Research on sensor-based activity recognition has, recently, made significant progress and is attracting growing attention in a number of disciplines and application domains. However, there is a lack of high-level overview on this topic that can inform related communities of the research state of the art. In this paper, we present a comprehensive survey to examine the development and current status of various aspects of sensor-based activity recognition. We first discuss the general rationale and distinctions of vision-based and sensor-based activity recognition. Then, we review the major approaches and methods associated with sensor-based activity monitoring, modeling, and recognition from which strengths and weaknesses of those approaches are highlighted. We make a primary distinction in this paper between data-driven and knowledge-driven approaches, and use this distinction to structure our survey. We also discuss some promising directions for future research.

Sensor-Based Activity Recognition

Gait Analysis. An Introduction

The classical Poisson-Boltzmann (PB) theory of electrolytes assumes a dilute solution of point charges with mean-field electrostatic forces. Even for very dilute solutions, however, it predicts absurdly large ion concentrations (exceeding close packing) for surface potentials of only a few tenths of a volt, which are often exceeded, e.g. in microfluidic pumps and electrochemical sensors. Since the 1950s, several modifications of the PB equation have been proposed to account for the finite size of ions in equilibrium, but in this two-part series, we consider steric effects on diffuse charge dynamics (in the absence of electro-osmotic flow). In this first part, we review the literature and analyze two simple models for the charging of a thin double layer, which must form a condensed layer of close-packed ions near the surface at high voltage. A surprising prediction is that the differential capacitance typically varies non-monotonically with the applied voltage, and thus so does the response time of an electrolytic system. In PB theory, the capacitance blows up exponentially with voltage, but steric effects actually cause it to decrease above a threshold voltage where ions become crowded near the surface. Other nonlinear effects in PB theory are also strongly suppressed by steric effects: The net salt adsorption by the double layers in response to the applied voltage is greatly reduced, and so is the tangential "surface conduction" in the diffuse layer, to the point that it can often be neglected compared to bulk conduction (small Dukhin number).

Steric effects in the dynamics of electrolytes at large applied voltages: I. Double-layer charging

Ambient Intelligence (AmI) is a new paradigm in information technology aimed at empowering people's capabilities by means of digital environments that are sensitive, adaptive, and responsive to human needs, habits, gestures, and emotions. This futuristic vision of daily environment will enable innovative human-machine interactions characterized by pervasive, unobtrusive, and anticipatory communications. Such innovative interaction paradigms make AmI technology a suitable candidate for developing various real life solutions, including in the healthcare domain. This survey will discuss the emergence of AmI techniques in the healthcare domain, in order to provide the research community with the necessary background. We will examine the infrastructure and technology required for achieving the vision of AmI, such as smart environments and wearable medical devices. We will summarize the state-of-the-art artificial intelligence (AI) methodologies used for developing AmI system in the healthcare domain, including various learning techniques (for learning from user interaction), reasoning techniques (for reasoning about users' goals and intensions), and planning techniques (for planning activities and interactions). We will also discuss how AmI technology might support people affected by various physical or mental disabilities or chronic disease. Finally, we will point to some of the successful case studies in the area and we will look at the current and future challenges to draw upon the possible future research paths.

/pdf/a-survey-on-ambient-intelligence-in-healthcare-wkvwfodws9.pdf

A Survey on Ambient Intelligence in Healthcare

In this paper, a textile-based strain sensor has been developed to create a respiration belt. The constituent materials and the knitted structure of the textile sensor have been specifically selected and tailored for this application. Electromechanical modeling has been developed by exploiting Peirce's loop model in order to describe the fabric geometry under static and dynamic conditions. Kirchhoff's node and loop equations have been employed to create a generalized solution for the equivalent electrical resistance of the textile sensor for a given knitted loop geometry and for a specified number of loops. A laboratory test setup was built to characterize the prototype sensor and the resulting equivalent resistance under strain levels up to 40%, and consistent resistance response levels have been obtained from the sensor which correlate well with the modelled data. Production of the respiration belt was realized by bringing together knitted sensor and a relatively inelastic textile strap. Both machine simulations and real-time measurements on a human subject have been performed in order to calculate average breathing frequencies under different static and dynamic conditions. Also, different scenarios have been performed, such as slow breathing and rapid breathing. The sensory belt was located in either the chest area or in the abdominal area during the experimental measurements and the sensor yielded a good response under both static and dynamic conditions. However, body motion artefacts affected the signal quality under dynamic conditions and an additional signal-processing step was added to eliminate unwanted interference from the breathing signal.

