zaslavsky, D. Chakraborty, “Complex Activity Recognition using Context Driven Activity Theory in Home Environments”, in Proceedings of Smart Spaces and Next Generation Wired/Wireless Networking (ruSMART 2011), St Petersburg, Russia, 2011. (pdf)