Hybrid dynamical systems (HDS) are time dependent systems that have both discrete and continuous behavior, such as a bouncing ball or jumping leg. The BioRobotics Lab studies HDS in order to better understand locomotion and develop controls for robotics. As controllers are developed, they need to be tested in the hybrid dynamical system (HDS) either through simulation or implementing in physical robotics. While simulated system environments are useful in early controller development, they are unable to capture all the intricacies of a physical system. Thus, testing controllers in a physical system can provide better insights for future development. We developed a hopping robot for the purpose of easily implementing controllers through software. The hopping robot is designed to be modular and easily configured for different dynamics. Simulations of the robot were also developed so that researchers can use both simulated and physical environments. This tool will help researchers develop new controllers and could bring new insights to studying hybrid dynamical systems.

Rehabilitation research has demonstrated the benefits of early powered mobility intervention for children with disabilities, from improving developmental skills to empowering children to better participate in family and community life. However, traditional powered mobility devices are often stigmatizing, costly, and require specialized transportation due to their size and weight. One alternative early powered mobility option that has sought to address several of these concerns, The Go Baby Go Mobility and Socialization Project, provides children with disabilities a means of socially welcoming, early independent mobility experiences through custom safety and accessibility modifications to commercially available toy ride-on cars. The Go Baby Go project has resulted in promising pilot research along with community-based outreach and education in collaboration with families, clinicians, and engineers. However, a means to efficiently track car performance in real-world environments without the presence of a researcher has been lacking. Therefore, the purpose of our project was to develop a customized data logger and companion Arduino code with the ability to collect real-time data from the cars as families use them in their homes and communities. Our multidisciplinary team has created and implemented a system which gathers car performance data automatically via integrated sensors, including the number and duration of switch activations, frequency and duration of use, outdoor location, distance traveled, and driving terrain. Housed in a simple, waterproof food storage container, the data logger is integrated into the car’s electronics and powered by the car battery, with data stored on a micro-SD card. Preliminary analysis of results from eight cars in local communities is ongoing and will be shared; initial feasibility of our system for real-world tracking without undue research presence or caregiver reporting burden is promising. Future research goals include full quantitative analysis of car use patterns to improve technology design and implementation in community settings.

Exposure to particulate matter (PM) is associated with adverse health effects including asthma, cardiopulmonary diseases, and lung cancer. These adverse health effects are responsible for the premature death of 7 million people worldwide each year. To reduce the consequences of air pollution, real-time individual exposure data is essential for health risk assessment and tailored intervention strategies. Portability and cost are significant barriers for consistent and personalized tracking. Unfortunately, the only accurate data regarding levels and types of air pollutants come from static monitoring stations that do not allow personalized analysis. Aero-Spec was developed to provide real-time analysis of air quality particles while also delivering a compact and low-cost approach. It is a wearable exposure monitor that combines real-time measurements of PM using an optical particle sizer (OPS). The OPS sensor allows for individual particle sizing and error reduction associated with the particles’ optical properties. Personal PM exposure data is provided to the user through a smartphone application such that intervention actions could be triggered given the detection of poor air quality. For personalized analysis, indoor localization is used over the global positioning system (GPS) for a more accurate and localized diagnosis. Aero-Spec utilizes this network technology to track the wearable sensors and report back localized air particle data. In conclusion, using a portable and low-cost approach to real-time air particle analysis, Aero-Spec delivers an effective and personalized solution for air quality monitoring. Recently, in a movement to monitor their employees’ exposure to particle data in real-time, Boeing has collaborated with the Sensors, Energy, and Automation Laboratory (SEAL) for further research.

