
Automatic Cat Feeder with Breed Identifier Using YOLOv8
Client
- Confidential
Category
- Embedded Systems & IoT
The Problem
In a multi-cat household every cat has different dietary needs, and portioning the right amount for each one by hand at every feeding is impractical. This feeder set out to solve that automatically, recognizing which cat has come to eat and serving a portion sized for it without anyone standing over the bowl.
What We Built
We built a feeder that watches the bowl with a camera and uses a custom-trained vision model to tell thirteen cat breeds apart, from Abyssinian to Sphynx. A load cell under the platform weighs the cat and a second under the bowl weighs the food, so the feeder can pick a portion sized for that breed and weight band and shake food out until the bowl hits the target. A touchscreen shows live detection, rules cap each cat at two feeds a day, drop overweight cats to a fixed portion, and record how much was actually eaten when the cat steps off. Committed bench logs running from October 2024 through January 2025 show it working on real hardware day after day. That is the part that separates a demo from a device.
Results
- 13 breeds
told apart on camera by a custom-trained vision model, from Abyssinian to Sphynx
- Weighed twice
one load cell under the platform weighs the cat, a second under the bowl weighs the food
- Portion by breed
food is shaken out to a target weight and stops there, capped at two feeds a day per cat
- 4 months on the bench
daily logs from October 2024 through January 2025 show it running on real hardware
Techstack

