This dissertation presents a comprehensive exploration encompassing the design, development, control and the application of reinforcement learning-based force planning for the autonomous grasping capabilities of the innovative assistive robotic exoskeleton gloves. Exoskeleton devices have emerged as a promising avenue for providing assistance to individuals with hand disabilities, especially those who may not achieve full recovery through surgical interventions. Nevertheless, prevailing exoskeleton glove systems encounter a multitude of challenges spanning design, control, and human-machine interaction. These challenges have given rise to limitations, such as unwieldy bulkiness, an absence of precise force control algorithms, limited portability, and an imbalance between lightweight construction and the essential functionalities required for everyday activities. To address these challenges, this research undertakes a comprehensive exploration of various dimensions within the exoskeleton glove system domain. This includes the intricate design of the finger linkage mechanism, meticulous kinematic analysis, strategic kinematic synthesis, nuanced dynamic modeling, thorough simulation, and adaptive control. The development of two distinct types of series elastic actuators, coupled with the creation of two diverse exoskeleton glove designs based on differing mechanisms, constitutes a pivotal aspect of this study. For the exoskeleton glove integrated with series elastic actuators, a sophisticated dynamic model is meticulously crafted. This endeavor involves the formulation of a mathematical framework to address backlash and the subsequent mitigation of friction forces. The pursuit of accurate force control culminates in the proposition of a data-driven model-free force predictive control policy, compared with a dynamic model-based force control methodology. Notably, the efficacy of the system is validated through meticulous clinical experiments. Meanwhile, the low-profile exoskeleton glove design with a novel mechanism engages in a further reduction of size and weight. This is achieved through the integration of a rigid coupling hybrid mechanism, yielding pronounced advancements in wearability and comfortability. A deep reinforcement learning approach is adopted for the real-time force planning control policies. A simulation environment is built to train the reinforcement learning agent. In summary, this research endeavors to surmount the constraints imposed by existing exoskeleton glove systems. By virtue of advancing mechanism design, innovating control strategies, enriching perception capabilities, and enhancing wearability, the ultimate goal is to augment the functionality and efficacy of these devices within the realm of assistive applications. / Doctor of Philosophy / This dissertation presents a comprehensive exploration encompassing the design, development, control and the application of reinforcement learning-based force planning for the autonomous grasping capabilities of the innovative assistive robotic exoskeleton gloves. Exoskeleton devices hold significant promise as valuable aids for patients with hand disabilities who may not achieve full recuperation through surgical interventions. However, the present iteration of exoskeleton glove systems encounters notable limitations in terms of design, control mechanisms, and human-machine interaction. Specifically, prevailing systems often suffer from bulkiness, lack of portability, and an inadequate equilibrium between lightweight construction and the essential functionalities imperative for daily tasks. To address these challenges, this research undertakes a comprehensive exploration of diverse facets within the exoskeleton glove system domain. This encompasses a detailed focus on mechanical design, control strategies, and human-machine interaction. To address wearability and comfort, two distinct exoskeleton glove variations are devised, each rooted in different mechanisms. An innovative data-driven model-free force predictive control policy is posited to enable accurate force regulation. Rigorous clinical experiments are conducted to meticulously validate the efficacy of the system. Furthermore, a novel mechanism is seamlessly integrated into the design of a new low-profile exoskeleton glove, thereby augmenting wearability and comfort by minimizing size and weight. A deep reinforcement learning based control agent, which is trained within a simulation environment, is devised to facilitate real-time autonomous force planning. In summary, the overarching objective of this research lies in rectifying the limitations inherent in existing exoskeleton glove systems. By spearheading advancements in mechanical design, control methodologies, perception capabilities, and wearability, the ultimate aim is to substantially enhance the functionality and overall efficacy of these devices within the sphere of assistive applications.

