Inspiration
We were inspired by Iron Man to make this robotic arm. In my dreams we would make a full suit, but for now we wanted to work on the arm. We also hope that this arm can help people who need an extra arm or an attachment of their normal body to deal with dangerous substances.
What it does
It is a gesture controlled robotic arm. It copies your arm through a list of 10 different signals, they are: peace, thumbs up, pointing, open hand, fist, cool, call, B, pinky, three. It senses your hand with a camera. After a task is completed with it, a scaled blue print in generated for the user to view the prototype of the task they completed for future refrence.
How we built it
Before the hackathon we knew we wanted to make a robot in some capacity, so we pre printed using our own CAD, basic components of robot, we could customize later how we would want. Next for the code we used a collaborative IDE to share code and program the python for the robot arm together.
Challenges we ran into
One challenge we went through is getting the 3d printed pieces. We were able to do this using a friend's printer. Another challenge we ran into is how to connect the 3d printed pieces. We used super glue for this which worked to our standards. Also de-buging logic code was difficult, which took time during the coding process, we mitigatied this using common debugging techniques and outside assitance. Plus if anyone ran into a bug, another person could view the code with a fresh set of eyes.
Accomplishments that we're proud of
We are extremly proud of ourselves for figuring out the blue print generation part of the system. When we started the hackathon, we belived the gesture control would be difficult, but we instead spent the most time using online resources to learn how to create a code that would draw out movements of the motors on the robot. This was a huge learning oppourtunity for us.
What we learned
As stated before, we learned how to de bug logic issues, create a program to generate drawings from moter movements, and most importantly teamwork, since we had to get thorugh some conflicts, but in the end became a stronger team.
What's next for Blueprint Bot
In the future, blue print bot will include a more robust bueprint mapping, that can utlize AI api to be more appealing and detailed. Also the software itself will include a better UI so that it is better as a real product for its engineering and industrial use case.
AI usage citations
Gemini 3.5 Flash was used on our google slide show to generate a voice over. The presentation itself used no AI assistance, and the voice over is a built in feature for google slides.
Claude Haiku 4.5 was used to help de-bug our own original code, for example we were coding the gesture tracking, and ran into a bug where it was not detecting gestures, we uploaded the code to Claude, and it detected a logic issue in 3 lines of the code. We also used claude for image generation of the blue print schematic of the robot arm, this isn't a techincal usage, just a design element of the pitch deck.
Log in or sign up for Devpost to join the conversation.