Weft-Knitted Strain Sensor for Monitoring Respiratory Rate and Its Electro-Mechanical Modeling

Silver – silver chloride electrodes (Ag/AgCl) for the detection of chloride ions were fabricated using thick-film technology. Five different formulations were prepared and chloride responses were investigated over time. Almost identical and near Nernstian responses were observed over the first 162 days with an average chloride sensitivity for all formulations of -51.12 mV ± 0.45 mV per decade change in chloride concentration compared with a value of -50.59 mV ± 0.01 mV over 388 days for the best two formulations. After 6-months continuous immersion in tap water, pastes formulated with a glass binder began to exhibit a loss in sensitivity whilst those formulated from a commercial thick-film dielectric paste remained functional for the best part of a year. This difference in lifetime performance is attributed to the inclusion of proprietary additives in the commercial paste aiding adhesion and minimising AgCl leaching. The mechanical and chemical robustness of these electrodes has been demonstrated through their ability to detect changing levels of chloride when immersed in soil columns. This particular capacity will make them an invaluable tool in the fields of hydrology, agricultural science, soil science and environmental science.

Screen-printed potentiometric Ag/AgCl chloride sensors: Lifetime performance and their use in soil salt measurements

This paper summarises preliminary work comparing conductive yarns, knitting structures and yarn compositions in order to integrate smart sensor strips into a surrounding garment as a kinematic measurement tool. The conductive areas of the garment were to be used as a strain-sensitive material; ultimately measuring knee joint movement. In total, thirty sample fabrics were developed using conductive yarns; six of which were chosen to be tested for responsiveness during repeated strain. Preliminary tests showed good levels of responsiveness to strain and acceptable levels of recovery.

/pdf/fabric-based-strain-sensors-for-measuring-movement-in-37bre0s3e1.pdf

Fabric-based Strain Sensors for Measuring Movement in Wearable Telemonitoring Applications

Large structures such as buildings, ships and aircraft often contain numerous components which are exposed to a failure risk due to corrosion processes. Aside from the extra cost of repair and replacement, there are also potential health and safety issues which need to be addressed. Corrosion detection is not a new concept and there exists a variety of methods to detect and evaluate corrosion. The analysis method that is the focus of this work is capillary electrophoresis (CE); commonly used for a number of biological and chemical processes, such as drug, food and water quality analysis, DNA and protein separation and so on. An alternative method to high pressure (or performance) liquid chromatography (HPLC), CE boasts high analysis speeds as well as low limits of detection. Both are of crucial advantage for use in a pharmaceutical market; however the application of CE for in-situ or portable corrosion monitoring has not been investigated in great depth.

An Investigation into Separation Enhancement Methods for Miniaturised Planar Capillary Electrophoresis Devices

This study shows that the practical scaling of a hollow cathode thruster device to MEMS level should be possible albeit with significant divergence from traditional design. The main divergence is the need to operate at discharge pressures between 1-3bar to maintain emitter diameter pressure products of similar values to conventional hollow cathode devices. Without operating at these pressures emitter cavity dimensions become prohibitively large for maintenance of the hollow cathode effect and without which discharge voltage would be in the hundreds of volts as with conventional microdischarge devices. In addition this requires sufficiently constrictive orifice diameters in the 10?m – 50 ?m range for single cathodes or <5?m larger arrays. Operation at this pressure results in very small Debye lengths (4 -5.2pm) and leads to large reductions in effective work function (0.3 – 0.43eV) via the Schottky effect. Consequently, simple work function lowering compounds such as lanthanum hexaboride (LaB6) can be used to reduce operating temperature without the significant manufacturing complexity of producing porous impregnated thermionic emitters as with macro scale hollow cathodes, while still operating <1200°C at the emitter surface. The literature shows that LaB6 can be deposited using a variety of standard microfabrication techniques.

Andrew Cranny

Papers

Screen-printed potentiometric Ag/AgCl chloride sensors: Lifetime performance and their use in soil salt measurements

Fabric-based Strain Sensors for Measuring Movement in Wearable Telemonitoring Applications

An Investigation into Separation Enhancement Methods for Miniaturised Planar Capillary Electrophoresis Devices

On-a-chip microdischarge thruster arrays inspired by photonic device technology for plasma television

Improving separation resolution of capillary electrophoresis to aid corrosion monitoring by dynamic control of the EOF