Дисертації з теми "Human activity monitoring"

The main aim of the project is to develop an algorithm which will classify the activity performed by a human who is carrying a smart phone. The day to day life made humans very busy at work and during daily activities, mostly elderly people who are at home have an important need to monitor their activity by others when they are alone, if they are inactive for a long time without movement, or in some situations like if they have fallen down, became unconscious for sometime or seized with a cardiac arrest etc… will help the observer to know the state of activity of person being monitored. In this project we develop an algorithm to know the activity of a person using accelerometer available in Smartphone. We have extracted the Smartphone accelerometer data using an application called accelerometer data logger version 1.0 available in Smartphone market and have processed the data in Matlab for classifying the different activities of human being into static and dynamic activity, if the activity is dynamic then further classification into walking or running is performed with the algorithm. We implemented smoothening filters for data analysis and statistical techniques like standard deviation, mean and signal magnitude analysis for activity classification. This classification algorithm will let us know the type of activity either static or dynamic and then classify the position of the user, such as walking, running or ideal, which can provide useful information for the observer who is monitoring the activities of wearer, and which will help the wearer for his daily living. To bring out the extensive use of algorithm and to provide valuable feedback for wearer regarding his activities, energy spent by user during the activities was calculated at a given time using regression methods and was implemented in the algorithm. The developed model was able to estimate the energy spent by the user, the observations recorded were almost similar to the treadmill data which is taken as a standard for our model and the mean error is not more than ±2 for 30 observations. The final results when compared with the standard model was proved to be 93 % accurate on average of 30 subjects data which is used for verifying the algorithm developed. With these set of results we have come to a conclusion that algorithm can be easily implemented in a real time Smartphone application with low false predictions and can be implemented with low computational cost and fast real-time response. In future our classification algorithm can also be used in military applications where one can know what the soldier is doing without actually seeing him and additionally it can be proved as a support system in athlete’s health monitoring and training.
I denna modell har vi utvecklat en algoritm för aktivitetsklassificeringoch energiförbrukning uppskattning , vilket hjälper oss i övervakningen daglig mänsklig aktivitet med större noggrannhet . Resultaten valideras med standard energiförbrukning teknik och aktivitetsklassificeringsvideoobservationer. Vi vill att denna modell ska integreras i smarta mobiltelefoner för att ge slutanvändaren en vänlig atmosfär utan att lägga några komplicerade funktioner för hantering av utrustningen . Denna modell är mycket användbart i klinisk uppföljning av patienterna , kommer det att hjälpa oss att övervaka gamla , sjuka och utvecklingsstörda personens aktivitetsidentifiering och hjälper oss i nära övervakning av patienterna men fysiskt att vara borta från dem . Våra bärbara MEMS baserade treaxlig accelerometer system baserat smartphone kompatibel algoritm tillsammans med andra fysiologiska övervakningsparametrarkommer att ge korrekt övervakning rörelse och energiförbrukning uppskattning för klinisk analys . Denna modell är användbar för analys och övervakning av grupp -och enskilda individer , vilket kommer att leda till att spåra deras rörelser och en framgångsrik räddningsaktion för att rädda dem från dödliga sjukdomar och förebygga risker när de är skadade . Framtida arbete kommer att vara kontinuerlig övervakning av ämnen enskild aktivitet tillsammans med gruppaktivitet . Identifiera hållning övergång av olika aktiviteter i en kort tid som att springa till sittande , sittande till stående , står att krypa etc.
0091-7660885577