Fossil Fuels are associated with contemporary energy crises and climate change. The combustion of fossil fuels is leading to increased greenhouse gas emissions, which in turn have increased the overall earth’s temperatures and is predicted to grow at an alarming rate. Fuel cells are alternative, sustainable sources of energy that uses hydrogen (or hydrogen-rich fuel) and oxygen to generate electricity through electrochemical processes. I conducted a survey, focused on Bellevue College’s (BC) Chemistry Department, that indicated broad support for a simple hydrogen proton exchange membrane (PEM) fuel cell lab to be incorporated into the introductory chemistry curriculum. I will design a lab that will spark student interest in sustainability and expose students to real-world electrochemistry applications while addressing electrochemical, thermodynamic, transport phenomena, and clean energy concepts. The educational goals of this lab are to promote a deeper conceptual understanding of electrochemistry, to improve quantitative reasoning, and to improve explanations of observed scientific phenomena. I collaborated with BC’s Chemistry Department to determine learning outcomes and a systematic process to quantifiably assess fuel cell labs from other institutions. This information was used to design an effective lab and lesson plan surrounding fuel cells. Four fuel cell labs were evaluated: (1) A pre-constructed Hydrogen PEM Fuel Cell from Horizon Fuel Cell Technologies (2) A microbial fuel cell, (3) A fuel cell using platinum electrodes that is bathed in an acid solution (4) A fuel cell using graphite electrodes that is immersed in an acid solution. This research produced an economical and introspective laboratory experience that utilized basic laboratory equipment and materials. The results were presented in an engineering framework that details how aspects of the lab promote critical thinking and engagement, addresses learning objectives, and was cost-effective.

Wireless power transfer has many applications, from powering biomedical implants to wireless sensors. For more practical use, however, several challenges must be overcome, such as a lack of efficiency and power leakage to other nearby electronics. These issues become especially difficult to tackle with a moving receiver. To combat these problems, an array of magnetically coupled coils was designed. Previous work has shown the capabilities of this system when one coil of the array is supplied with power. In this work, I explore the possible benefits of having two coils in the array driven with power instead, studying the interaction between the different coils. By adjusting parameters such as the phase relationship between the two transmitters’ signals, we aim to optimize power delivery to specific targets and simultaneously minimize leakage to other areas. I tested different configurations of the system in a series of experiments and analyzed measured data to determine which setup is most favorable. Afterwards, I evaluated the efficiency of the configuration compared to the previous single-transmitter case. This provides better insight into how the coils in the array magnetically interact with one another, which will inform future design decisions. It can also eventually lead to better solutions for delivering high power to selected targets within a given space. This can create flexible, efficient, and safe wirelessly charged electronic implants for a variety of biomedical applications, enabling further research in the field and the development of novel solutions.

Reinforcement Learning (RL) is a subcategory of machine learning, in which an agent (the decision maker) observes its environment and executes the best course of actions to maximize rewards. This is similar to teaching a pet to perform tricks using treats as positive reinforcement. Our research compares different RL methods on low-performance devices like a Raspberry Pi in real-time, real-world environments. RL has gained popularity recently with breakthroughs from DeepMind’s paper, Playing Atari with Deep Reinforcement Learning, where an agent learns to play Atari games from raw pixels and from DeepMind’s AlphaGo (DeepMind, https://deepmind.com/research/alphago) program that was the first computer program to beat a world champion Go player. RL projects like AlphaGo have utilized big data, powerful computing resources, and simulated environments that do not require real-time interaction to train the machine learning models. Our group compares the effectiveness of different RL methods on an accessible level of computing power on offline devices that an average consumer could acquire. The team constructs a physical environment for the robot to navigate, creates an OpenAI Gym environment that our agents will use to control the robot and get feedback from the environment. We train our agents using different RL methods to optimally navigate the environment and avoid collisions. We then compare the performance of the different methods in our physical real-time environment. Reinforcement Learning in small, offline devices could pave the way for a variety of devices that learn over time without being connected to a network. Imagine a small Mars rover that learns to navigate its environment efficiently over time.