Soft robotic sensors (SRSs) are a class of pliable, passive sensors which vary by some electrical characteristic in response to changes in geometry. The properties of SRSs make them excellent candidates for use in wearable motion analysis technology. Wearable technology is a fast-growing industry, and the improvement of existing human motion analysis tools is needed. Prior research has proven the viability of SRSs as a tool for capturing motion of the foot-ankle complex; this work covers extensive effort to improve and ruggedize a lab tool utilizing this technology. The improved lab tool is validated against a camera-based motion capture system to show either improvement or equivalence to the previous prototype while introducing enhanced data throughput, reliability, battery life, and durability.

Human activity recognition based on wearable sensors’ data is quite an attractive subject due to its wide application in the fields of healthcare, wellbeing and smart environments. This research is also focussed on predictive performance comparison of machine learning algorithms for activity recognition from wearable sensors’ (MHEALTH) data while employing a comprehensive process. The framework is adapted from well-laid data science practices which addressed the data analyses requirements quite successfully. Moreover, an Analysis Tool is also developed to support this work and to make it repeatable for further work. A detailed comparative analysis is presented for five multi-class classifier algorithms on MHEALTH dataset namely, Linear Discriminant Analysis (LDA), Classification and Regression Trees (CART), Support Vector Machines (SVM), K-Nearest Neighbours (KNN) and Random Forests (RF). Beside using original MHEALTH data as input, reduced dimensionality subsets and reduced features subsets were also analysed. The comparison is made on overall accuracies, class-wise sensitivity and specificity of each algorithm, class-wise detection rate and detection prevalence in comparison to prevalence of each class, positive and negative predictive values etc. The resultant statistics have also been compared through visualizations for ease of understanding and inference. All five ML algorithms were applied for classification using the three sets of input data. Out of all five, three performed exceptionally well (SVM, KNN, RF) where RF was best with an overall accuracy of 99.9%. Although CART did not perform well as a classification algorithm, however, using it for ranking inputs was a better way of feature selection. The significant sensors using CART ranking were found to be accelerometers and gyroscopes; also confirmed through application of predictive ML algorithms. In dimensionality reduction, the subset data based on CART-selected features yielded better classification than the subset obtained from PCA technique.

The purpose of this thesis was to determine the proportion of participants enrolled in a community-based weight loss program that would accept and use a wearable device (Fitbit) if included as part of the program. A sample of 526 newly enrolled, adult, female weight loss program participants (BMI ≥ 30 kg/m2 ) were recruited. Participants were randomized to either a Fitbit experimental condition or no-Fitbit control condition, and received emailed information on program features. The experimental condition email also included a free Fitbit offer. The full sample (n=526) was 44±12.6 years old with a BMI of 37±6.2 kg/m2. The proportion of experimental sample (n=266) that accepted and synced was 50% and 23%, respectively. Twenty-two participants in the control condition (8%) also independently obtained and synced a Fitbit. Ninety-nine percent passively declined (did not respond to request for Fitbit color and size information). Those that declined were older (46±13.4 vs. 42±11.3 years of age, p=.001) and weighed less (214±38.9lbs. vs. 231±41.3lbs., p=.01) than those who accepted. Those in the experimental sample who synced were younger (42±10.0 vs. 45±13.2 years of age, p=.012), and weighed more (237±45.2lbs. vs. 217±38.1lbs., p=.002) than those who accepted but did not sync. This thesis provides preliminary support that 23% of participants will accept and sync a free wearable device. These data can be used for decision making, combined with effectiveness and cost data, and research on wearable activity trackers and community, incentive, and web-based weight loss. / Master of Science

This paper is a study of the aspects that are important to consider when designing an unobtrusive interaction for wearable devices in the area of extreme sports. The work is based on an analytical study of seemly-unrelated areas with the common facet that they all call for an unobtrusive interaction in their devices. The findings of the analysis conclude that it is necessary a change of direction in the design because it is not possible to design an unobtrusive interaction that relies on active manipulation. Activity theory and affective computing present theoretical principles with the potential to be used as a framework for HCI and solve the mentioned issues. Finally, I design the user interface of a specific case in the areas of mountain biking and skiing to use it as design-oriented research. An essential aspect of this case is the use of expert feedback and video simulations to drive the design process. Another important point is the definition of the situations and variables that will be observed by the system to adapt itself so it is able to continue being unobtrusive and helpful through the changes.

