Over the last week, I have been preparing for my final presentation night by rehearsing my speech and presentation in front of my classmates and teacher. This practice rehearsal went very well, in my opinion, as I covered all the points I wanted to, except for one about my future goals with this robotic hand project and college in general. To prevent this, my teacher advised me to add a slide to my presentation specifically about this and suggested adding pictures of the professionals I interviewed to make it feel more personal. As for the actual robotic hand, everything works as is, but I plan to code a few more fun concepts, like having the hand count from one to five or even ask what the audience wants it to do to make it more engaging and further present my project. Unfortunately (or fortunately), I will miss my presentation night at Independence High School because my robotics team qualified for the state championship, which is on the same day. So, instead, I will be giving my presentation at Liberty High School on Tuesday the following week, which gives me a bit more time to prepare.

Over the last week, I have been preparing for my final presentation night and working on my final product to make it presentation-ready. I have reused the same design for my invitation and program from last year to maintain consistency with my branding, updating small design elements based on last year's feedback. Furthermore, I also developed my visual aid (slides) for my final product, which I have kept purposefully minimal because I plan to focus on the physical robotic hand, the 3D CAD model, and the code on my laptop. As for the actual robotic hand, I have implemented some quality-of-life elements, such as creating spaces for the hot glue on the fingertips so that the tendon is not glued to the surface but remains part of the finger. I printed the robotic hand for a test, but there was a manufacturing issue: the thumb’s hole did not print, and the first layer of one of the servos’ spools was not properly adhered to the print bed. Regardless, the hand still worked wonderfully with the new TPU insert design I had implemented. Over the next few weeks, I will be working on my final speech and just getting ready for the big night.

Over the last week, I have been super busy with the two competitions my robotics team has been participating in. Nevertheless, I managed to carve out some time to continue incorporating the new design revelation I had tested last time into the entire hand. This was a painstakingly slow process, as I also discovered some issues with my old design. This led me to delete the master sketch for the servo placements in the palm and redo it much more efficiently, using the appropriate tools and constraints I learned through this very process of iteration. While I was at it, I also added more clearance for the servos’ spool and wires. Moving forward, I will talk with my mentor about the new design and re-prioritize my list of design features in order to finish on time for my practice final product speech next week. Some ideas I know I have to implement are reducing the spool size and mounting the electrical components on the rear side. For the first issue, I will ask one of my engineering teachers at the CTE center if I can use the school’s resin printers for my project, as they can print much more finely than my standard FDM printer. Since my current spools are as big as the metal horn attachments that come with the servos, it is impossible to make them smaller, so I would need a resin printer to print a smaller spool that directly attaches to the small pinion (gear) on the servo, removing the need for the horn attachment altogether. However, the resin printer has the downside of requiring further post-processing, which can be harmful if done improperly.

Over the last couple of weeks I have been thinking and working about an alternative mechanism to solving the issue with my servos being too weak to stretch the thermoplastic polyurethane (TPU) inserts I had on the backside of the fingers to passively return the fingers back to its original position. I had tried countless iterations of different finger and TPU inserts, even going as far as to buy and try and try a weaker TPU material. This was all to no avail, so after a lengthy conversation with my mentor about solutions through over-volting the servos to squeeze out even the tiniest amount of torque from them, I CADed up a design I was thinking about where instead of the stretching the TPU, the finger crushes the TPU down when flexing, so that when the servo stops applying a force, the TPU would expand to its original size, thereby causing the finger to extend. After a couple of hours of editing the design by moving the TPU inserts from the back to the top side of the finger and a couple hours of printing the new finger, spool, and TPU inserts, I tried this idea. I sat down at my table and using all my fingers to hold down the one robotic finger and servo, I ran the code. To my utter surprise the finger actually curled in on itself and the TPU didn’t fall out. Once again my design shattered my expectations by actually extending on its own. So, now I have to repeat this design onto all the fingers and move onto the other issues I had listed with my mentor in previous meetings.