A spacewalk, or extravehicular activity (EVA), is one of the most mission critical and physically and cognitively challenging tasks that crewmembers complete. With next-generation missions to the Moon and Mars, exploration EVA will challenge crewmembers in partial gravity environments with increased frequency, duration, and autonomy of operations. Given the distance from Earth, associated communication delays, and durations of exploration missions, there is a monumental shift in responsibility and authority taking place in spaceflight; moving from Earth-dependent to crew self-reliant. For the safety, efficacy, and efficiency of future surface EVAs, there is a need to better understand crew health and performance. With this knowledge, technology and operations can be designed to better support future crew autonomy. The focus of this dissertation is to develop and evaluate a psychophysiological monitoring tool to classify cognitive workload during an operationally relevant EVA task. This was completed by compiling a sensor suite of commercial wearable devices to record physiological signals in two human research studies, one at Virginia Tech and one at NASA Johnson Space Center. The approach employs supervised machine learning to recognize patterns in psychophysiological features across different psychological states. This relies on the ability to simulate, or induce, cognitive workload in order to label data for training the model. A virtual reality (VR) Translation Task was developed to control and quantify cognitive demands during an immersive, ambulatory EVA scenario. Participants walked on a passive treadmill while wearing a VR headset to move along a virtual lunar surface. They walked with constraints on time and resources, while simultaneously identifying and recalling waypoints in the scene. Psychophysiological features were extracted and labeled according to the task demands, i.e. high or low cognitive workload, for the novel Translation Task, as well as for the benchmark Multi-Attribute Task Battery (MATB). Predictive models were created using the K Nearest Neighbor (KNN) algorithm. The contributions of this dissertation span the simulation, characterization, and modeling of cognitive state. Ultimately, this work tests the limits of extending laboratory psychophysiological monitoring to more realistic environments using wearable devices, and of generalizing predictive models across participants, times, and tasks. This work paves the way for future field studies and real-time implementation to close the loop between human and automation.
Doctor of Philosophy
A spacewalk is one of the most important and physically and mentally challenging tasks that astronauts complete. With next-generation missions to the Moon and Mars, exploration spacewalks will challenge astronauts in reduced-weight environments (1/6 and 1/3 Earth's gravity) with longer, more frequent spacewalks and with less help from mission control. To keep astronauts safe while exploring there is a need to better understand astronaut health and performance (physical and mental) during spacewalks. With knowledge of how astronauts will respond to high workload and stressful events, we can plan missions and design tools that can best assist them during spacewalks on the Moon and Mars when help from Earth mission control is limited. Traditional tools of quantifying mental state are not suitable for real-time assessment during spacewalks. Current methods, including subjective surveys and performance-based computer tests, require time and attention to complete and cannot assess real-time operations. The focus of this dissertation is to create a psychophysiological monitoring tool to measure mental workload during a virtual reality (VR) spacewalk. Psychophysiological monitoring uses physiological measures, like heart rate and breathing rate, to predict psychological state, like high workload or stress. Physiological signals were recorded using commercial wearable devices in two human research studies, one at Virginia Tech and one at NASA Johnson Space Center. With machine learning, computer models can be trained to recognize patterns in physiological measures for different psychological states. Once a model is trained, it can be tested on new data to predict mental workload. To train and test the models, participants in the studies completed high and low workload versions of the VR task. The VR task was specifically designed for this study to simulate and measure performance during a mentally-challenging spacewalk scenario. The participants walked at their own pace on a treadmill while wearing a VR headset to move along a virtual lunar surface, while balancing their time and resources. They were also responsible for identifying and recalling flags along their virtual path. Ultimately, this work tests the limits of extending laboratory psychophysiological monitoring to more realistic environments using wearable devices, and of generalizing predictive models across participants, times, and tasks. This work paves the way for future field studies and real-time implementation to close the loop between human and automation.

Urbanisation in this fast-paced world although possessing some lucrative advantages causes some serious problems to its inhabitants. Green spaces are important in highly urbanised societies for adequate restoration of mental health and physical well being. This study focuses on understanding people’s behaviour in green spaces. To enable this, this study was designed with a video of volunteers in a greenspace. In order to automate the data collection required to observe the participants and study their behavioural patterns, computer science aided interventions and machine learning algorithms were employed. YOLOv4 enabled the detection of objects using a regression-based approach to accurately determine the position of the bounding boxes. Using the bounding box coordinates, experiments were conducted with several use cases like hotspot detection and crowd detection. Further using transfer learning, attempts were made to recognize the actions of humans in the videos. The experiments were evaluated using the mean Average Precision technique and achieved good results for the use cases mentioned above. With implications in hotspot detection and crowd detection, the outcome of the study can contribute towards a better and efficient object detection and action recognition.

Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2014.
50
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 155-160).
Non-intrusive load monitoring (NILM) has three distinct advantages over today's smart meters. First, it offers accountability. Few people know where their kWh's are going. Second, it is a maintenance tool. Signs of wear are detectable through their electrical signal. Third, it provides awareness of human activity within a network. Each device has an electrical fingerprint, and specific devices imply associated human actions. From voltage and current measurements at a single point on the network, non-intrusive load monitoring (NILM) disaggregates appliance-level information. This information is available remotely in bandwidth-constrained environments. Four real-world field tests at military micro grids and commercial buildings demonstrate the utility of the NILM in reducing electrical demand, enabling condition-based maintenance, and inferring human activity from electrical activity.
by Mark Daniel Gillman.
S.M.

