The present research objective is to develop a human activity recognition (HAR) system by exploiting in-memory computing (IMC) and machine learning technologies. However, while most of the developed HAR systems have been focused on the identification of macro activities, little attention has been addressed to systems with a narrower scope. After analyzing market potential and challenges, the development of a sitting posture recognition system emerged as the most promising route. From a literature review study, passive solutions (office chairs with ergonomic design) were proved to be scarcely effective, active solutions (posture identification through sensors) have not been developed to define a scalable product. In order to address such research gaps an integrated system will be created, composed by a smart office chair and a smartphone application providing warnings to the user. The HAR system will be integrated with the office chair, by using a sensors matrix integrated on the seat. This innovative product could lead to drastically improve the effectiveness of office chairs in maintaining a correct position as well as reducing the impact of backaches on users’ health condition.
PRINCIPAL ACADEMIC TUTOR
Prof. Daniele Ielmini, Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
OTHER ACADEMIC TUTOR
Prof. Andrea Aliverti, Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano
Prof. Fabrizio Bonani, Dipartimento di Elettronica e Telecomunicazioni, Politecnico di Torino
EXTERNAL INSTITUTIONS
STMicroelectronics, IMM-CNR
TEAM MEMBERS
Irene Chiocchia, Management Engineering, PoliMi
Riccardo Galvani, Biomedical Engineering, PoliMi
Kevin Montano, Nanotechnology Engineering, PoliTo
Marcello Neri, Electronics Engineering, PoliTo
Andrea Scotti, Computer Engineering, PoliMi
Gian Luca Dolso, Engineering Physics, PoliMi