On this page... (hide)
1. Week 1: June 23 - 29th
This week marks the kickoff of my eight weeks at Stanford and I was introduced to the lab. My mentor EmJ gave me an overview of the problem we are aiming to tackle with my project which involves grasping satellites with gecko adhesives. We brainstormed ideas for a working prototype (intuitively named Gecko Lasso) and experimentation to test the most critical functionality limits of our design. I am simultaneously working on the TadBot project in collaboration with the Biology Department, which mainly consists of repairs to the older models that have become worn or discovered by the parents as fraudulent over time. I was able to meet our primary collaborator Billie and learn more about the project from both a biological and engineering standpoint. In order to gain access to essential facilities and equipment, a bulk of my time was spent completing safety trainings this week. On Thursday the lab had a summer kickoff cookout/hike which was a refreshing activity to set goals and build camraderie. We wrapped up the week hosting a robotics lab tour for a group of local K-12 students from the DaVinci Camp. They were forward thinkers and asked lots of great questions. The plan for next week is to begin the initial design phase of prototyping in OnShape for the Gecko Lasso!
. 
2. Week 2: June 31 - July 6th
Monday: I began the week by developing a technical presentation outlining the conceptual design and objectives of the Gecko Lasso project. This included drafting a simplified CAD model to visually communicate the core mechanical structure of the lasso mechanism.
Tuesday: The SURF cohort was invited to tour the SLAC National Accelerator Laboratory. This immersive experience offered a detailed look into the history, infrastructure, and scientific legacy of the two-mile (3.2 kilometers) linear accelerator. We discussed SLAC’s foundational role in particle physics, including its contributions to probing subatomic structures beyond the scale of protons and electrons, called neutrinos. The facility’s current utilization of synchrotron radiation—produced by the rotational motion of high-energy electrons—to generate X-ray light sources was particularly relevant, since this technology supports research efforts across the Stanford community. In the afternoon, Mark led an introductory workshop on how to use an Arduino. Although I have learned about basic circuitry through coursework, this session provided exposure to Arduino’s useful programmable control capabilities—particularly relevant for potential automation in our lasso mechanism. A part two of this workshop has been scheduled for next week!
Wednesday: I presented a preliminary slideshow to EmJ summarizing the Gecko Lasso’s overall design concept, key variables of interest, and their associated control strategies. The presentation included computed estimates for variable parameters such as contact torque, target spin rate, and deployment force thresholds, based on literature-derived data and anticipated system dynamics. Our discussion focused on the interdependencies between physical parameters—such as mass, inertia, and surface contact area—and their implications for successful adhesion and capture. This feedback session refined our approach to constraint definition and clarified directions for simulation and prototyping.
Thursday: The day began with finalizing the Bill of Materials (BOM) for the first iteration of the gecko lasso prototype. This involved quantifying all required components—mechanical, electronic, and adhesive—and referencing in-lab inventory. Items unavailable through existing stock or external vendors were flagged for custom fabrication. I subsequently developed a CAD assembly of the full experimental setup. In the afternoon, I began fabricating components at the Product Realization Lab (PRL). Using laser cutting techniques, I manufactured the initial prototype of the deployment arm from available PRL scrap stock.
Friday: I spent the 4th of July with my SURF cohort. We went into San Francisco to see the fireworks show at Pier 39.
Saturday: Kyla and I went into the city where we biked to Golden Gate Park and Ocean Beach. The wind was strong and the climate was cold, however the sand at Ocean Beach was surprisingly warm.
3. Week 3: July 7 - 13th
Monday: I completed part two of the Arduino workshop with the other undergraduates, led by Mark. We learned how to implement a voltage divider through simulation via TinkerCAD and on a physical breadboard. By using a force sensing resistor and LED light, we were able to program the Arduino to quantify how much force was being applied to the force sensing resistor. This mechanism relied on piezoresistivity, the principle that force and resistivity are related. In the case of an FSR, they have an inverse relationship. Afterwards I familiarized myself with the basic anatomy of an OpenRB-150 microcontroller, essentially a smaller version of the Arduino used in the workshop. Being able to program this microcontroller will prove useful when manually altering the torque produced by the Dynamixel motors we plan to use when testing the Gecko Lasso prototype.
