Custom Dog Playing Fetch with ML Used AI
This project uses a custom YOLOv5 model to track and follow a 3d printed ball to help a dog play fetch using a live camera. The model is trained to find the ball in each video frame and track its movement as it runs to get the ball and comes back. By looking at where the ball is and how big it looks in the video, the system can tell how far the ball is from the dog.
Unfortunately, it is a very complicated process if you want to change the ball design as the whole model is based off of training images of this specific blue and red ball. Sorry 😢
click here to access replit if demo does not work (the demo is just a simple project that shows the basics how the ball is being detected): https://replit.com/@IshaanAhuja-algebros314/BallDetectionDemo#templates/index.html
Followers
No followers yet
Ship Your Project
Get ready!
Once you ship this you can't edit the description of the project, but you'll be able to add more devlogs and re-ship it as you add new features!
Timeline
I couldnt have called this a finished project without having AI in it. I have finally added a microphone and a speaker in the dog and it can not talk to people. Currently its using an API to get the answers to the questions but it is still pretty fast in responding. Finally I am done with the project!
designed and 3d printed a robotic dog tail with 2 axis of movement and coded it so it wags the tail when it finds the ball
worked on making the models docs and describing the steps to install it yourself and make a custom model without doing all the research