With the advancement of computer technology, demand for more accurate and intelligent monitoring systems has also risen. The use of computer vision and video analysis range from industrial inspection to surveillance. Object detection and segmentation are the first and fundamental task in the analysis of dynamic scenes. Traditionally, this detection and segmentation are typically done through temporal differencing or statistical modelling methods. One of the most widely used background modeling and segmentation algorithms is the Mixture of Gaussians method developed by Stauffer and Grimson (1999). During the past decade many such algorithms have been developed ranging from parametric to non-parametric algorithms. Many of them utilise pixel intensities to model the background, but some use texture properties such as Local Binary Patterns. These algorithms function quite well under normal environmental conditions and each has its own set of advantages and short comings. However, there are two drawbacks in common. The first is that of the stationary object problem; when moving objects become stationary, they get merged into the background. The second problem is that of light changes; when rapid illumination changes occur in the environment, these background modelling algorithms produce large areas of false positives.
These algorithms are capable of adapting to the change, however, the quality of the segmentation is very poor during the adaptation phase. In this thesis, a framework to suppress these false positives is introduced. Image properties such as edges and textures are utilised to reduce the amount of false positives during adaptation phase. The framework is built on the idea of sequential pattern recognition. In any background modelling algorithm, the importance of multiple image features as well as different spatial scales cannot be overlooked. Failure to focus attention on these two factors will result in difficulty to detect and reduce false alarms caused by rapid light change and other conditions. The use of edge features in false alarm suppression is also explored. Edges are somewhat more resistant to environmental changes in video scenes. The assumption here is that regardless of environmental changes, such as that of illumination change, the edges of the objects should remain the same. The edge based approach is tested on several videos containing rapid light changes and shows promising results. Texture is then used to analyse video images and remove false alarm regions. Texture gradient approach and Laws Texture Energy Measures are used to find and remove false positives. It is found that Laws Texture Energy Measure performs better than the gradient approach. The results of using edges, texture and different combination of the two in false positive suppression are also presented in this work. This false positive suppression framework is applied to a smart house senario that uses cameras to model ”virtual sensors” to detect interactions of occupants with devices. Results show the accuracy of virtual sensors compared with the ground truth is improved.

The way in which infants play with objects can be indicative of their developmental progress and may serve as an early indicator for developmental delays. However, the observation of children interacting with toys for the purpose of quantitative analysis can be a difficult task. To better quantify how play may serve as an early indicator, researchers have conducted retrospective studies examining the differences in object play behaviors among infants. However, such studies require that researchers repeatedly inspect videos of play often at speeds much slower than real-time to indicate points of interest. The research presented in this dissertation examines whether a combination of sensors embedded within toys and automatic pattern recognition of object play behaviors can help expedite this process. For my dissertation, I developed the Child'sPlay system which uses augmented toys and statistical models to automatically provide quantitative measures of object play interactions, as well as, provide the PlayView interface to view annotated play data for later analysis. In this dissertation, I examine the hypothesis that sensors embedded in objects can provide sufficient data for automatic recognition of certain exploratory, relational, and functional object play behaviors in semi-naturalistic environments and that a continuum of recognition accuracy exists which allows automatic indexing to be useful for retrospective review. I designed several augmented toys and used them to collect object play data from more than fifty play sessions. I conducted pattern recognition experiments over this data to produce statistical models that automatically classify children's object play behaviors. In addition, I conducted a user study with twenty participants to determine if annotations automatically generated from these models help improve performance in retrospective review tasks. My results indicate that these statistical models increase user performance and decrease perceived effort when combined with the PlayView interface during retrospective review. The presence of high quality annotations are preferred by users and promotes an increase in the effective retrieval rates of object play behaviors.

