This paper describes the approach used to identify elderly people’s needs and attitudes towards applying ambient sensor systems for monitoring daily activities in the home. As elderly are typically unfamiliar with such ambient technology, interactive tools for explicating sensor monitoring –an interactive dollhouse and iPad applications for displaying live monitored sensor activity data– were developed and used for this study. Furthermore, four studies conducted by occupational therapists with more than 60 elderly participants –including questionnaires (n=41), interviews (n=6), user sessions (n=14) and field studies (n=2)– were conducted. The experiences from these studies suggest that this approach helped to democratically engage the elderly as end-user and identify acceptance issues.
DOCUMENT
On April 16 and 17, 2020, the third edition of the Sensor data challenge was held by The Hague University of Applied Sciences, Statistics Netherlands, Utrecht University and the National Institute for Public Health and the Environment. The Sensor data challenge provides hardware (various sensors, raspberryPI) and software to teams with a mix of expertise in electronics, mechatronics, data science, user experience and industrial design. Teams need to design a tool and demonstrate its feasibility and relevance for one of the presented challenges. The third challenge had sensor measurements for living and working as the central theme. Winning solutions of the two previous editions have been the starting point for large-scale ongoing research projects. We like to present a brief summary of the solutions presented by the participating teams at the third challenge and offer the winning team the opportunity to share and discuss their ideas at the BigSurv20 with a larger audience.
VIDEO
A novel type of application for the exploration of enclosed or otherwise difficult to access environments requires large quantities of miniaturized sensor nodes to perform measurements while they traverse the environment in a “go with the flow” approach. Examples of these are the exploration of underground cavities and the inspection of industrial pipelines or mixing tanks, all of which have in common that the environments are difficult to access and do not allow position determination using e.g. GPS or similar techniques. The sensor nodes need to be scaled down towards the millimetre range in order to physically fit through the narrowest of parts in the environments and should measure distances between each other in order to enable the reconstruction of their positions relative to each other in offline analysis. Reaching those levels of miniaturization and enabling reconstruction functionality requires: 1) novel reconstruction algorithms that can deal with the specific measurement limitations and imperfections of millimetre-sized nodes, and 2) improved understanding of the relation between the highly constraint hardware design space of the sensor nodes and the reconstruction algorithms. To this end, this work provides a novel and highly robust sensor swarm reconstruction algorithm and studies the effect of hardware design trade-offs on its performance. Our findings based on extensive simulations, which push the reconstruction algorithm to its breaking point, provide important guidelines for the future development of millimetre-sized sensor nodes.
DOCUMENT
The maturing field of Wireless Sensor Networks (WSN) results in long-lived deployments that produce large amounts of sensor data. Lightweight online on-mote processing may improve the usage of their limited resources, such as energy, by transmitting only unexpected sensor data (anomalies). We detect anomalies by analyzing sensor reading predictions from a linear model. We use Recursive Least Squares (RLS) to estimate the model parameters, because for large datasets the standard Linear Least Squares Estimation (LLSE) is not resource friendly. We evaluate the use of fixed-point RLS with adaptive thresholding, and its application to anomaly detection in embedded systems. We present an extensive experimental campaign on generated and real-world datasets, with floating-point RLS, LLSE, and a rule-based method as benchmarks. The methods are evaluated on prediction accuracy of the models, and on detection of anomalies, which are injected in the generated dataset. The experimental results show that the proposed algorithm is comparable, in terms of prediction accuracy and detection performance, to the other LS methods. However, fixed-point RLS is efficiently implementable in embedded devices. The presented method enables online on-mote anomaly detection with results comparable to offline LS methods. © 2013 IEEE.
DOCUMENT
An energy harvesting device for obtaining energy from drops without needing of moving the drops along the device, in a reduced scale and combinable with othertypes of harvesting devices, the energy harvesting device comprising one or more triboelectric generators comprising a bottom electrode, a friction or triboelectric element placed over the bottom electrode, and at least two top electrodes placed over the triboelectric element and defining at least one gap between them, exposing the triboelectric element to the external environment so that on contacting a drop of liquid makes an electrical connection between the top electrodes varying the capacitance of the triboelectric generators and alternatively for functioning as a power unit for a sensor or as a self-powered sensor producing an electrical signal generated by the contact of the liquid with the electrodes.
DOCUMENT
In wheelchair sports, the use of Inertial Measurement Units (IMUs) has proven to be one of the most accessible ways for ambulatory measurement of wheelchair kinematics. A three-IMU configuration, with one IMU attached to the wheelchair frame and two IMUs on each wheel axle, has previously shown accurate results and is considered optimal for accuracy. Configurations with fewer sensors reduce costs and could enhance usability, but may be less accurate. The aim of this study was to quantify the decline in accuracy for measuring wheelchair kinematics with a stepwise sensor reduction. Ten differently skilled participants performed a series of wheelchair sport specific tests while their performance was simultaneously measured with IMUs and an optical motion capture system which served as reference. Subsequently, both a one-IMU and a two-IMU configuration were validated and the accuracy of the two approaches was compared for linear and angular wheelchair velocity. Results revealed that the one-IMU approach show a mean absolute error (MAE) of 0.10 m/s for absolute linear velocity and a MAE of 8.1◦/s for wheelchair angular velocity when compared with the reference system. The twoIMU approach showed similar differences for absolute linear wheelchair velocity (MAE 0.10 m/s), and smaller differences for angular velocity (MAE 3.0◦/s). Overall, a lower number of IMUs used in the configuration resulted in a lower accuracy of wheelchair kinematics. Based on the results of this study, choices regarding the number of IMUs can be made depending on the aim, required accuracy and resources available.