With the rapid growth and use of wearable devices over the last decade, the advantages of using portable wearable devices are now been utilised for day to day activities. These wearable devices are designed to be flexible, low profile, light-weight and smoothly integrated into daily life. Wearable transmission lines are required to transport RF signals between various pieces of wearable communication equipment and to connect fabric based antennas to transmitters and receivers; the stitched transmission line is one of the hardware solutions developed to enhance the connectivity between these wearable devices. Textile manufacturing techniques that employ the use of sewing machines alongside conductive textile materials can be used to fabricate the stitched transmission line. In this thesis the feasibility of using a sewing machine in fabrication of a novel stitched transmission line for wearable devices using the idea of a braided coaxial cable have been examined. The sewing machine used is capable of a zig-zag stitch with approximate width and length within the range of 0-6 mm and 0-4mm respectively. The inner conductor and the tubular insulated layer of the stitched transmission lines were selected as RG 174, while the stitched shields were made up of copper wires and conductive threads from Light Stiches®. For shielding purpose, the structure is stitched onto a denim material with a conductive thread with the aid of a novel manufacturing technique using a standard hardware. The Scattering Parameters of the stitched transmission line were investigated with three different stitch angles 85°, 65° and 31° through simulation and experiments, with the results demonstrating that the stitched transmission line can work usefully and consistently from 0.04 to 4GHz. The extracted Scattering parameters indicated a decrease in DC loss with increased stitch angle and an increase in radiation loses, which tends to increase with increase in frequency. The proposed stitched transmission line makes a viable transmission line but a short stitch length is associated with larger losses through resistance. The DC losses observed are mainly influenced by the resistance of the conductive threads at lower frequencies while the radiation losses are influenced by the wider apertures related to the stitch angles and increase in frequency along the line. The performances of the stitched transmission line with different stitch patterns, when subjected to washing cycles and when bent through curved angles 90° and 180° were also investigated and results presented. Also, the sensitivity of the design to manufacturing tolerances was also considered. First the behaviour of the stitched transmission line with two different substrates Denim and Felt were investigated with the results indicating an insignificant increase in losses with the Denim material. Secondly, the sensitivity of the design with variations in cross section dimensions was investigated using numerical modelling techniques and the results showed that the impedance of the stitched transmission line increases when the cross sectional dimensions are decreased by 0.40mm and decreases when the cross sectional dimensions are increased by 0.40mm. Equally, repeatability of the stitched transmission line with three different stitch angles 85°, 65° and 31° were carried out. The results were seen to be consistent up to 2.5GHz, with slight deviations above that, which are mainly as a result of multiple reflections along the line resulting in loss ripples. The DC resistance of the stitched transmission line with three different stitch angles 85°, 65° and 31° corresponding to the number of stitches 60,90 and 162 were computed and a mathematical relationship was derived for computing the DC resistance of the stitch transmission line for any given number of stitches. The DC resistance computed results of 25.6Ω, 17.3Ω and 13.1Ω, for 31°, 65° and 85° stitch angles, indicated an increase in DC resistance of the stitch with decrease in stitch angle which gives rise to an increase in number of stitches. The transfer impedance of the stitched transmission line was also computed at low frequency (< 1GHz) to be ZT=(0.24+j1.09)Ω, with the result showing the effectiveness of the shield of the stitched transmission line at low frequency (< 1GHz).