Wearable sensors and systems have become increasingly popular for diverse applications. An emerging technology for physical activity assessment is Smart Textile Systems (STSs), comprised of sensitive/actuating fiber, yarn, or fabric that can sense an external stimulus. All required components of an STS (sensors, electronics, energy supply, etc.) can be conveniently embedded into a garment, providing a fully textile-based system. Thus, STSs have clear potential utility for measuring health-relevant aspects of human activity, and to do so passively and continuously in diverse environments. For these reasons, STSs have received increasing interest in recent studies. Despite this, however, limited evidence exists to support the implementation of STSs during diverse applications. Our long-term goal was to assess the feasibility and accuracy of using an STS to monitor human activities. Our immediate objective was to investigate the accuracy of an STS in three representative applications with respect to occupational scenarios, healthcare, and activities of daily living. A particular STS was examined, consisting of a smart socks (SSs), using textile pressure sensors, and smart undershirt (SUS), using textile strain sensors. We also explored the relative merits of these two approaches, separately and in combination. Thus, five studies were completed to design and evaluate the usability of the smart undershirt, and investigate the accuracy of implementing an STS in the noted applications. Input from the SUS led to planar angle estimations with errors on the order of 1.3 and 9.4 degrees for the low-back and shoulder, respectively. Overall, individuals preferred wearing a smart textile system over an IMU system and indicated the former as superior in several aspects of usability. In particular, the short-sleeved T-shirt was the most preferred garments for an STS. Results also indicated that the smart shirt and smart socks, both individually and in combination, could detect occupational tasks, abnormal and normal gaits, and activities of daily living with greater than 97% accuracy. Based on our findings, we hope to facilitate future work that more effectively quantifies sedentary periods that may be deleterious to human health, as well as detect activity types that may be help or hinder health and fitness. Such information may be of use to individuals and workers, healthcare providers, and ergonomists. More specifically, further analyses from this investigation could provide strategies for: (a) modifying a sedentary lifestyle or work scenario to a more active one, and (b) helping to more accurately identify occupational injury risk factors associated with human movement.
PHD

Adverse health effects from exposure to air pollution, although at present only partly understood, are a global challenge and of widespread concern. Quantifying human exposure to air pollutants is challenging, as ambient concentrations of air pollutants at potentially harmful levels are ubiquitous and subject to high spatial and temporal variability. At the same time, individuals have their very own unique activity-patterns. Hence exposure results from intertwined relationships between environmental and human systems add complexity to the assessment process. It is essential to develop a deeper understanding of individual exposure pathways and situations occurring in people’s everyday lives. This is important especially with regard to exposure and health impact assessment which provide the basis for public health advice and policy development. This thesis describes the development and application of a personal monitoring method to assess exposure to fine particulate matter in a variety of microenvironments. Tools and methods applied are tested with respect to feasibility, intrusiveness, performance and potential for future applications. The development of the method focuses on the application in everyday environments and situations in an attempt to capture as much of the total exposure as possible, across a complete set of microenvironments. Seventeen volunteers took part in the pilot study, collected data and provided feedback on methodology and tools applied. The low-cost particle counter applied showed good agreement with reference instruments when studied in two different environments. Based on the assessment of the two instruments functions to derive particle mass concentration from the original particle number counts have been defined. The application of the devices and tools received positive feedback from the volunteers. Limitations are mainly related to the non-weatherproof design of the particle counter. The collection of time-activity patterns with GPS and time-activity diaries is challenging and requires careful processing. Resulting personal exposure profiles highlight the influence of individual activities and contextual factors. Highest concentrations were measured in indoor environments where people also spent the majority of time. Differences between transport modes as well as between urban and rural areas were identified.

This thesis explores a concept by Sonvox AB called VoiceUp. The concept is that of a mobile application that communicates with a wearable voice measuring device and visualizes voice information in a way that helps singers. Sonvox current main product is VoxLog—a system for long-term voice monitoring, mostly used for research purposes. Sonvox believes that their voice analysis technology could be relevant to a larger audience. The main goal for the thesis is to explore if a wearable voice measuring device could be relevant to singers, and in what ways. To do this, a needs analysis was conducted where song teachers and singers were interviewed. In order to draw statistical conclusions about the occurrence of needs, a survey was conducted where people with an interest for singing were the targeted respondents. Based on the result from the needs analysis, the VoiceUp concept was refined, resulting in an idea of a product that measures and visualizes how much the user sings and speaks with the aim to increase singers motivation to practice singing more regularly. Based on theory related to self-tracking, a design proving the concept was created, resulting in a mockup and a simple prototype. The mockup and the prototype can together be seen as one example of how self-tracking technology could be relevant to singers.