DOCUMENT
Purpose People with dementia (PwD) often present Behavioral and Psychological Symptoms of Dementia, which include agitation, apathy, and wandering amongst others, also known as challenging behaviors (CBs). These CBs worsen the quality of life (QoL) of the PwD and are a major source/reason of (increased) caregiver burden. The intricate nature of the symptoms implies that there is no “one size fits all solution”, and necessitates tailored approaches for both PwDs and caregivers. To timely prevent these behaviors assistive technology can be utilized to guide caregivers by enabling remote monitoring of contextual, environmental, and behavioral parameters, and subsequently alarming nurses on early-stage behavioral changes prior to the presentation of CBs. Eventually, the system should propose an intervention/action to prevent escalation. In turn, improvement in QoL for both caregivers and PwD living in nursing homes (NHs) is expected. In the current project “MOnitoring Onbegrepen Gedrag bij Dementie met sensortechnologie” (MOOD-Sense), we aim to develop such a monitoring system. The strengths of this new monitoring system lie in its ability to align with the individual needs of the PwD, utilization of a combination of wearables and ambient sensors to obtain contextual data, such as location or sound, and predict or monitor CBs individually rather than in groups, thus facilitating person-centered care, based on ontological reasoning. The project is divided into three parts, Toolbox A, B and C. Toolbox A focuses on obtaining insight in which behaviors are challenging according to nurses and how they are described. Previous studies utilize clinical terminology to describe or classify behavior, we aim to employ concrete descriptions of behavior that are observable and independent of clinical terminology, aligning with nurses who are often the first to notice behavior and can be operationalized such that it can also be aligned with sensor data. As a result, an ontology will be developed based on the data such that sensor data can be integrated into the same conceptual information that standardizes the communication in our monitoring system. Toolbox B focuses on translating data coming from various sensors into the concepts expressed in the ontology, and timely communicate situations of interest to the caregivers. In Toolbox C the focus is exploring interventions/actions employed in practice to prevent CBs. Method In Toolbox A we used a qualitative approach to collect descriptions of CBs. For this purpose, we employed focus groups (FGs) with nursing staff who provide daily care to PwD. In Toolbox B pilot studies were conducted. A set of experiments using sensors in NHs were performed. During each pilot, multiple PwD with CBs in NHs were monitored with both ambient and wearables sensors. The pilots were iteratively approached, which means that insights from previous pilot studies were used to improve consecutive pilot studies. Lastly, the elaboration of Toolbox C is ongoing. Results and Discussion Regarding Toolbox A four FGs were conducted during the period from January 2023 to May 2024. Each FG was comprised of four nurses (n = 16). From the FGs we gained insights into behavioral descriptions and the context of CBs. Although data analysis has to be performed yet, there are indications that changes preceding CBs can be observed, such as frowning or clenching fists for agitation or aggression. Further results will be available soon. Regarding Toolbox B a monitoring system, based on sensors, is developed iteratively (see Figure 1) and piloted in three consecutive NHs from January 2021 to December 2023. Each pilot was comprised of two PwD (n = 6). Analysis of sensor data is ongoing.
LINK
Anomaly detection is a key factor in the processing of large amounts of sensor data from Wireless Sensor Networks (WSN). Efficient anomaly detection algorithms can be devised performing online node-local computations and reducing communication overhead, thus improving the use of the limited hardware resources. This work introduces a fixed-point embedded implementation of Online Sequential Extreme Learning Machine (OS-ELM), an online learning algorithm for Single Layer Feed forward Neural Networks (SLFN). To overcome the stability issues introduced by the fixed precision, we apply correction mechanisms previously proposed for Recursive Least Squares (RLS). The proposed implementation is tested extensively with generated and real-world datasets, and compared with RLS, Linear Least Squares Estimation, and a rule-based method as benchmarks. The methods are evaluated on the prediction accuracy and on the detection of anomalies. The experimental results demonstrate that fixed-point OS-ELM can be successfully implemented on resource-limited embedded systems, with guarantees of numerical stability. Furthermore, the detection accuracy of fixed-point OS-ELM shows better generalization properties in comparison with, for instance, fixed-point RLS. © 2013 IEEE.
DOCUMENT
Sensor systems can be deployed in the homes of older adults living alone for functional health assessments. Their information is very useful for health care specialists. The problem lies in developing person independent models while facing a large variability in behavior. We address this problem by, first, proposing a new feature extraction method for data from ambient motion sensors. The method uses functional similarities between houses and daily structure to extract meaningful features. Second, we propose a change-based approach for analyzing data, taking difference scores of both the sensor features and health metrics. To evaluate our approach, experiments on longitudinal data were conducted, where the relationship between sensor data and health measurements was modeled with linear regression and (nonlinear) regression forests. These experiments show that the change-based approach yields better results and that the resulting models can be used as a reliable metric for (functional) health. In addition, feature analysis can help health care specialists understand relevant aspects of behavior. Prediction of health metrics is possible even with simple sensors. With such sensors, it is possible to detect problems and health decline in an early stage. This will have great impact on clinical practice.
DOCUMENT
A low-cost sensornode is introduced to monitor the 5G EMF exposure in the Netherlands for the four FR1 frequency bands. The sensornode is validated with in-lab measurements both with CW signals as for QAM signals and perform for both cases and for all frequency bands an error less than 1 dB for a dynamic range of 40 dB. This sensor is a follow up of the earlier version of our previously developed sensor and have substantial improvements in terms of linearity, error, and stability.
DOCUMENT