Tuesday: I spent most of the day working in Onshape, creating CAD models for the TadBot project and Gecko Lasso prototype. The TadBot project requires repairs to the tadpole's main apparatus for controlling it's tail movement frequency, which consisted of a stand attached to a double-sided canister holder. For Gecko Lasso, I focused on creating mounts for the various motors throughout the design.
Wednesday: Today, I am prototyping the Gecko Lasso with parts purchased online and around the lab. EmJ and I spent the day in PRL.
Thursday: I spent the entirety of the day in the Product Realization Lab designing and printing parts for our experimental setup for Gecko Lasso.
Friday: On Friday the lab had its first lab meeting with the entire group. After the meeting I picked up the Tadbot canister prints which were successful. I had to get creative when removing supports since they required a significant amount. The mount to connect the target motor to the hoop was also successfully printed and I drilled the holes using the drill press in BDML lab to create a tight fit for the screws I am using. EmJ and I met in the design loft to begin assembly of our experimental setup. The tensioning motor we planned to use generates too much torque, therefore we plan to use a Maxon motor instead. This one generates less torque which is useful for our purposes since a lower torque will allow tension to increase gradually enough for the string to conform to the target object's surface.
Saturday: My family came to visit and we spent the day in San Francisco. I have been to San Francisco every weekend since arriving in California. I feel like a native at this point; Karl and I are best buddies.
4. Week 4: July 14 - 20th
Monday: Today am finishing up part fabrication by printing the gripper attachment for the deployment arm and laser cutting the arms to connect the target motor to the hoop. Additionally, I will redesign and print the spool to fit the Maxon DC motor. With these pieces, the experimental setup will be ready for assembly. In order to attach the gecko adhesive to the retractable guiding boom, the initial idea was to use magnets since we want the attractive force to be strong in the normal direction and weak in shear. However, I had the idea to use another piece of adhesive since it is inherently anisotropic. The wedges will be engaged as the boom extends and disengage as the contact patch connects with the target object. I also resin printed the TadBot bodily components.
Tuesday: Today EmJ and I made significant progress on part fabrication. We spent the morning optimizing our part designs for the experimental setup. Some changes we made were to:
- Modify the deployment arm dimensions from 1.5 in. x 14 in. x 3/16 in. to 1in. x 8 in. x 3/16 in. The reason being that our setup simply does not require a longer arm, therefore we could decrease the amount of torque necessary to rotate it.
- Redesign the Dynamixel motor bracket to be adjustable along its x-axis and incorporate a pulley to prevent tangling of our lasso
- Eliminate the need for a deployment arm gripper by fastening the deployment arm directly to a Dynamixel motor; This modification also reorients the deployment arm in way such that its side is in the line of contact with the target debris. The tradeoff associated with this change is a siginificant loss of surface area for which the gecko adhesive could be attached, in exchange for increased bending stiffness.
Wednesday: Today was largely spent in the Engineering Design Loft where I worked to quickly manufacture parts mainly either by 3d printing or laser cutting. Additionally, I finished the resin printing the TadBot body components. This was my first time resin printing, and I quickly familiarized myself with the additional post-processing steps involved. I finished manufacturing all the parts necessary for the Gecko Lasso experiment and began fully assembling the setup. The assembly required various hand tools and hardware to secure parts in place and ensure a safe, functional setup. While the general setup is complete there are still a few optimizations to be made before running the experiment and collecting data.
Thursday: The goal for today was to conduct a small proof-of-concept experiment to prove that out experimental setup is viable. Since we are close to the data collection phase of the Gecko Lasso project, EmJ and I had another brainstorming session regarding which tests would be relevant to define the boundaries of functionality for our prototype.
