9:00 AM PDT Breakout 11: Engineering Panel D

Friday, July 30 9:00AM – 10:00AM

Location: Online via Zoom

The Zoom event has ended.

Seth Caines
University of Nebraska–Lincoln
Presentation 3
Are microplastic fibers that are usually found in textiles prevalent in Nebraska freshwater bodies?
Microplastics are increasingly being detected in freshwater systems and can have negative impacts on the environment including growth limitations of aquatic organisms (G. Chen, Y. Li, and J. Wang, 2021), degraded water quality through the sorption of inorganic pollutants and heavy metals (F. Yu et al., 2019), and a reduction in the dewaterability of activated sludge (J. Xu, X. Wang, Z. Zhang et al., 2021). Although microplastics have been detected in terrestrial and freshwater systems, we do not have a comprehensive understanding of the quantities and types of microplastics that are in freshwater environments. In this study, microplastics were collected from freshwater systems upstream and downstream of wastewater treatment plant discharges, as well as freshwater systems surrounded by both agricultural and suburban land use. The water column was sampled using grab sampling and net sampling approaches. Sediment was also sampled. Other conventional water quality parameters were collected including nitrate, phosphate, total solids, and total suspended solids. We hypothesize that there is a higher abundance of microplastics in freshwater systems impacted by wastewater treatment plant effluent and the most common type of microplastic that will be found will be either polyester, polyamide, or polypropylene, since those are microplastics that are commonly found in textiles. Our study will provide a better understanding of the source of microplastics in freshwater bodies in Nebraska as well as the dominant plastic types that occur in suburban and agricultural areas.
Alek Sepulveda
The University of Arizona
Presentation 1
Learning an Unwalked Path: Machine Learning Models for Predicting Pedestrian Trajectory
The number of pedestrian accidents, both fatal and non-fatal, continues to increase. As mobile compute power increases, the practicality of smartphones in helping this issue improves. Previous research in reducing pedestrian accidents has explored using embedded smartphone sensors with algorithms to improve the accuracy of pedestrian position and trajectory measurements, providing a warning to both the pedestrian and surrounding vehicles. In order to improve the effectiveness of these warnings, this paper investigates the effectiveness and viability of various models for predicting pedestrian trajectory using mobile phone sensors. With pedestrian motion being individualistic in nature, we present a diverse set of time-series datasets with natural raw motion and location data. This data is collected on a university campus to simulate the potential paths of university instructors and students. Several machine learning methodologies are employed and analyzed for accurately predicting the trajectory of a pedestrian across several prediction times using the data collected. The purpose of this machine learning study is to determine an appropriate model architecture to predict pedestrian trajectory both in accuracy and viability in terms of modern mobile computing power. Although this study is ongoing and we have not concluded our data analysis, we hope that our findings support future traffic system and Pedestrian Trajectory Prediction System (PTPS) research.
Noah Garcia
University of Nebraska–Lincoln
Presentation 2
Designing a Single-Leg Robotic Exoskeleton for Hemiparesis Patient Gait Assistance
The purpose of this study is to design a single-leg robotic exoskeleton for hemiparesis stroke patients in order to assist disabled individuals in performing natural gait patterns. Hemiparesis is a partial loss of muscular strength on one side of the body most commonly brought about by strokes, a leading cause of disability worldwide. Assistive devices such as canes and walkers are commonly used among stroke survivors, but they each carry serious drawbacks relating to asymmetric gait, safety, and ease of use; robotic exoskeletons may eliminate these drawbacks by directly aiding the patient in completing symmetrical gait patterns. Although there are many existing exoskeleton designs for augmenting human strength and other useful functions, this study’s design uniquely focuses on correcting hemiparesis patients’ gait while providing additional torques that help the patient propel forward and extend their leg, achieving a full range of motion. Existing exoskeleton and exosuit designs were examined, and literature regarding actuation technologies was explored to gauge feasibility related to the design intent. Using the information gathered, a design concept that most effectively satisfied the design’s functional requirements was synthesized. A 3D model of the design was then constructed using SolidWorks, and a dynamic analysis was performed to test various components of the design.