Section: New Results

Assistance

We are still going on in building a framework for customizable and modular assistive robotics including hardware, software and communication and medical monitoring [12]. The development of our platforms shows that we are now able to identify problematic issues for end-users, helpers and the medical community and to propose appropriate hardware/software solutions. But the most time consuming part of our work is related to evaluation and therefore experimentation: this involves legal/ethical issues (for which we contribute [11]), participation of the medical community (for evaluation and recruitment) and heavy administrative management.

Rehabilitation in an immersive environment

Participants : Artem Melnyk, Jean-Pierre Merlet.

Rehabilitation is a tedious and painful process and it is difficult to assess its trend. Using an immersive environment has shown to increase the patient motivation but is not sufficient regarding rehabilitation efficiency. First the visual feedback (event 3D) is not sufficient to provide a full immersive feeling as body motion is not involved. Controling body motion is also very important for therapists that currently must continuously correct the patient pose so that the rehabilitation exercise is the most efficient. We propose to add motion generators in the environment to reinforce realism (thereby increasing patient motivation) but also to allow therapists to use these generators to control the body pose so that they will be able to repeat rehabilitation exercises in a controlled context. Furthermore these generators are instrumented to provide information on the body pose and additional external sensors complete these measurements for rehabilitation assessment. We have developed 3 types of motions generators: one 6 d.o.f. motion base, a CDPR that is able to lift a patient and 2 multipurpose lifting columns. We are in the process of mixing visual feedback with these generators but unfortunately Inria-Sophia immersive room that we were planning to use is no more available. Hence we have been obliged to develop a renderer and this has slowed down the integration process. Still we have currently a preliminary experiment going on: our columns are able to modify the slope and inclination of a treadmill so that we can simulate a walk in the mountain while controlling the inclination to enforce the use of one specific leg. External and wearable sensors allow us to monitor the walking pattern and provide synthetic indicators on this pattern for the therapist.

Smart Environment for Human Behaviour Recognition

Participants : Mohamed Hedi Amri, Alain Coulbois, Aurélien Massein, Artem Melnyk, Jean-Pierre Merlet, Yves Papegay, Odile Pourtallier, Valérie Roy, Eric Wajnberg.

The general aim of this research activity focuses on long term indoor monitoring of frail persons. In particular we are interested in early detection of daily routine and activity modifications. These modifications may indicate health condition alteration of the person and may require further medical or family care. Note that our work does not aim at detecting brutal modifications such as faintness or fall.

In our research we envisage both individual and collective housing such as rehabilitation center or retirement home.

Our work relies on the following leading ideas :

• We do not base our monitoring system on wearable devices since it appears that they may not be well accepted and worn regularly,

• Privacy advocates adequacy between the monitoring level needed by a person and the detail level of the data collected. We therefore strive to design a system fitted to the need of monitoring of the person.

• In addition to privacy concern, intrusive feature of video led us not to use it.

The main aspect that grounds this work is the ability to locate a person or a group in their indoor environment. We focus our attention to the case where several persons are present in the environment. As a matter of fact the single person case is less difficult.

A natural way of being able to adapt the accuracy of localization (and consequently accuracy of monitoring), is to use a partition of the monitoring area in a finite number of elementary zones ; the number of zones together with their geometry being closely related with the pursued level of monitoring. In practice these zones will be materialized by sensors barriers that detect the passage of a person from one zone to another. Henceforth each zone are polygonal.

We are following several directions :

• monitoring system design,

• material development,

• data gathering and analysis,

• experimentation.

Monitoring system design

Monitoring systems provide information that can range from a broad measure of the daily activity to a precise analysis of the ability of a person performing a task (cooking, dressing, ...) and its evolution.

The broad range of needs and contexts, together with the large variety of available sensors implies the necessity to carefully think the design of the monitoring system. An appropriate system should be inexpensive and forgettable for the monitored person, should respect privacy but collect necessary data, and should easily adapt to stick to new needs. We aim to provide an assisting tool for designing appropriate monitoring systems.

There is an ongoing PhD work about to be defended : metrics have been defined to evaluate quality of sensors solutions and placement to infer people behaviors inside a smart environments, and a methodology for optimal design of smart environments has been developed.

Material development, signal processing and data fusion

Based on the experimentation we initiated in 2016 in Ehpad Valrose and this year at the Institut Claude Pompidou in Nice, we designed, developed and tested a new class of multi-sensors barriers to overcome difficulties arising from wider environment, reflection properties of walls and light exposures.

These multi-sensor barriers contains a selection of infra red distance sensors and motion sensors of passive infrared type. Dedicated signal processing, and fusion of the different sets of signals provide information on crossing time, direction of crossing, speed and size of crossing person, object, or group. This last information is helpful to differentiate for example a person using a wheelchair, a valid person (e.g medical staff), or an elderly.

Data analysis for activity recognition

Data are issued from long-term recording during on-going experimentation and from simulation tools developed on purpose from probabilistic models.

At Ehpad Valrose, a monitoring system is installed since last year in the first floor. Area of monitoring is restricted to the hallway that leads to the individual rooms of six residents. Residents are proposed several activities (social or cultural activities, physical activities, meals) and have to use the hallway when participating to those activities. Hence few people are entering the experimentation area, with significant characteristics : elderly with rolling chairs or rollators, autonomous elderly, technical staff with trolleys, visitors, caregivers. On this experiment, we focus our analysis on inferring from data, quantitative indicators on the activity of each resident. Data analysis generates several possible scenarii of activities. They are discriminated, based on distinction of people when crossing barriers and on additional knowledge on their habits.

Experimentation

A new monitoring system has been installed in Institut Claude Pompidou in Nice. Area of monitoring is an half floor dedicated to a consultations and day care unit of the Centre Mémoire de Ressources et de Recherche, together with a clinical research unit of the Cobtek team. It consists in 15 rooms articulated on a simple network of hallways, including offices, training rooms, consultation rooms, waiting area, toilets, resting room for staff, delimited by an entrance and a communication to the research center. Patients, service staff, clinical staff, researcher are using the monitored area. The aim of this experiment is to determine identify the different flows of people between different zones to qualify uses of the facilities and interaction between the different groups of users.

Twenty barriers have been installed for a total of 77 sensors and kilometers of cable deployed. Installation and set-up of the system led to several electrical, electronical and technical problems, efficiently solved by the engineers of our group that played a crucial role.