Inspiration
I used to box when I was younger. It's a great way to remain fit. The problem is that It's quite boring to punch a boxing bag for 10x3 minutes. So I had this idea in 2020 during covid why not create a gaming app for this.
What it does
Think of the following. The user gets ready for a workout session on front of his/her boxing bag, puts on two silicone wrists both with a 6axis motion sensior in them, opens the app, sensors connects on BLE, hits the button "Matchmaking", the app would find him/her a sparring partner, one the session starts, you can start punching the bag. The app would record the user's hit types as well as when hitting a hidden combo, there would be sound animations and bonus damage in the combat log. User could also find special items which would give the user special attributes and abilities which eventually alter his/her total DPS/total damage. There would be a streak system and a very competitive ranking/seasonal system.
"It's a game while doing fittness, or the other way around"
How we built it
The project is nowhere near ready yet. I am building it with the latest GPT model.
Challenges we ran into
Arduino programming is not easy for chatgpt, also maintaining a stable BLE connection while not overloading the antennas was a challange. mapping the punch types (cross, jab, hook, uppercut) is also challanging, but I am in the right path. I think I will be able to have the app recognize the punch types with a 95% accuracy by the end of summer.
Accomplishments that we're proud of
The reason why I started this because I reliazed that with zero developer knowledge I can make a php website in about an hour with the guikdance of chatgpt, so I though why not start this now. I am proud that I haven't parked the project yet, because gpt gave me a solution for each difficult (for me) tasks.
What we learned
I learnt a lot about how databases work, how software developement work in general, github, react packages, some coding as well
What's next for Boxing gaming app
I would like to be able to say by the end of Summer, that the project has merit and will work. When I will see the punch types being recognized by the app, I knew that the heavy lifting is done. Currently I managed to design the principles are punch recognition. I am trying to find the best prompt for gpt to give me the recognition formula.
next step is a kickstarter campaign hopefully after summer so I can afford the first phase of testing with like 20-30 users, then second phase with more. The device cost with low MOQ is very expensive.
cheers Lalo
Built With
- arduino
- cpp
- php
- react
- react-native
- supabase
- typescript
Log in or sign up for Devpost to join the conversation.