The market for wearable devices is continuously growing and has seen an in- crease in interest and demand this past year, specifically smartwatch devices. With several big players entering and trying to take place in the market the number of devices and platforms grow. This leads to device and software fragmentation like the one seen in the world of smartphones. In this paper I discuss and compare the two smartwatch platforms Android Wear and Apple Watch in terms of possibilities, limitations and differences. Research is done to find cross-platform development possibilities for these platforms. Extensive theoretical background of both APIs is researched and presented. An app for both smartwatch platforms is developed with integration of the WebSocket protocol to function as a remote control for a Video-On-Demand web service. This is done to showcase the cross-platform possibilities and differences of the platforms. As a result the biggest differences are out- lined and a conclusion is made that cross-platform development for these platforms can be challenging but is possible on certain levels.

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

<p>  </p> <p>High-resolution enclosed channels/fibers are highly demanded by different disciplines such as microfluidic channels for chemical synthesis, bioreactors for drug metabolism, magnetic locomotor for drug delivery, and wearable devices for motion detection. However, the current fabrication techniques for enclosed channels/fibers are restricted to a few millimeters in size. Their manufacturing often involves time and energy-consuming multi-step processes with insufficient resolution. In this work, we demonstrate a novel co-flow-enabled fabrication method to resolve the technological restrictions in the fabrication of high-resolution enclosed channels/fibers with efficient production time, controllable morphologies, and high throughput manner.</p> <p>An epoxy-based enclosed microfluidic channel was first built. A non-reactive paraffin oil and a liquid resin were pumped into a 3D-printed co-flow generator and worked as core and shell fluids, respectively. The epoxy resin was cured by external heat stimulus. As a result, the reaction region was limited between the generator wall surface and the boundary of core flow, eliminating the need for precise control over the curing system. The experiment was successfully conducted to cure build resin channel inside copper and resin tubes with good shell thickness.</p> <p>Conductive hollow hydrogel microfibers were also fabricated by this method. Sodium Alginate and Calcium Chloride were chosen as the shell and core flows, respectively. The ionic crosslinking happens at the boundary of two flows and expands outwards across the radial direction. Thus, the diameter of the hollow channel can be easily adjusted by tuning the flow rate and the size of the core flow injection needle. PEDOT: PSS, a conductive polymer, was mixed with Sodium Alginate to impart fibers with excellent electrical conductivity. The synthesized hollow microfibers have shown their functionality in stretching movement detection by serving as a fundamental building element of motion sensors. </p>

<p>  </p> <p>In this study, we present advancements in smart contact lenses, highlighting their potential as minimally or non-invasive diagnostic and drug delivery platforms. The eyes, rich in physiological and diagnostic data, make contact lens sensors an effective tool for disease diagnosis. These sensors, particularly smart contact lenses, can measure various biomolecules like glucose, urea, ascorbate, and electrolytes (Na+, K+, Cl-, HCO3-) in ocular fluids, along with physical biomarkers such as movement of the eye, intraocular pressure (IOP) and ocular surface temperature (OST).</p> <p>The study explores the use of continuous, non-invasive contact lens sensors in clinical or point-of-care settings. Although promising, their practical application is hindered by the developmental stage of the field. This thesis addresses these challenges by examining the integration of contact lens sensors, covering their working principle, fabrication, sensitivity, and readout mechanisms, with a focus on monitoring glaucoma and eye health conditions like dry eye syndrome and inflammation.</p> <p>Our design adapts these sensors to fit various corneal curvatures and thicknesses. The lenses can visually indicate IOP through microfluidic channels' mechanical deformation under ambulatory conditions. We also introduce a colorimetric hydrogel tear fluid sensor that detects pH, electrolytes, and ocular surface temperature, indicating conditions like dry eye disease and inflammation.</p> <p>The evaluation of these contact lens sensors includes in vivo/vitro biocompatibility, ex vivo functionality studies, and in vivo safety assessments. Our comprehensive analysis aims to enhance the practicality and effectiveness of smart contact lenses in ophthalmic diagnostics and therapeutics.</p>