Personal informatics systems help people collect and reflect on behavioral information to better understand their own behavior. Because most systems only show one type of behavioral information, finding factors that affect one’s behavior is difficult. Supporting exploration of multiple types of contextual and behavioral information in a single interface may help. To explore this, I developed prototypes of IMPACT, which supports reflection on physical activity and multiple types of contextual information. I conducted field studies of the prototypes, which showed that such a system could increase people’s awareness of opportunities for physical activity. However, several limitations affected the usage and value of these prototypes. To improve support for such systems, I conducted a series of interviews and field studies. First, I interviewed people about their experiences using personal informatics systems resulting in the Stage-Based Model of Personal Informatics Systems, which describes the different stages that systems need to support, and a list of problems that people experience in each of the stages. Second, I identified the kinds of questions people ask about their personal data and found that the importance of these questions differed between two phases: Discovery and Maintenance. Third, I evaluated different visualization features to improve support for reflection on multiple kinds of data. Finally, based on this evaluation, I developed a system called Innertube to help people reflect on multiple kinds of data in a single interface using a visualization integration approach that makes it easier to build such tools compared to the more common data integration approach.

The use of autonomous robots in our society is increasing every day and a robot is no longer seen as a tool but as a team member. The robots are now working side by side with us and provide assistance during dangerous operations where humans otherwise are at risk. This development has in turn increased the need of robots with more human-awareness. Therefore, this master thesis aims at contributing to the enhancement of human-aware robotics. Specifically, we are investigating the possibilities of equipping autonomous robots with the capability of assessing and detecting activities in human teams. This capability could, for instance, be used in the robot's reasoning and planning components to create better plans that ultimately would result in improved human-robot teamwork performance. we propose to improve existing teamwork activity recognizers by adding intangible features, such as stress, motivation and focus, originating from human behavior models. Hidden markov models have earlier been proven very efficient for activity recognition and have therefore been utilized in this work as a method for classification of behaviors. In order for a robot to provide effective assistance to a human team it must not only consider spatio-temporal parameters for team members but also the psychological.To assess psychological parameters this master thesis suggests to use the body signals of team members. Body signals such as heart rate and skin conductance. Combined with the body signals we investigate the possibility of using System Dynamics models to interpret the current psychological states of the human team members, thus enhancing the human-awareness of a robot.
Användningen av autonoma robotar i vårt samhälle ökar varje dag och en robot ses inte längre som ett verktyg utan som en gruppmedlem. Robotarna arbetar nu sida vid sida med oss och ger oss stöd under farliga arbeten där människor annars är utsatta för risker. Denna utveckling har i sin tur ökat behovet av robotar med mer människo-medvetenhet. Därför är målet med detta examensarbete att bidra till en stärkt människo-medvetenhet hos robotar. Specifikt undersöker vi möjligheterna att utrusta autonoma robotar med förmågan att bedöma och upptäcka olika beteenden hos mänskliga lag. Denna förmåga skulle till exempel kunna användas i robotens resonemang och planering för att ta beslut och i sin tur förbättra samarbetet mellan människa och robot. Vi föreslår att förbättra befintliga aktivitetsidentifierare genom att tillföra förmågan att tolka immateriella beteenden hos människan, såsom stress, motivation och fokus. Att kunna urskilja lagaktiviteter inom ett mänskligt lag är grundläggande för en robot som ska vara till stöd för laget. Dolda markovmodeller har tidigare visat sig vara mycket effektiva för just aktivitetsidentifiering och har därför använts i detta arbete. För att en robot ska kunna ha möjlighet att ge ett effektivt stöd till ett mänskligtlag måste den inte bara ta hänsyn till rumsliga parametrar hos lagmedlemmarna utan även de psykologiska. För att tyda psykologiska parametrar hos människor förespråkar denna masteravhandling utnyttjandet av mänskliga kroppssignaler. Signaler så som hjärtfrekvens och hudkonduktans. Kombinerat med kroppenssignalerar påvisar vi möjligheten att använda systemdynamiksmodeller för att tolka immateriella beteenden, vilket i sin tur kan stärka människo-medvetenheten hos en robot.

The thesis work was conducted in Stockholm, Kista at the department of Informatics and Aero System at Swedish Defence Research Agency.

