Jacques Saraydaryan

Characterization of a multi-user indoor positioning system based on low cost depth vision (Kinect) for monitoring human activity in a smart home

An increasing number of systems use indoor positioning for many scenarios such as asset tracking, health care, games, manufacturing, logistics, shopping, and security. Many technologies are available and the use of depth cameras is becoming more and more attractive as this kind of device becomes affordable and easy to handle. This paper contributes to the effort of creating an indoor positioning system based on low cost depth cameras (Kinect). A method is proposed to optimize the calibration of the depth cameras, to describe the multi-camera data fusion and to specify a global positioning projection to maintain the compatibility with outdoor positioning systems. The monitoring of the people trajectories at home is intended for the early detection of a shift in daily activities which highlights disabilities and loss of autonomy. This system is meant to improve homecare health management at home for a better end of life at a sustainable cost for the community.

Detection of collaborative activity with Kinect depth cameras

The health status of elderly subjects is highly correlated to their activities together with their social interactions. Thus, the long term monitoring in home of their health status, shall also address the analysis of collaborative activities. This paper proposes a preliminary approach of such a system which can detect the simultaneous presence of several subjects in a common area using Kinect depth cameras. Most areas in home being dedicated to specific tasks, the localization enables the classification of tasks, whether collaborative or not. A scenario of a 24 hours day shrunk into 24 minutes was used to validate our approach. It pointed out the need of artifacts removal to reach high specificity and good sensitivity.The health status of elderly subjects is highly correlated to their activities together with their social interactions. Thus, the long term monitoring in home of their health status, shall also address the analysis of collaborative activities. This paper proposes a preliminary approach of such a system which can detect the simultaneous presence of several subjects in a common area using Kinect depth cameras. Most areas in home being dedicated to specific tasks, the localization enables the classification of tasks, whether collaborative or not. A scenario of a 24 hours day shrunk into 24 minutes was used to validate our approach. It pointed out the need of artifacts removal to reach high specificity and good sensitivity.

A bio-inspired living lab as a robot exoskeleton

We propose a “bio-inspired” architecture of a living lab based on a robot operating system, in order to facilitate maintaining senior people autonomy at home. A highly personalized approach is possible in the personal home environment, which should adapt to its inhabitants along their life, providing a personalized robotic physical assistance together with a cognitive and mnesic support. The described architecture is completely modular, providing the opportunity to include more features, sensors and actuators in the living lab in the future. An example of indoor localization using depth cameras is described. It illustrates both the ability to use existing sensors with high interoperability and how the system can be extended. A scenario of activity treated by the system is also discussed.

Patrouille Multi-Agent Dynamique, application en Robotique au Service de Personnes Mobiles

Dans le contexte de la robotique de service, deployer une flotte de robots mobiles auservice d’humains qui evoluent dans de vastes espaces est un defi important. Nous nous interessons ici au probleme de servir regulierement et equitablement des personnes se deplacant librement avec un nombre limite de robots. Nous montrons qu’il s’agit d’un probleme de patrouille multi-agent dynamique, pour lequel nous formalisons les criteres de performance. Nous proposons et comparons quatre strategies de patrouille, deux etant fondees sur des solutions standards de la patrouille multi-agent, deux autres prennent en compte la mobilite et l’oisivete des personnes a servir. Nous presentons un simulateur combinant un modele de pietons (Ped-Sim) et un environnement robotique. Les strategies sont evaluees et comparees, en faisant varier les parametres du probleme, comme le nombre de robots, la topologie de l’environnement et la dispersion des individus dans l’environnement, montrant l’efficacite des nouvelles approches.