With the increased development of space programs globally, more space probes are being funded for sample-return missions. Versatility and power limitations are some of the critical issues that these extraterrestrial sampling rovers are seeking to overcome in order to reliably deploy autonomously. Exploring the rough and unpredictable surfaces of other celestial bodies requires more adaptable and energy efficient robotics, which is what makes bio-inspired origami structures so appealing. Leaf-like origami is reconfigurable, which means that it can walk, jump, grasp, and actuate other useful motions all within the same device. Typical robots have redundant actuators and structural systems, but origami devices can build up potential energy before converting it into a mechanical motion. Due to this, rigid origami design approaches have the potential to be more compact, versatile, and energy efficient than conventional devices. Studying reconfigurable origami-based robotics can lead to devices that can be transformed into multiple configurations for various tasks. To design a versatile origami-based structure, I followed up on the research that Professor Jinkyu Yang and the members of his research group had on leaf-like origami, specifically leaf-out origami. This origami structure shows multi-transformable features. This means that the structure can be configured in a stable-stored or stable-deployed shape without having an external power supply maintaining its configuration. By finalizing the design, fabrication method, and discovering the optimized folding patterns for jumping and grasping motions, we will be able to start implementing the structure into more complex systems. Designing more structurally efficient systems may be the most practical solution we currently have to combat power and versatility limitations for autonomous space probes.

Natural disasters pose a great threat to power systems. They can cause major damage to power lines and substation equipment, resulting in widespread outages. It can then take days to weeks and millions of dollars to restore power to customers. Power system resilience is concerned with reducing the impact of disasters on power systems. The purpose of this study is to evaluate the resilience of one power utility in southwest Washington, the Grays Harbor PUD (GHPUD), to earthquakes, earthquake liquefaction, and tsunamis. This study will consider the likelihood of these disasters, potential effects on the GHPUD’s electrical system and its customers, and possible mitigation techniques. I have collected seismic hazard curves, response spectra, and liquefaction susceptibility data at GHPUD substation sites from US Geological Survey maps and the M9 project. I have also collected data on tsunami inundation depth at substation sites from Washington Geological Survey maps. With this information I have determined which of their substations are most at risk from these disasters. This data will be used to determine possible impacts on substation equipment and to estimate the probability of equipment failure. When completed, the results of this study can allow the Grays Harbor PUD and other power utilities to be more prepared for future disasters and to be more informed of potential resilience improvements.

Quantum sensing applications require diamonds with high concentrations of high-fidelity NV centers. Here we find we can significantly increase the NV center density in high-purity diamond by over a factor of ten by simple annealing. We perform fifteen anneals starting with 800 °C up to 1100 °C. Throughout each anneal, we track thousands of individual NV centers in a large experiment volume (350x350x500 um³) using a custom confocal microscope. Peak NV density was observed to occur at 980 °C. Spectroscopy measurements also show near ideal NV quantum characteristics for the newly formed centers. With this process, we can optimize NV formation for magnetic field sensing and quantum entanglement applications.

Chemically induced dimerization (CID), in which two proteins dimerize only in the presence of a small molecule, has been widely used to control cell signaling, regulatory, and metabolic pathways, and used as logic gates for biological computation in living mammalian cells. However, few naturally occuring CID systems and their derivatives are currently available. Creating a CID system with desired affinity and specificity for any given small molecule remains an unsolved problem for computational design and other protein engineering approaches. To address this challenge, we have used a novel strategy to select CID binders from a vastly diverse combinatorial nanobody library. We have created new CID systems that can sense cholecalciferol and artemisinin. We are validating CID biosensors by a yeast three-hybrid system and built structural models to understand the small molecule-induced dimerization. Our work is a proof-of-concept that can be generalized to create CID systems for many applications.

In this research, we looked at the possible use of an Air Breathing Pulsed Plasma Thruster (AB-PPT) for propelling an aircraft that could potentially act as an atmospheric satellite at atmospheric altitudes > 25 km. One of the advantages of operating at such altitudes is that the aircraft is not subject to highly variable weather conditions. An atmospheric satellite is an aircraft that can perpetually fly around the same area, and can be launched at significantly reduced cost when compared to conventional satellites while providing similar services such as communication and imaging. At the desired altitude, conventional blade-based propellers are too inefficient due to low background pressure, but it is still too low of an altitude for conventional space propulsion, prompting the use of AB-PPT. The AB-PPT is an adaptation of the conventional PPT which is a pulsed device that uses an electric discharge to ionize and expel the solid propellant such as PTFE at high speeds. For the AB-PPT, a large amount of voltage discharge generated across the two coaxial electrodes ionizes and expels the background air instead. In order to identify the most efficient AB-PPT design, three different configurations were tested using a pendulum-based thrust stand to measure the thrust efficiency, a Rogowski coil to measure the current per discharge, and high-speed video. One configuration has been identified, and we are currently investigating an alternative electric connection to further increase the amount of thrust per discharge. Future work includes plasma diagnostics of the AB-PPT such as Langmuir probe, and B-dot probe, in addition to the development of more robust electronics capable of delivering higher power, and the development of the final version of the AB-PPT for patenting purposes.