Over the past couple of weeks, I have been focusing on solving the problem of the thermoplastic polyurethane (TPU) inserts being too strong for my servos. I have iterated countless designs for both the robotic fingers and the TPU inserts. I have experimented with different dimensions for the TPU inserts, testing both small and large. I have also messed around with the print settings for TPU and even tried 0.5mm thick TPU inserts. Unfortunately, none of these could be fully pulled by the servo. So, I followed my mentor’s suggestion and bought a roll of lower shore hardness TPU filament. The current roll of TPU filament I have is 95a, and I had bought a spool of 85a TPU, which will arrive this week. Hopefully, drastically reducing the elastic strength of the TPU inserts will enable the fingers to work as intended, with the servos able to adequately curl them, and the TPU inserts acting passively as a spring to return them to their original position. My goal this week is to finally put an end to this problem by either succeeding with the weaker TPU or completely redoing my design to use a different mechanism for finger extension.

Unfortunately, we had a week-long school holiday due to the winter storm from January 26th to the 29th, so the ISM Showcase had to be postponed to a later date. Due to this unexpected time off, I was able to focus my research and study the effects of changing my 3D printing settings for Thermoplastic Polyurethane (TPU) inserts in my first iteration of the robotic hand. The problem I have with this prototype is that the TPU inserts return the fingers to their natural resting position after curling them, but I think the servos are too weak to pull the fingers with TPU inserts. I will do further testing on this to confirm that it is, in fact, an issue with the TPU being too strong. Nevertheless, optimizing the TPU strength to control passive extension motion and force is a beneficial feature I did not plan to include or optimize when designing this mechanism. After I solve this problem, I will move on to testing its grip strength with various objects and modifying the design for my second iteration. I am considering adding more TPU inserts to the contact surfaces to improve grip, moving from a simple plastic contact surface to a conformative and adaptive type of grip. I hope that for my second iteration I will be able to properly control all five fingers without any help from my own human hands and code a program that can hold onto objects.

Over the last couple of weeks, I have been focused on preparing for the ISM Showcase this Thursday (29th), where all ISM students from the district present their research and original work. One thing I am proud of from these weeks is my utilization of Canva's feature to create custom-sized designs. So, I used the real-world size of the trifold (48x36 in) in Canva and added a couple of lines 12 inches from the edges of the design to represent the trifold. This way, I could accurately design the board with proper scaling when I went to print it out. Furthermore, since it was real-world scaling on Canva, I could easily replicate the design in real life by just copying down the placements. However, I will need to figure out how to power my robotic hand and charge my laptop at the event, since it will be held in the middle of a football field.
Furthermore, I met with Mr. Rahmati, my mentor last year, to discuss my original work and the future iterations I will need to work towards. After this interview, I asked the big question of mentorship. Fortunately, he said yes, noting that this year I actually had the resources to bring my project to life and that he is willing to support me through it.

Happy New Year! I had a very relaxing break, not working on anything. Nevertheless, since the school started, I have made progress on my robotic hand. In fact, I even 3D-printed my first iteration of the entire robotic hand, in which I could place all my servos and actually have the hand work, since I had all the electronics and basic code done. Unfortunately, this 100g print is useless because the 3D model I had downloaded last (academic) year in ISM 1 for my servo was incorrect compared to the one I physically bought. This waste of 100g can be attributed to one singular letter. I downloaded a model of the “SG90S” servo, but I bought the better “MG90S”, which is only a couple of millimeters longer. Moreover, the model underrepresented the width of the wires exiting the servo, resulting in the holes for the wires being too small by just a couple of millimeters. There is a learning moment from this failed print, however, as I realized that there is nothing holding the servo down in the palm. So, my next iteration will use the correct servo model and include a mechanism to firmly keep the servo in place.

