Finite Element Analysis, Representation, and Interpretation of Soft Robotics Kinematics and Dynamics

Rapid progress has been made in recent years to improve the accuracy, precision, and intelligence of robots as humans seek to use them to make their lives more comfortable and convenient. Robots are increasingly incorporated into our daily lives, as well as into industry and manufacturing, causing the market demand for them to grow. Despite the number of tasks robots can successfully perform, traditional rigid robots nonetheless have the potential to cause harm to humans and their immediate environment, creating safety concerns. These risks have led to the development of a relatively new field of robotics in which soft materials are employed to limit the possibility of damage and injury. Unlike rigid robots, soft robots have more degrees of freedom and the ability to adapt to their environments, allowing for a wider range of motion and tasks they can undertake. Unfortunately, analyzing the motion and the ability to apply forces of soft robots is challenging, primarily due to their nonlinear behavior and properties. For this project, we used finite element modeling techniques to explore the
practicability of two types of soft robotic actuators: Fiber-Reinforced Elastomeric Enclosures
(FREEs) and McKibben actuators. A comparison of experimental and finite element results confirmed the accuracy of the model and allowed the workspace achieved with a module comprised of multiple FREEs to be studied. Furthermore, we were able to establish a more effective soft robotic design by considering the role of each system parameter, maximizing the range of displacement and rotation of the actuators.

Trajectory Gaze Path Analysis and Isolating Areas of Interest in Eye-Tracking Data for Autism Spectrum Disorder Studies

According to the CDC, about 1 in 59 children have autism spectrum disorder or ASD, representing a significant percentage of the population. Unfortunately, this condition often remains undiagnosed until later in childhood, which, in turn, delays many clinical treatments that could improve social functioning outcomes.

Researchers have identified abnormal visual attention as a hallmark symptom of ASD. With this finding, ASD researchers commonly deploy eye-tracking systems in their experiments. A typical experimental setup assesses how participants look at objects encapsulated within one static image or a stimulus. Eye-tracking systems collect real-time gaze data over a short, fixed time period. ASD experts have found that children without ASD generally focus more on objects associated with socializing, such as people or food items, than on inanimate objects. In contrast, children with ASD tend to focus on both categories of images with no preference. Heat maps, currently used in the clinical setting, forgo clinically crucial information about how children cognitively prioritize stimuli over time.

To better understand the cognitive process for prioritization of stimuli between children with and without ASD, clinical researchers need novel methods that yield visuals that show how participants prioritize stimuli over time. My work under the guidance of Dr. Brian King is developing multiple novel algorithms intersecting between the computer science fields of data mining and machine learning, including density-based clustering, object detection, and image classification. Further, we make these data visualization algorithms accessible to end-users through an interactive graphical user interface (GUI) encapsulated within a software toolkit.

Measuring Lower Limb Muscle Activity and
Kinematics in Variable Foot Strike Gaits

Anterior knee pain affects roughly 23% of adults and 29% of adolescents, and many cases go
untreated. Prior research has aimed to identify underlying causes of knee pain,
and while exact causes can be unknown for individuals, differences in muscle activity, gait
patterns, morphology, and loading are key contributors. To better understand links between
muscle activity and kinematics, we aimed to measure changes in surface electromyography of knee extensor muscles and others as a result of different gait patterns. A total of twenty subjects underwent surface electromyography measurements in the Bucknell motion analysis biomechanics lab and with the use of non-invasive surface EMG sensors that measure muscle activation. Specific activities and gait patterns include normal walking, toe-in/toe-out walking, heel-strike/toe strike, and normal running. Sensors were placed on the subject’s vastus medialis and lateralis, quadriceps, hamstring, and medial/lateral gastrocnemius. Following data collection, data processing included rectification, high/low pass filters, root mean square and moving envelope calculations, and normalization to maximum voluntary contraction EMG. Statistically significant patterns were identified in EMG profiles both intra-subject and between subjects, with the Vastus Medius and Vastus Lateralis showing the most variation in activation, and toe-in/toe-out walking showing the greatest activation. In several subjects, the activation profile of both the Rectus Femoris/Hamstrings and Gas Med/Gas Lat were not statistically different from themselves but were different than the Vas Med/Vas Lat ratio. The next step is to discuss clinical relevancy and how our data can inform pain prevention.