Human body is a complex system organized at different levels such as cells, tissues and organs, which contributes to 11 important organ systems. The functional efficiency of this complex system is evaluated as health. Traditional healthcare is unable to accommodate everyone's need due to the ever-increasing population and medical costs. With advancements in technology and medical research, traditional healthcare applications are shaping into smart healthcare solutions. Smart healthcare helps in continuously monitoring our body parameters, which helps in keeping people health-aware. It provides the ability for remote assistance, which helps in utilizing the available resources to maximum potential. The backbone of smart healthcare solutions is Internet of Things (IoT) which increases the computing capacity of the real-world components by using cloud-based solutions. The basic elements of these IoT based smart healthcare solutions are called "things." Things are simple sensors or actuators, which have the capacity to wirelessly connect with each other and to the internet. The research for this dissertation aims in developing architectures for these things, focusing on IoT-based smart healthcare solutions. The core for this dissertation is to contribute to the research in smart healthcare by identifying applications which can be monitored remotely. For this, application-specific thing architectures were proposed based on monitoring a specific body parameter; monitoring physical health for family and friends; and optimizing the power budget of IoT body sensor network using human body communications. The experimental results show promising scope towards improving the quality of life, through needle-less and cost-effective smart healthcare solutions.

Le premier appareil est un ensemble de détection portable pour la médecine nucléaire. Il permet de compter des rayons gamma, sur deux voies suivant une période d'échantillonnage variable et durant plus de 24 heures. Le second appareil mesure 4 paramètres (fréquence cardiaque, activité, températures ambiante et cutanée) durant plus de 15 jours

Nous avons déployé en milieu hospitalier des Habitat Intelligent pour la Santé composé d'un réseau de capteur infrarouges passifs et de l'infrastructure informatique permettant une fusion de données d'actimétrie en vue de dépister des variations d'état de santé de patients fragiles. Le choix de service de moyen et de long séjour avait pour objet d'avoir des lieux se rapprochant le plus possible d'un domicile tout en ayant un personnel suffisant comme étalon de l'HIS. Ceci nous a permis de conclure : 1. Que l'HIS était acceptable et peu intrusif pour la population bénéficiaire potentielle. 2. Que les capacités de détection d'alarme de l'HIS peuvent être modélisées et comparées à d'autres dispositifs mesurant la même grandeur (activité) dans le cadre de tableaux de fragilité ciblés. 3. Nous avons vérifié notre modèle sur un exemple de chute chez un patient avec maladie à corps de Lewy. 4. Que l'HIS peut répondre à une gestion de risque dans la norme ISOFDIS 14971. 5. Qu'en absence de soutien par la solidarité nationale, l'HIS peut être amorti dès la première année dans le cas d'une dame âgée fragile faisant des décompensations cardiaques. Il y a donc une pertinence médico-socio-économique à mettre un dispositif de type HIS sur le marché potentiel que nous avons démontré par notre travail scientifique
An activity monitoring health smart home (HIS type) based upon an infrared sensor network and the related data fusion was deployed in intermediate and long-term geriatric wards. Frail elderly people in those facilities are the closest substitute to home-dwellers, yet the staff provides the standard for activity monitoring. We could show that: 1. HIS is little intrusive and readily accepted. 2. We may model the capabilities of HIS to trigger an alarm and compare it to other devices measuring activity in specific frail populations. 3. We proved our model on the clinical case of a falling Lewy's bodies disease patient. 4. ISOFDIS 14971 norm on risk management may apply to HIS. 5. Returns may be expected from the first implementation year if an HI Sis deployed at the home a frail elderly lady with cardiovascular condition. Therefore, proposing a commercial service using HIS is medically, socially and economically relevant

碩士
國立成功大學
醫學資訊研究所
102
Human Activity Recognition and consequent Human Energy Expenditure monitoring have become one of the emerging researched topics nowadays. With newly available sensors and wearable devices getting significant attention, it brought the idea to develop an integrated system for such purposes with running on a smartphone device. In this thesis, we aim to explore various methods for building a parameterized, continuous and energy efficient system to monitor human daily energy expenditure running on a smartphone. We design and implement a system for Activity Recognition and test it on four public data sets. The experimental results show that the performance of our system is comparable with the previous approaches using multiple sensor data, while our approach uses only single stream of accelerometer data.