Over the past few weeks and during Thanksgiving break, I have interviewed multiple professionals who all affirmed my designs and plans and advised me to get on my feet and start actually printing and wiring to iterate on my designs meaningfully. My progress so far has been to refine my designs based on my interviews, 3D print them, buy the electrical components, and begin basic prototyping. I have been experimenting with the servos and Pico with some code. Specifically, I am learning the low-level workings of the servos and the Pico through the PCA9685. Once I understand how these work, I can move on to using the servos for their intended purpose: curling a finger. I plan to iterate on the spool's design until it consistently pulls the tendon back to curl the finger fully. Then I will move on to repeating this for all five fingers and to get the first prototype of the entire robotic hand.
As for mentorship, I yet again have two strong candidates: my previous mentor and a student at TAMU pursuing a master's in robotics. The conversations with each of them have been super beneficial, with their own benefits. I think I will keep talking to them and slowly introduce the idea of mentorship by mentioning bi-weekly calls or something, and see if they have any problems due to their schedules.

Over these past couple of weeks, I have been thinking more about CADing and building my robotic hand because I need to plan out where to place my servos. As per my previous plans, I am currently working on embedding the motors into the palm of the hand and positioning them such that all five do not interfere with one another. Alongside this, I am designing a spool to attach to the horns of the servo, rather than its pinion. But the horn's length causes my spool to be bigger than my palm height. The obvious course of action is to simply increase the height of the palm to accommodate the spool. However, doing so would make it look bulky in my opinion. So, instead, I plan on keeping the same height but having cutouts in the backside of the hand in which the spool can stick out. These extrusions can be covered with the Raspberry Pi and power supply mounts on top, maintaining their aesthetic appeal. Nevertheless, all this planning won't matter if I can not physically get all five servos to fit in the palm. If this is the case, I will just move it externally and not worry about the size of the spools.
My plans for the next few weeks are to acquire some jumper wires and either purchase a new Raspberry Pi Pico or keep my 3 B+ Pi to test out my designs and complete a prototype by the end of the year. I plan to have at least one finger actuated by a servo using tendons for my original work. Obviously, an entire hand would be preferable.
As for interviews, I have two more scheduled for this week: one with my mentor from last year and another with a college student at TAMU who is studying robotics for his master's degree. Hopefully, these go well and provide even more insights into my development.

I have been reaching out to professionals, mostly college professors and students in robotics-related classes and labs. Out of the numerous cold emails I sent, I received only two replies. One of which was simply a denial, and suggested I ask other professors. The second one, however, agreed to a short conversation. The conversation was productive as we discussed his work in robotics at UT Austin and gained some vital insights into my own project, a 3D printable tendon-driven robotic hand. He recommended that I not worry too much about embedding the servos into the palm of the hand, and could simply mount them onto an external part, such as a forearm. Taking this into consideration, I have decided to try to embed the servos until the end of the month, and only if this proves unsuccessful will I mount the servos externally. Ultimately, I hope to get the robotic hand working for my original work so I can move on to (maybe) developing a robotic arm for the hand or make it more dexterous, especially if the servos are external.
For my third interview, I plan to reconnect with my mentor from last year, showcase my progress, and ask for his insights into the electrical aspects of my project, specifically regarding the controller and power supply. Lastly, I plan to ask Professor Justin Hart (the interview mentioned above) for another interview in a couple of weeks to showcase my progress.

The past couple of weeks have been busy for me with all my tests, college applications, and ISM. Now, I have shifted my focus from research papers to interviewing professionals in the field. Although I already have warm contacts from last year, I want to reach out to new people and connect with professionals. So, so far, I have created a list of professionals and companies that are oriented around robotics. I will use this list of contacts to conduct cold emails or cold calls this week and hopefully secure an interview with at least one of them. In doing so, I will deviate from my usual connections and step outside my comfort zone. Furthermore, I plan to interview my PLTW Digital Electronics teacher because her class is precisely what my electronics course will be based on.
On the other hand, my work with the robotic hand has hit a roadblock as I encounter a couple of obstacles. Firstly, I need to figure out a way to attach a spool to the servos; this is a problem because the pinion on the servo has tiny gear teeth that are too small for my 3D printer. Secondly, I also need to plan out my electronics and design the layout to mount them onto the robotic hand.