ProPANE Notetaking Assistive Technology
The Propane project is a collaborative research project between Electrical & Computer Engineering and Education at Bucknell University. Our goal is to develop an assistive technology that will support note taking for college students with learning disabilities (LD) and English Language Learners (ELLs) in the lecture-based classroom to improve their content learning and academic performance. While research has shown that effective note taking leads to better performance and content mastery, students with LD and ELLs may struggle with this task in lecture-based classrooms. The purpose of the project is to reduce students’ cognitive load and free students’ working memory space to absorb lecture content.
Our approach is to create a smartphone application that will be used by the student to capture the lecture. The student will submit a video of the lecture and the application will use various image processing techniques and segmentation algorithms to extract the key information from the lecture. The extracted information is returned to the student for them to further review and or annotate. The extraction of the targeted information is a unique aspect of this project and presents major technical challenges but can potentially support effective and efficient note taking. By allowing the student to focus on the lecture and not note taking, it allows them to engage in other ways and use their working memory space on activities and discussion.

Visual Characterization of Aponeurosis Microstructure

Aponeurosis is a tendinous sheath-like tissue found in many muscle-tendon units that covers the muscle belly and transitions into tendon. Little research has been done to understand how collagen fiber microstructure contributes to aponeurosis stiffness and mechanical function. The goal of this study was to use scanning electron microscopy (SEM) to characterize the microstructure of aponeurosis tissue be comparing waviness values measured in unstretched and stretched tissue. It is hypothesized that the waviness of the collagen fibers will decrease after tissue has been stretched. Porcine shoulder tissue was dissected to obtain 40×10 mm specimens. Unstretched samples (n=10) were fixed in 10% formaldehyde, while stretched samples (n=10) were fixed at five percent strain. Images were taken at 50, 100, 1k, and 3.5k magnification. Waviness of the stretched and unstretched aponeurosis was quantified as the ratio of the true length of the collagen fiber to the tangent of two endpoints of that same collagen fiber. Aponeurosis exhibited a hierarchical structure, similar to that of tendon or ligament, with collagen-rich fascicles, fibers, and fibrils. Waviness in the collagen fibers was observed at lower magnifications (100 μm), while at higher magnifications the sheet-like structure of the collagen fibrils was seen (1,000 μm). Unstretched tissue exhibits a high degree of waviness or collagen crimp (1.17 ± 0.21) compared to tissue that has been fixed under a five percent stretch (1.05 ± 0.06, p = 6.615e-08). Future work will include using computational modeling to study the effect of collagen structure on aponeurosis mechanics.

Development of Guidelines for Playground Surfacing Based on Field Testing

Many different playground surfaces are used in the United States for fall-related injury protection. These surfaces are categorized as unitary or loose-fill and there is a strong interest in quantifying their performance to ensure they are meeting head-injury safety standard ASTM F1292, which dictates a compliant head injury criteria (HIC) metric of less than 1000.

Analysis for this study was performed from a randomized sample of 103 public playgrounds across the United States, which was collected by the National Program of Playground Safety on behalf of the United States Consumer Product Safety Commission (CPSC). Field testing data included analysis of safety surfacing, including fall height, depth of the surface material, and resulting head injury criteria (HIC) score to interpret performance trends in ASTM-compliant HIC scores.

The results indicate for all materials that as the fall height increases, the surfacing’s performance decreases. A similar trend applies for surface depth—as the surface depth decreases, the surfacing’s consistency decreases. While all the surfaces display similar overall trends, there are a few differences as well. Some surface materials appear to reach a fall height at which performance decreases regardless of surfacing depth. Other surfaces demonstrate a linear relationship between fall height and surface depth. Preliminary data tables showing the relative performance of surface depth to specific fall height ranges were developed for each surface material. From these preliminary data tables, recommendation tables were created as a cost-efficient alternative to using an impact testing device.

Characterizing Anterior Knee Mechanics through
Patellofemoral Joint Modeling

Patellofemoral pain affects nearly 25% of the general population, and is particularly prevalent in athletes and military personnel. It remains unclear as to why some people experience patellofemoral pain and others do not, however joint mechanics certainly play an integral role in pain development and disease progression. The mechanics of the patellofemoral joint are driven by multiple factors such as kinematics (the way people move and walk), morphology (the person-specific shape of the musculoskeletal tissues), and the loads that act about the surrounding area of the joint (muscle forces and activation patterns). Broadly, the aim of the research is to develop a computational tool that will help to investigate the contributions of these factors to anterior knee mechanics and thus joint pain longer term. A musculoskeletal model of the patellofemoral joint will be developed in OpenSim – an open source musculoskeletal modeling software – based on the OpenKnee data set, which is freely available. First, segmented MRI scans are used to generate the following geometries: the distal femur, distal femoral cartilage, the patella, patellar cartilage, the proximal tibia, and tendon and ligament insertion/attachment sites. These geometries are then smoothed and refined and imported into OpenSim, an open source musculoskeletal modeling software, with cartilage as contact geometries. Femoral and tibial kinematics will be defined along with knee extensor muscle forces, thus enabling the simulation of patellofemoral joint contact. Long term goals include studying the effects of different loading cases, different patellofemoral joint morphologies, and different kinematics on joint contact pressure.