Friday: Today was the 2nd full group meeting in which EmJ and I presented our progress on the Gecko Lasso project. I explained our motivations for the project as well as the physics behind the feasibility of our design and a video of our optimized experimental setup.
Saturday: My program went to Santa Cruz for a beach day! The weather was beautiful and the setting unusually portrayed the stereotypical version of California: sunny, hot, dry, and sandy. A perfect day for the beach.
5. Week 5: July 21 - 27th
Monday: Today I am focusing on the automatic controls of our experimental system, including motor and servos control. Thanks to Mark's Arduino workshop I am able to use my knowledge of basic Arduino functions to create a program that allows the motor responsible for spinning our target to maintain a constant velocity specified by the user. Since there are two motors in our experimental assembly, I will need to program two Arduinos using separate power supplies. The bigger motor used for the target requires a velocity controller so that the miniscule contact force applied by the deployment arm on the target does not influence the Lasso's adhesion. The smaller Dynamixel motor used to rotate the deployment arm requires a torque controller in order to minimize the amount of force applied by the deployment arm.
Tuesday: The big target motor is spinning, however the velocity controller overestimates its speed resulting in a lower measured velocity than expected. Additionally, it is experiencing difficulty regulating its pulse width modulation (PWM) in response to external force meaning that the PID controller implemented is not measuring velocity correctly. I used the initial mold I printed last week and followed the TadBot body making protocol. I only did one to ensure that my technique was correct. I edited some parts of the protocol, adding tips from my experience that I thought would make some steps a bit more clear for future users.
Wednesday: Today we are refocused on the control systems of the Gecko Lasso. The motor spins, and the controls are functional because it responds to external force. The issue now is that it's rotary motion is very jerky. The PID controller I implemented needs to be tuned so that the velocity adjustments are less abrupt. I am also troubleshooting the second circuit and corresponding C++ code to control the small Dynamixel motor since it is not resulting in the rotational motion expected. I am testing various Arduinos as controllers, voltage supplies, and C++ commands to pinpoint the issue. If the Dynamixel motor continues to be unresponsive, a Maxon motor (known to exhibit exceptional torque control capabilities) will be used instead. For the TadBot, I resin printed more molds for the assembly of the bodies and took inventory of the controllers in circulation.
Thursday: With the PID controller for the big target motor tuned properly, the measured velocity was within an acceptable margin of error of the target velocity and did not jerk as much when regulating its velocity. As it turns out the small Dynamixel motor was unresponsive due to an ID communication error in the code. With this resolved, both motors of the experimental setup are now functional. Since the big target motor still has the potential to jerk, a threshold velocity (with error bars) will be set to maintain constant velocity values across trials when collecting data. For TadBot, I filled the molds with silicone epoxy which will be fully dried and ready for assembly as soon as tomorrow.
Friday: Initial Data Collection Phase of Gecko Lasso + Rob Howe is coming to speak during group meeting + Backpacking Logistics, courtesy of EmJ. She and I then ran some initial tests to collect data and made a few adjustments to our lasso, including adding a Kapton tail to the end of the adhesive. We observed that gecko adhesive with less stiffness adheres more consistently than stiffer gecko adhesive. We believe the addition of the Kapton tail may have potentially resulted in some false positives, however this may not be the case. Upon consulting Mark, we plan to make adjustments on Monday.
Sunday: I went hiking in Muir Woods with EmJ and friends where we became Junior Rangers!
6. Week 6: July 28th - August 3rd
Monday: Testing with EmJ, redefined success/failure after realizing how qualitative our data collection methodology is. Additionally ran into some circuitry issues.
Tuesday: What did Captain Edward A .Murphy Jr.'s Law imply? Anything that can go wrong shall go wrong. Well, today was the embodiment of his philosophy. EmJ and I spent the entirety of the day troubleshooting hardware components, software, and circuit connections.
