Researchers have developed a fully automated system to dismantle used clothing for high-quality textile recycling. The technology uses artificial intelligence and laser technology to process garments every ten seconds.
- The researchers have created this automated system to identify, sort, and disassemble garments at high speed to address critical global waste issues.
- The prototype system uses machine learning and laser technology to identify and remove non-recyclable elements like zippers, logos, and mixed materials.
- The technology has been developed by Rochester Institute of Technology's Golisano Institute for Sustainability in partnership with Nike, Goodwill, and Ambercycle.
THE TECHNICAL ANGLE: The process begins with a conveyor-fed imaging station where three specialised cameras generate high-resolution, multi-dimensional maps of garments. This allows for fibre composition analysis down to the millimetre level, enabling precise identification of materials.
- The system leverages artificial intelligence and machine vision to identify and remove non-recyclable elements from clothing, which proved uniquely challenging for the team.
- Unlike traditional manufacturing automation, used clothing presented unpredictable variables requiring the system to make on-the-spot decisions.
- Vision-guided algorithms were developed to identify features like logos, collars, and cuffs, while interpreting infrared reflections to define fibre types with precision.
- The data is then passed to a robotic laser-cutting system that cut features with precision and speed, without damaging reusable materials.
- Once cut, garments advance to a robotic sorting gantry, which place clean material into separate bins for recycling at roughly ten-second intervals.
RESEARCH TEAM: Programme manager Mark Walluk led the research team, which consisted of staff engineers Ryan Parsons, Nick Spears, Sri Priya Das, Ronald Holding, and Christopher Piggot. Associate research professor Abu Islam also contributed to the project, with team members bringing expertise in mechanical engineering, robotics, and computer engineering technology.
- Key collaborators included Ambercycle, a Los Angeles-based company pioneering polyester recycling, and Goodwill of the Finger Lakes, which provided garments for testing.
- Nike contributed industry guidance in the project's early stages, offering insights into commercial textile manufacturing and recycling challenges faced by the fashion industry.
- The work began in 2023 and was funded through a grant of nearly £1.3 million from the REMADE Institute, a public-private partnership.
- The team presented their work at a global REMADE conference in Washington, D.C., in April, showcasing their technological achievements to industry leaders.
ECONOMIC AND ENVIRONMENTAL IMPACT: The system was built with scalability and real-world complexity in mind, making it both economical and ready to replicate. The technology is being seen as a step toward a more circular economy for the fashion industry.
- The new technology can transform post-consumer clothing into high-quality, reliable feedstock, making these materials not only viable but preferable options.
FUTURE DEPLOYMENT: Though still in the pilot phase, the technology is reported to be attracting interest globally from recyclers in the United States, Europe, South Asia, and Latin America. The team anticipates transitioning the system to partners for continued testing and potential deployment.
- The system will undergo further testing with industry partners later in the year, moving toward commercial implementation.
- Global interest from multiple regions suggest strong market demand for automated textile recycling solutions, indicating potential for widespread adoption.
- The technology was designed to be scalable and economically viable, with the team confident about replicating the system across different markets.