Electroformed Heat Pipe Heat Exchanger
As the necessity to shift towards renewable forms of energy becomes ever more apparent in the wake of the consequences of climate change, one major barrier that has largely yet to be solved is the problem of energy storage. It is technologically possible to meet our energy demands solely with renewables, and likely many times over. However, when energy is generated is a critical factor to meet our necessary load demands each day and when most renewables are active is not controllable. Many different forms of energy storage already exist but with many issues such as cost, scaling, and environmental impact. We are exploring the development of low-cost energy storage mediums, such as salt, sand, or dirt. However, these mediums have very low thermal conductivity and thus it becomes necessary to synthesize different techniques to accomplish a feasible storage method. Principally, we are investigating the combination of a heat pipe and the topology of a bio-inspired heat exchanger. The highly complex geometries under consideration would possess a high surface area to volume ratio, facilitating conduction in low-conductivity media, but must be fabricated with non-traditional techniques such as electroforming. If we are able to successfully incorporate both of these elements into a thermal energy storage, it would lead to the cheap and economical storage of energy allowing us to justify larger scale use of renewable energy within our power grid.

Deep Brain Stimulation of the Subthalamic Nucleus and its Effect on Gait in Parkinson Disease

Deep brain stimulation (DBS) is a surgical procedure where electrodes are implanted in the brain to modulate specific regions with electricity. This stimulation can alleviate gait disturbances in Parkinson disease (PD). DBS modeling can be used to estimate the spatial extent of stimulation, enabling individualized treatment. Although the standard DBS target for PD is the subthalamic nucleus (STN), a generalized approach may not be optimal for every patient due to the diversity of their symptoms. Better outcomes may be obtained by stimulating regions around the STN.
Forty PD patients who received bilateral STN DBS were included in this study. For each patient, the location of therapeutic stimulation was calculated using tissue activation models built from individualized imaging data and stimulation settings. The volume of tissue activation (VTA) was used to quantify STN and external (non-STN) activation in the lateral-medial, anterior-posterior, and dorsal-ventral directions. The relationship between STN/external activation and symptom improvement (gait, freezing of gait, postural stability, and total gait) was evaluated. A similar analysis was performed for electrode location (the distance between the active contact and STN centroid).
A significant positive relationship between anterior STN activation and total gait improvement was found (p < 0.01). No significant relationships were found for the external activation and electrode location analyses. Results suggest that more anterior STN stimulation may be preferable for patients whose primary symptoms are gait disturbances. Furthermore, VTAs may provide more information about stimulation location than active contacts and highlight the importance of patient- and symptom-specific targeting.

Characterizing Styrenic Triblock Copolymer with Static Light Scattering
Light scattering refers to the process in which a wavelength of light hits a particle and scatters in a pattern. This pattern can be interpreted to allow us to understand the characteristics of said particle. For my research, we used a known measurement technique called static light scattering (SLS) which measures the intensity of this scattering at multiple angles and multiple concentrations. This technique allows us to calculate constants such as the molecular weight, the A2 virial coefficient, and radius of gyration of the scattered particle. Using this information, the measured polymer characteristics are used to explore the relationship between structure and behavior of the polymer. The instrument used for these measurements is the BI-200SM Goniometer. The beginning part of the research project was to assemble, align, and calibrate this instrument for use. As soon as the instrument was prepared, we performed the SLS procedure on multiple polymers supplied by Kuraray America and Kraton Polymers. Using solutions of known concentrations, and by measuring the scattered intensity at multiple angles, we are able to organize the data into a Zimm plot. This Zimm plot is organized such that we are able to easily determine the important constants of the polymers by observing the trendlines of the data. Seventeen different styrenic triblock polymers were characterized using this instrument. A procedure for future measurements using the instrument was also developed. This instrument is currently being used in further research opportunities of dynamic light scattering (DLS) behavior of polymer gels.