Wednesday: Independent testing + experiment with the Roboclaw! So exciting! I spent all of today optimizing the experimental setup, particularly the electronics. I familiarized myself with the Roboclaw controller, DC power supply and the Motion Studio and Dynamixel Wizard software. Upon doing so I was able to complete a functional experimental setup, with a defined success and failure region for the gecko-inspired adhesive. With the Roboclaw, I was able to implement a functional velocity controller on the target hoop.
Thursday: The bulk of today was spent on data collection. EmJ and I created a system to qualitatively determine success or failure of adhesion using a slow motion camera and data points were stored in an Excel file. We tested all three radii at 10 different velocities, conducting 6 trials for each velocity. The velocities were increased until consistent failure was observed in the adhesive sample. We used the same sample for every trial, cleaning it between trials to ensure consistency throughout the experiment.
Friday: Today the lab went to the Stanford Outdoor Center to rent/purchase supplies for the annual backpacking trip. We acquired the essentials: tents, hiking boots & poles, sleeping bags & mats, backpacks, and bear cans. Food and snacks will be purchased tomorrow, tents have been assigned, and carpooling has been arranged. EmJ and I are wrapping up a written document to submit to a conference before we leave for the trip. I created a short video summarizing the difference between success and failure of adhesion to supplement our written work.
Sunday: EmJ, Mark, and I completed and submitted our written work early this morning, and are now completely ready to embark on the annual BDML Backpacking trip! This year the destination is the Trinity Alps in Northern California, more specifically Canyon Creek (known for its beautiful scenery). This is a new experience for me, as I have never camped as long as 4 nights and I have been preparing myself. The drive took about 6 hours, and we arrived at the trail head at about 2:30pm which was a bit later than the other carpool groups. We hiked about 3 miles out of the 8 it will take to reach our final destination, the Canyon Creek lake and decided to set up camp for the night.
7. Week 7: July 28th - August 4th
Monday: Today, we spent most of the day hiking the remainder of our 8 mile path. When we finally reached the lake, the sight was beautiful.
Tuesday: Today was a relaxing, unplugged day. I played cards with lab mates, climbed a mountain, enjoyed some leisurely reading in a hammock by the lake, and played volleyball. According to a Chinese proverb, a picture is worth 1000 words. So, here are some pictures from the trip:
Wednesday: Despite the fact that the hike up to our campsite took about 12 hours total, we climbed down in less than 5 hours! The drive back to campus took about 6 hours, and we arrived by evening.
8. Week 8: August 11th - August 17th
Monday: Today marks the first day of the last week of the SURF 2025 program, which means it is time for final presentations! For the SURF program I am required to give both a short research talk and a poster presentation. I spent today working with EmJ to create/adjust figures and refine my poster and slideshow.
Tuesday: My poster was printed and I went to pick it up today. It was quite large at 44" x 56", however I am pleased with the way it printed. Since my research talk is tomorrow, I am practicing with EmJ while she provides feedback. Additionally, I talked with Seongheon about developing an electrode that will allow force sensing on the gecko-inspired adhesive patch. This is beneficial because it will allow us to quantify the area of adhesive making true contact with a surface.
Wednesday: All of today is dedicated to SURF program research talks for the SURF program. I am one of the first to present, but listening to the diversity of research being conducted by my cohort was insightful and inspiring. It was a powerful reminder of possibility and that despite the current governmental disapproval of science, we as scholars refuse to cease discovery & innovation.
Thursday: This morning Mark introduced me to a visiting professor and I gave a brief spiel about my project. My poster presentation will take place in the evening and until then I intend to wrap up the Gecko Lasso project by organizing the electronic documentation (CAD, Excel, Powerpoint, etc.) so that the next researcher to continue the project will be able to easily follow what has been done. This is a very important and somewhat overlooked part of research because it allows a project to seamlessly progress as personnel constantly transitions.
Friday: This is my last full day at Stanford for the summer, and I am packing my bags to return to Maryland. I am grateful to the entire lab for their support and engagement throughout my eight weeks here. I have learned many valuable hard and soft skills that I can now bring with me to spaces that I occupy in the future.
Saturday: 8 hour flight back to MD (including the time difference).