Yes
Human activity recognition from sensor readings have proved to be an effective approach in pervasive computing for smart healthcare. Recent approaches to ambient assisted living (AAL) within a home or community setting offers people the prospect of more individually-focused care and improved quality of living. However, most of the available AAL systems are often limited by computational cost. In this paper, a simple, novel non-wearable human activity classification framework using the multivariate Gaussian is proposed. The classification framework augments prior information from the passive RFID tags to obtain more detailed activity profiling. The proposed algorithm based on multivariate Gaussian via maximum likelihood estimation is used to learn the features of the human activity model. Twelve sequential and concurrent experimental evaluations are conducted in a mock apartment environment. The sampled activities are predicted using a new dataset of the same activity and high prediction accuracy is established. The proposed framework suits well for the single and multi-dwelling environment and offers pervasive sensing environment for both patients and carers.
Tertiary Education Trust Fund of Federal Government of Nigeria and by the European Union’s Horizon 2020 research and innovation programme under Grant Agreement H2020-MSCA-ITN-2016 SECRET-722424

The desire to send humans to Mars will require a change in the way that extravehicular activity (EVA) is performed; in-space crews (including those within a vehicle or habitat monitoring others conducting EVA) will need to be more autonomous and that will require them to monitor large amounts of information in order to ensure crew safety and mission success. The amount of information to perceive and process will overwhelm unassisted intra-vehicular (IV) crewmembers, meaning that automation will need to be developed to support these crews on Mars while EVA is performed (Mishkin, Lee, Korth, & LeBlanc, 2007). This dissertation seeks to identify the information requirements for the performance of scientific EVA and determine which information streams will need to be allocated to in-space crew and which are the most effective streams to automate. The first study uses Mars rover operations as a homology—as defined by von Bertalanffy (1968)—to human scientific exploration. Mars rover operations personnel were interviewed using a novel method to identify the information requirements to perform successful science on Mars, how that information is used, and the timescales on which those information streams operate. The identified information streams were then related to potential information streams relevant to human exploration in order to identify potential function allocation or automated system development areas. The second study focused on one identified mission-critical information stream for human space exploration: monitoring astronaut status physiologically. Heart rate, respiration rate, and heart rate variability measurements were recorded from participants as they performed field science tasks (potentially tasks that are similar to those that will be performed by astronauts on Mars). A statistical method was developed to analyze this data in order to determine whether or not physiological responses to different tasks were statistically different, and whether any of those differences followed consistent patterns. A potential method to automate the monitoring of physiological data was also described. The results of this work provide a more detailed outline of the information requirements for EVA on Mars and can be used as a starting point for others in the exploration community to further develop automation or function allocation to support astronauts as they explore Mars.

abstract: Movement disorders are becoming one of the leading causes of functional disability due to aging populations and extended life expectancy. Diagnosis, treatment, and rehabilitation currently depend on the behavior observed in a clinical environment. After the patient leaves the clinic, there is no standard approach to continuously monitor the patient and report potential problems. Furthermore, self-recording is inconvenient and unreliable. To address these challenges, wearable health monitoring is emerging as an effective way to augment clinical care for movement disorders. Wearable devices are being used in many health, fitness, and activity monitoring applications. However, their widespread adoption has been hindered by several adaptation and technical challenges. First, conventional rigid devices are uncomfortable to wear for long periods. Second, wearable devices must operate under very low-energy budgets due to their small battery capacities. Small batteries create a need for frequent recharging, which in turn leads users to stop using them. Third, the usefulness of wearable devices must be demonstrated through high impact applications such that users can get value out of them. This dissertation presents solutions to solving the challenges faced by wearable devices. First, it presents an open-source hardware/software platform for wearable health monitoring. The proposed platform uses flexible hybrid electronics to enable devices that conform to the shape of the user’s body. Second, it proposes an algorithm to enable recharge-free operation of wearable devices that harvest energy from the environment. The proposed solution maximizes the performance of the wearable device under minimum energy constraints. The results of the proposed algorithm are, on average, within 3% of the optimal solution computed offline. Third, a comprehensive framework for human activity recognition (HAR), one of the first steps towards a solution for movement disorders is presented. It starts with an online learning framework for HAR. Experiments on a low power IoT device (TI-CC2650 MCU) with twenty-two users show 95% accuracy in identifying seven activities and their transitions with less than 12.5 mW power consumption. The online learning framework is accompanied by a transfer learning approach for HAR that determines the number of neural network layers to transfer among uses to enable efficient online learning. Next, a technique to co-optimize the accuracy and active time of wearable applications by utilizing multiple design points with different energy-accuracy trade-offs is presented. The proposed technique switches between the design points at runtime to maximize a generalized objective function under tight harvested energy budget constraints. Finally, we present the first ultra-low-energy hardware accelerator that makes it practical to perform HAR on energy harvested from wearable devices. The accelerator consumes 22.4 microjoules per operation using a commercial 65 nm technology. In summary, the solutions presented in this dissertation can enable the wider adoption of wearable devices.
Dissertation/Thesis
Human activity recognition dataset
Doctoral Dissertation Computer Engineering 2020

