MINI SORTING MACHINE
MINI SORTING MACHINE
Sorting Objects using a Raspberry Pi and On-Device Machine Learning
Sorting Objects using a Raspberry Pi and On-Device Machine Learning
Embedded Files
2024-2025
Project Summary
Project Summary
The Goal
The Goal
My lab partner and I were tasked with designing a machine-learning-powered sorting machine that could sort objects of different shapes, sizes, and colors. As this was built for the lab's SmartCT program (see below), it had to be powered by a Raspberry Pi and made using 3D-printed parts.
The Problem
The Problem
Though machine-learning has become much more accessible in recent years, it still often requires powerful hardware to have any viable applications in industry. In addition, the steep learning curve associated with training a model makes the power of ML difficult to harness as a newcomer.
The Result
The Result
The current version of the sorting machine sorts different colored balls into one of 2 bins. The machine can sort 40 objects in 90 seconds, and features an automatic dispenser that drops each ball into the sorter at a constant speed. It uses a Raspberry Pi and a Coral USB accelerator to run an on-device ML model.
Overview of the SmartCT Program
Overview of the SmartCT Program
The Smart Computational Thinking (SmartCT) project is a collaboration between the MSU Computer Science Department and College of Education. The goal of this ongoing project is to make machine-learning more accessible for high-schoolers by developing robotics "kits" that students can build and use to learn about ML and computer vision.
Final Design
Final Design
sorting_demo.MOVContributions
Contributions
I wrote the Python code for the following major device processes:
Page updated
Google Sites
Report abuse