Abstract
The following chapter describes new functions and applications that the authors have developed for the capacitive sensor system SensFloor® during the last five years. Some of these features have already found their way into the market while others are still in the research phase. From the first large-scale installations in 2012 up to now, the focus was put on applications for healthcare and ambient assisted living (AAL). However, as will be described later in this chapter, the authors have realized projects in many other domains as well, such as medical assessments, retail, security, and multimedia. The chapter starts out with a description of the underlying technology. After that, examples of the various domains of application are presented. The authors conclude with a summary and future research plans.
TopIntroduction: A Sensor System Based On A Large-Area Capacitive Floor
SensFloor® is a large-area capacitive sensor floor, installable beneath all kind of flooring – invisible and discreet. Persons walking across the floor trigger signals, which are sent wirelessly to a transceiver. This system can calculate the number of persons on the floor, their direction and speed as well as detect falls. Several standard-interfaces are available for client-specific data analysis infrastructure. This sensor floor offers a variety of applications in health care, Ambient Assisted Living, home automation, security and multimedia. (Steinhage & Lauterbach, 2011). Although the result of these functions is obvious to the user, the sensor system itself remains invisible and does not interfere with the material or design of the floor covering in any way. In this respect, the technology presented here is an example of a new class of systems summarized under the expression Ambient Assisted Living (AAL). As this expression indicates, these systems’ main purpose is to provide a wide range of Smart Home comfort functions, as well as to assist handicapped people within their daily life (Lauterbach, Steinhage & Techmer, 2013; Lauterbach & Steinhage, 2012). An example is the recognition of falls and the following automatic activation of an alarm call. Depending on the installation site (nursing home, hospital, senior residence, private home) and the existing infrastructure, these alarms are either transmitted to the nurses’ ward and/or to their wireless phones, to care service providers or to the mobile phones of relatives or neighbours.
However, fall detection is just one of many possible functions in this domain. Based on the high spatial and temporal resolution of the sensor floor, the number and exact location of all people in the room can be recorded together with the velocity and direction of their movements. These data can be used to enhance peoples’ comfort and safety e.g. by automatically adapting heating, air condition and lighting to the presence and whereabouts of people. Experts can detect changes in the health status of a person by comparing typical movement patterns over a long period. On a short time scale, automatically recording, visualizing and analysing the gait patterns of patients can increase the accuracy and objectivity of physiological assessments of the health status.
Standard technologies for movement recognition and emergency detection are based on conventional infrared-, ultrasonic- or radar motion sensors. The acquisition of peoples’ exact location and movement tracking requires more advanced systems usually based on camera image processing or wireless identification tags. In addition to the technical problems brought about by varying lighting conditions, blind spots caused by furniture in the room and the still unsolved computational task of robustly detecting arbitrarily dressed persons in a video image, cameras installed in every room may interfere with the inhabitant’s desire for privacy. The latter does also hold for wireless identification tags as they allow for a labelled behavioural protocol of individuals. In addition, systems like these are not ambient as they require a visible installation.
In particular, in the case of elderly or handicapped persons knowing their current location or behavioural status is crucial. Many wearable sensor systems have been developed for this purpose already. There exist alarm buttons, for instance, which can be pressed in emergencies. This requires, however, that the person is conscious and still able to press the button. Automatic sensors, such as accelerometers, can detect specific situations like a fall, for instance. These entire wearable sensors require, however, that the users carry them at all times, even in the bathroom. This may become cumbersome for the user and in addition, other people might directly associate these devices with disabilities of their carriers. The adequate operation, charging and maintenance of these devices often overburdens the growing group of patients that suffer from diseases like dementia.
SensFloor® relies on a much more direct way of detection: a grid of sensors underneath the flooring detects local capacitive changes in the environment brought about by humans walking or standing on the floor. By design, this method does not allow for an identification of individuals. However, the persons’ location is acquired very accurately based on the spatial resolution of the sensor grid. By collecting and processing the sensor patterns over time, it is possible to assign movement trajectories to the persons, which allows for a variety of applications.
Key Terms in this Chapter
Movement Trajectory: Graphical representation of the walking path of people.
Redundancy: The fact that instances of important data elements or physical objects exist multiple times such that the loss of one element does not impair the overall function.
Capacitive Sensor: Sensor that measures an electrical property of matter even from a distance. Certain materials like metals, water, or body parts have a high capacitance.
Wireless Transmission: Cables are not required for exchanging sensor and configuration data.
Fall Detection: Recognizing a fall from characteristic sensor information.
Proximity Sensing: Measuring properties of objects from a distance.
Living Lab: Apartment equipped with sensors that serves as test environment for investigating the interaction between people and technology in realistic living situations.
Ambient Assisted Living (AAL): Research field for evaluating technology and services to allow elderly or people in need of care to stay longer in their private homes instead of moving to nursing homes.
Textile Underlay: Polyester fleece with built-in electronics modules for capacitance sensing underneath the floor covering.
Pressure Sensor: Device that measures a mechanical indentation caused by force or weight.
Internet of Things (IoT): Aim to connect distributed objects with the internet in order to collect or access information about the local environment of these objects.
Gait Pattern: Visual representation of a sequence of steps of a person.
Medical Assessment: Observation of the physical or behavioral characteristics of a person in order to classify the health status.