Going into the ISM Professional Symposium for the second time, I already knew what to expect, so I prepared myself for the onslaught of polite small conversations with enthusiastic professionals from a variety of fields and students with the same drive and passion as I. I really enjoyed seeing the massive number of students in ISM this year. It did backfire a bit since there weren't quite enough professionals for every group of students. Nevertheless, it all worked out, and I got to meet people both related and unrelated to my career field of robotics. I received advice on my future based on potential colleges and jobs. Surprisingly, most professionals said that, in the long run, a college degree is not as significant as I had believed it to be. Many ended up in a fulfilling job that was not related to their college education. Then came the mock interviews. This year, I made it my goal to sit down with anyone, regardless of their career. This mindset proved invaluable as I sat down with a retired teacher and an editor who also helps students with their college applications. So, I received a lot of impactful and personalized advice based on me and my interests. All in all, the symposium was amazing and equally as beneficial as always, and I wish it continues striving for the same level of quality for the upcoming students.

This past week, I have been preparing for my second Professional Symposium. Firstly, I have been thinking about the answers to the typical questions that will be asked of me during interviews and networking with professionals and students. Do not get me wrong, I have not been memorizing scripted answers; instead, I am thinking about bullet points that I should talk about, given any questions. For example, suppose someone asks me about my ISM topic/project. In that case, I am not going to go into depth about robotics and AI; instead, I will keep it simple and talk about how I am researching and developing a robotic hand that can grasp objects on its own using computer vision. I could then expand on it by talking about why I chose this: scalability. I can continue this project by developing a robotic arm for the hand and then improving the computer vision system. Secondly, I have refreshed my resume to make it more concise while holding all my new experiences with the help of my teacher and classmates. Finally, other than preparing for the symposium, I have also begun my research with an excellent article that has similar functionality and design to mine. One obstacle I am facing, however, is that this article references another that discusses how they designed a two-DOF thumb using just one motor, which would be a game-changer for me. Unfortunately, it is blocked behind a paywall and requires me to ask for access directly from the authors, which is unlikely.

Having begun my second year in ISM, have learned a few tricks here and there to conduct my research more efficiently and productively. For example, I utilize the databases that my school and district provide completely free of charge, which contain every topic that anyone would need for their research. Furthermore, Gale databases allow users to explore their topic via a topic finder wheel/tile. This tool has enabled me to expand my prompting beyond rudimentary searches of “robotic hands” or “computer vision.” Instead, I input a broad category into the topic finder, like “automating robotics with computer vision” or “3D printing for robotics.” Then, the topic finder would provide numerous sub-topics under this and several articles/papers that fall under those categories. By using this, I have managed to save countless research papers related to 3D printing for robotics, approximately twenty. Imagine the number of highly credible sources I can get for all the other research areas, like computer vision, electronics, and even something as simple as inverse kinematics. Other than just research, I have also begun preparing for the ISM Symposium by refining my resume and preparing not only to answer questions, but also to ask questions to professionals in any field, even if they share no connection to robotics. I look forward to revitalizing my ISM journey and also to inspiring other students to reach for the stars.

Firstly, I returned to the Independent Study and Mentorship (ISM) program for a second time to continue developing my professionalism, research skills, and dedication to my project. ISM allowed me to explore robotics engineering and solidified my passion for pursuing it as a career. I concluded the year with a 3D design of a tendon-driver robotic hand. However, my lack of hands-on 3D printing left the design with significant flaws. While I had researched 3D printing, I had never actually designed and printed a functioning model. Over the summer, after touring some exceptional colleges and even staying at one for a week, I gained unrestricted access to a Bambu Lab P1S (without AMS). This allowed me to understand better what mechanical systems could realistically be 3D printed. Using this experience, I completely restructured my robotic fingers and hand, and continued iterating based on prior and new research. I will delve deeper into the exact changes in future blogs, but for now, just know that I successfully 3D printed a robotic hand, albeit without motors or any controlling system. My next step is to dive deeper into the electrical aspects of robotics, specifically the servos, Raspberry Pis, and circuit design. I'm focusing on using a Raspberry Pi 3B+ that I had lying around for the central controller for the hand. This phase will bridge my mechanical progress with the electronics necessary to make the hand fully functional. The next phase after this would be to integrate these components and then program them.