I graduated with a B.S. in Materials Science & Engineering at the University of Washington, where I developed strong engineering, science, and manufacturing skills from classwork, student teams, and industry internships. I have four years of CAD and design experience with a focus on design for manufacturing and assembly; here's some of my work.
Advanced Robotics at the University of Washington (ARUW.org)
The Sentry is a fully autonomous robot in the RoboMaster competition with two independent firing turrets mounted on a larger turret that allows the assembly to rotate independently of the chassis. It utilizes a swerve drive system that allows the wheels to rotate in any direction, allowing for agile, omnidirectional movement without sacrificing torque or speed.
I led an engineering team of six people to design, manufacture, assemble, wire, and test the Sentry for the 2023 season. My design work was primarily on the swerve drive system, the large turret housing both firing turrets, and integrating all of the subsystems, and I contributed significantly to the manufacturing, assembly, and wiring.
The 2024 swerve drive system was a massive improvement over the previous year's design. Changes to the 3D printed structure and component layout made it more stable, compact, lighter, and easier to assemble. A custom wheel with COTS tread was also used to allow for greater flexibility in wheel grip and wheel-pulley keying options.
The Testbed is a special non-competition robot designed to aid the software team in developing odometry, vision, and other various software systems. The sliding vertical plates allow the vision team to test the particle filter auto-selection ML model on various setups of plate height, radius from center, and high-to-low plate offset. The Testbed is also equipped with two odometry modules, an optical flow sensor, and a 6-DoF IMU to help the controls team test auto-navigation, path planning, and new hardware implementations.
I designed, manufactured, assembled, and wired the Testbed.
The odometry module is a follower wheel system used to track robot position throughout a match, achieved by embedding a diametric magnet in the end of the wheel axle and reading the changing magnetic field with an encoder. The assembly has linear suspension such that any robots with odometry modules will not bottom out on them when engaging their drive suspension.
I designed, manufactured, assembled, and stress tested eight odometry modules for use on various robots.
I developed two SOLIDWORKS tools to speed up the workflow of the mechanical team.
1 - Custom Hole Wizard database: because the RoboMaster competition uses metric dimensions, we design all of our robots in metric. However, the machine shop we use only has imperial tooling. This database has remapped all hole sizes (close, normal, loose fits) from ANSI Metric to the closest imperial drill size.
2 - Configured fastener models: using design tables, data from McMaster, and a custom Python program to convert the data to a SOLIDWORKS-friendly format, I created 11 models of fasteners that have all sizes/lengths we keep stocked as configurations. Using these, designers don't need to replace parts as fastener size/length changes in an assembly.
Capstone Project
I was the group leader on a capstone project sponsored by Boeing and Hexcel. The purpose of this project was to design a manufacturing and processing method that improves the manufacturing efficiency and optimizes the processing of composite panels made from Hexcel Nature Range NR-M78.1-LT/48%/UD210P/Flax (flax fiber/bio-epoxy resin) prepreg.Â
We tested mechanical properties (tensile strength/modulus, elongation at break, degree of cure) of samples cured with vacuum pressure only vs. 90psi applied pressure from autoclave. The 90psi method resulted in a slight strength increase; however, we decided to move forward with the vacuum method, as we felt the increase in manufacturing difficulty and higher barrier to usage was not worth the slight strength increase and did not align with the sustainability goals of the material. We then tested the same properties for vacuum pressure with various temperature/soak times and found that higher temperatures resulted in stiffer plates and lower temperatures resulted in plates with higher elongation. I manufactured panels, tested samples, and led the data analysis and final report writing.