It is led by a team led by Mark Walluk, Program Manager, consisting of Staff Engineers Ryan Parsons '17 (Mechanical Engineering), Knickspears '24 (Robotics and Manufacturing Engineering Technology), Sri Priyadas, Ronald Holding, Christopher Pipoto '91 (Computer Engineering and Technology), and ABU ISLABE from the Associate Research Association of ABU ISLABE. recovery.
The process begins with a conveyor feed station where three specialized cameras generate high-resolution multidimensional maps of clothing and fiber composition analysis generates high-resolution multidimensional maps up to millimeter levels.
Researchers at RIT's Golisano Institute for Sustainability have developed an AI-powered, fully automated system that identifies, sorts and disassembles clothing for textile recycling. Process clothing every 10 seconds using advanced imaging, machine vision and robotic cutting. Supported by Remake Institute's fundraising, it aims to transform consumer clothing into high-quality, recyclable items.
The system then leverages artificial intelligence and machine vision to identify and remove unrecycled elements from clothing, proving a unique challenge for the team.
“In traditional manufacturing, these automations have been used for decades and are predictable,” said Islam, which collaborated with DAS on AI integration. “We know which part is coming next, exactly where we're going. With used clothing, all items are different. That unpredictability means that the system needs to decide on the field.”
To address that challenge, Islam and DAS have developed visual guidance algorithms that identify functions such as logos, colours, cuffs, and interpret infrared reflections for the definition of fiber types. That data is passed to a robotic laser cutting system that cuts these features with accuracy and speed without damaging reusable materials. Once cut, the clothes go to the robot sorting gantry and place clean materials in separate bins for recycling. This prototype can process new clothing approximately every 10 seconds according to research.
Walluk said the system is economical and ready to replicate as it is built with scalability and real-world complexity in mind.
“We will not solve the global textile waste problem, but it is a step towards a more circular economy,” Wark said. “Today, recyclers prefer post-industrial fabrics because of predictable material properties. We are working to move beyond that step by converting consumer garments into high-quality, reliable ingredients.
Key collaborators include Ambercycle, the pioneering polyester recycled company from the Los Angeles-based company, and Finger Lakes' friendly friendship, which provides clothing for testing and insight into the resale and reuse market. Nike provided industry guidance early in the project.
The work, which began in 2023, was funded through a grant of approximately $1.3 million from the Remade Institute, a public-private partnership focused on the development of circular manufacturing solutions. The team announced their work at the global remodeling conference held in Washington this April.
“Textile recycling is an important global challenge and we are proud to be able to work with industry leaders to drive meaningful solutions,” he said. Nabil Nasr, director of GIS and CEO of The Remade Institute. “This initiative not only creates a significant environmental impact, but it is also a key area of growth and innovation for us at GIS.”
Although still in the pilot stage, the technology has already attracted global interest from recyclers in the US, Europe, South Asia and Latin America. The team hopes to move the system to partners later this year for continuous testing and potential deployment.
fibre2fashion News Desk (RR)
