Hi! I'm a robotics engineer and researcher who just graduated from Olin College of Engineering.
As a researcher, I've derived novel models and algorithms based on biomechanics in snakes, octopuses, and elephants, and I've used these to design and control soft robot arms to safely work alongside humans. I published this work as the first author in my 3rd year, and was invited to give a conference talk as the sole undergrad speaker, where it won 3rd best poster at a workshop.
As an engineer, I've developed robotics controls and sensing algorithms on hydrofioiling boats and construction robots. This includes firmware, nonlinear controllers, SLAM algorithms, simulators, and motion capture, both in C++ and Python.
I am currently building novel genome sequencing technologies at XGenomes as a microfluidics engineer while applying to PhD programs this fall. I'm interested in applying computational modeling to better understand biological systems, and using this understanding to improve engineered solutions.
A Fast and Model Based Approach for Evaluating Task-Competence of Antagonistic Continuum Arms
Bill Fan, Jacob Roulier, Gina Olson
Determining whether a proposed continuum arm design can complete a specified task and load is computationally expensive due to the need for simulation. We introduce a novel method of analysis that eliminates the need for simulation, speeding up computation by 80x, and creates novel metrics for evaluating arm designs against tasks. Olin, UMass | Under review for Robosoft 2025.
Beyond "How's it going?": A Collaborative Autoethnographic Study by Early Instructors in a First-Year Engineering Studio Course
Esme E Abbot, Berwin Lan, Luke Raus, Bill Fan, Zachary del Rosario
Studio teaching is increasingly popular, but teaching in one is difficult, especially for new instructors. We studied how new instructors learn to teach in a studio environment through a collective ethnography of four undergrad course assistants. We identified challenges new instructors faced and strategies they in response. Olin | ASEE 2024.
Linear Kinematics for General Constant Curvature and Torsion Manipulators
Bill Fan*, Farhan Rozaidi, Capprin Bass, Gina Olson, Melinda Malley, Ross L Hatton
Inspired by octopus arms, helically-wound muscles greatly increase continuum arm workspaces, but they are more difficult to model and control. We present a novel application of Lie theory to simplify the modeling of constant twist soft robots, and unify modeling across different designs. Lightning talk at Robosoft 2023, and 3rd place best workshop poster. Olin, OSU, UMass | Robosoft 2023
In my 3rd year I co-founded Twist Lab - a soft robotics research group with 10 students. As PM I was responsible for raising over $7,500 in lab funding from grants, supported projects in the design, control, and simulation of soft robotics, and helped students apply to fellowships.
During my gap year before college, I built an autonomously hydrofoiling boat prototype at an early stage startup capable of multiple minutes of airtime. We added motorized hydrofoils with ROS controls to an existing boat hull, developed a simulation, and used it to implement a cascaded PID controller.
As the lead course assistant, I (re-)designed the final-project system for the Feedback Controls course at Olin, and created kits for all students. In the month-long project the students levitate a permanent magnet using an electromagnet and a hall-effect sensor. Gif by Krishna Suresh.
In my scientific computing class, I implemented a particle based model of the non-linear phase change of glass using Python. To render the results, I implemented a modern OpenGL GPU-programming shader pipeline, complete with a phong lighting model and full camera movement in 3D space.