Dissertação de mestardo em Industrial Eletronics and Computers
Sarcopenia, the age-associated loss of skeletal muscle mass, has been postulated to be a major factor in the strength decline with ageing. Considering the increase in the number of people with muscle weakness, the monitoring of a person’s muscular activity becomes a necessity. This need is also present in the sports area, since the muscles monitoring allows an improvement in the athlete’s technique and may also help preventing possible injuries. The standard method for muscle monitoring is the electromyography signal acquisition, although it presents various problems, like their lack of ergonomics, requiring hairless skin and gel inserted in it, and the need of complex electronics, demanding several hardware filters, since the EMG raw data is full of noise. This dissertation consists in developing a wearable prototype to monitor the user’s muscular activity, through force sensing resistors sensors, and recognize the toe-off gait event. The sensors data are processed by a microcontroller and are sent to a desktop application through wireless connection, or saved in a memory card for a later analysis. This project was also integrated in the robotic system SmartOs. The force sensors output signals were validated by comparing them to the EMG signals. These trials were divided in two groups: static trials, in which the subjects performs specific gestures several times, and dynamic trials, where the subject walks in different paces (slow, medium and fast). Both signals showed some similarity between them, although their similarities were more obvious in the static trials because of their more simple and linear signals. Several regression methods were validated in order to convert the FSR in EMG signals, but the results showed poor results, discarding this possible implementation. The gait event toe-off recognition algorithm was also validated in the dynamic trials performed. The results were satisfactory, showing a high accuracy percentage and low delay times. This dissertation project should provide an easier way to monitor muscles, discarding the needs of complex electronics and hairless skin and providing a clean signal with few noise.
Sarcopenia, a perda de massa muscular esquelética associada à idade, tem sido postulada como um fator importante no declínio de força com o envelhecimento. Com o aumento do número de pessoas com fraqueza muscular, uma monitorização da atividade muscular de uma pessoa torna-se uma necessidade. Esta necessidade também está presente na área do desporto, em que a monitorização muscular permite uma melhoria na técnica do atleta ou prevenir possíveis lesões. O método padrão para a monitorização muscular é a aquisição do sinal EMG, embora apresente vários problemas, como a sua falta de ergonomia, exigindo pele depilada e inserção dum gel específico, e a necessidade de eletrónica complexa, composta por vários filtros, uma vez que os sinais EMG contém muito ruído. Esta dissertação consiste em desenvolver um protótipo vestível para monitorizar a atividade muscular do utilizador através de sensores piezoresistivos e reconhecer o evento da marcha toe-off. Os dados dos sensores são processados por um microcontrolador que envia os dados para uma aplicação gráfica por comunicação wireless ou então são guardados num cartão de memória para uma futura análise. Este sistema também foi integrado no sistema robótico SmartOs. Os sinais provenientes dos sensores de força foram validados, comparando-os com os sinais EMG. Estes testes foram divididos em dois grupos: testes estáticos, onde a pessoa realiza movimentos específicos repetidamente, e testes dinâmicos, onde a pessoa caminha em diferentes velocidades (lenta, média e rápida). Os testes mostraram alguma semelhança entre os dois sinais, embora estas semelhanças foram mais visíveis nos testes estáticos devido ao facto dos seus sinais serem mais simples e lineares que nos testes dinâmicos. O algoritmo de reconhecimento do evento toe-off foi validado nos testes dinâmicos realizados, mostrando resultados satisfatórios tais como altas percentagens de precisão e curtos atrasos temporais. Este projeto deverá fornecer uma maneira mais fácil de monitorizar os músculos, não necessitando de eletrónica complexa ou de ter a pele depilada e a inserção de gel, fornecendo assim um sinal livre de muito ruído.