The study of the behavior of multiple agents, specifically human and machine, in a dynamic environment is challenging due to the unpredictable individual behaviors. Humans will naturally formulate beliefs about the machine’s behavior, which would directly affect their future decisions. Our research aims to develop a framework for the study of human-machine dynamic interactions. With the imperfect information humans and machines have about each other and their environment, a game-theoretic approach was done to study the natural model of their interactions. We derive theoretical models for steady-state (i.e. equilibrium) and transient (i.e. learning) behaviors of humans interacting with other agents (humans and machines). We also design experiments to validate our theory. A haptic testbench, in the shape of a robotic arm, is used as a dynamic simulation platform for studying the trajectories of the human/machine interaction, allowing us to study both theoretically and experimentally. The robotic arm has a position control system that supports a wide variety of human/machine experiments. The user is provided with visual and haptic feedback, which allows for experiments to be designed to study the sensorimotor learning processes. The robotic arm is built using direct-drive brushless motors, force sensors, an open-source ODrive motor controller, and an arm lever. The motor firmware is designed in C/C++, and integrated with a user-interface in Python. With the wide variety of potential applications, we hope our research will give insights into the different natures of human motion and be a fundamental platform for technological breakthroughs in the medical field.

Dementia is a neurological condition that impairs memory, cognitive functions, and the performance of daily activities. It is estimated that 5.5 million Americans suffered from dementia in 2018. Currently, the predominant method used to treat persons with dementia is through pharmacological means. However, through reviewing current research, we have found evidence to suggest that cognitive behavior therapies (CBT) may be more beneficial to patients, with the potential to reverse some symptoms of dementia. Our study aims to improve CBT through virtual environments. Virtual Reality (VR) is not merely a visual platform but an immersive environment that stimulates the brain in the same way real environments do. However, in VR we can track certain markers like head and eye movement, metrics that can be valuable in tracking the progression of the disease as well. We have developed a virtual experience that can calm and stimulate seniors while being able to track a person’s ability to perform free recall and spatial recognition through a series of memory games. This experience has the potential to have both therapeutic and diagnostic capabilities, as well as provide insight into how VR can affect memory. Over the past 6 months, our research team has worked with seniors at the Greenwood Senior Center to test different virtual reality platforms such as the Oculus Go and HTC Vive, as well as various virtual environments. Using our experience from the center our team is creating a platform tailored to fit seniors who are affected by dementia. In the near future, our team will be volunteering at a local assisted living facility to work with these seniors. This project will offer understanding of how implementing VR therapy can improve patients’ morale and cognitive capabilities, as well as exploring the use of VR as a diagnostic tool.

Roborun is a simulation game that utilizes crowdsourcing to identify legged movement patterns for land-based robotic movement. By using a game controller, keyboard, or by providing a set of instructions to execute in order (e.g. “rotate front leg X degrees”, “move back leg forward Y meters”, etc.), players can control joint torques and leg movements in order to navigate a virtual two-legged robot through several 2D obstacle courses that contain varied terrain and movable boxes. The game can be played through a web browser on either a computer or smartphone, making the game accessible to players across multiple platforms. In future development, we intend to implement a scoring system based on efficiency and speed of course solutions and will replicate the best scoring solutions from players on a commercially-available robot in our testing laboratory. It is our hope that their solutions will help develop reliable robotic movement algorithms and shed some insight into the dynamics of land-based movement.