SUPR
Unsupervised Sorting Robot
Dnr:

NAISS 2024/22-1479

Type:

NAISS Small Compute

Principal Investigator:

James Waguespack

Affiliation:

Linköpings universitet

Start Date:

2024-11-13

End Date:

2025-02-01

Primary Classification:

10207: Computer Vision and Robotics (Autonomous Systems)

Webpage:

Allocation

Abstract

Automation is an important part of industry, as it helps keep costs down and frees human workers to focus on less menial tasks. We would like to help contribute to the automation of tedious tasks. To this end, we are attempting to create a method for an unsupervised robot to sort novel items into a given number of separate bins. The criteria for the sorting are to be decided by the system with no input from a human. We will analyze ways to encode object properties such as shape and color using computer vision models, as well as clustering algorithms for deciding which objects will belong to which bins.