Applying machine learning to the recycling industry

18 October 2021

The world generates 2 billion tonnes of domestic solid waste annually but less than ten per cent is recycled because the current recycling process is extremely inefficient. We’re working with a start-up that is developing a revolutionary robotic system to significantly increase recycling efficiency.

Danu Robotics is an Edinburgh-based clean tech company that is looking to improve the efficiency of recycling through automation. Currently, most recycling centres and plants operate by having human pickers sort through recycled goods on conveyor belts – any recyclable of the wrong category (eg any soda can in the paper recycling) is manually removed from the conveyor belt and either sorted into the correct category or disposed of if not recyclable. This process has many disadvantages: sorting recycling is a thankless and repetitive task; sorters are prone to error (especially late in a shift); and, as sorting rates are determined by the number of people working the conveyor belt, sorting can quickly become a bottleneck in the recycling process.

Here is where Danu Robotics steps in by offering a robotic sorting system that can swiftly seek and sort the recyclables on the belt, with the aim of removing anything that shouldn’t be there. To do this, the system needs to know what to pick, and where to pick it. For this, Danu Robotics is developing a machine learning solution – before deployment of the robotic sorting system, Danu Robotics will collect video footage of the conveyor belt with which to train a machine learning agent. This agent will learn to categorise every item on the belt. Once the initial training is complete, this machine learning algorithm will direct the robotic sorting system to sort out waste items efficiently and effectively.

So where does EPCC fit into all of this? Xiaoyan Ma, the founder and CEO of Danu Robotics, is an alumna of the EPCC MSc programme in HPC with Data Science. Early in 2021, Xiaoyan got in touch with us regarding a small project for which she would like help from EPCC. With an Interface Innovation Voucher (matched in time by EPCC) plus support from the EuroCC@UK project, we had the seed of an ongoing collaboration.

We have been providing assistance where we can, specifically helping to develop an algorithm that will ensure communication between the machine learning agent and the robotic sorting system goes smoothly, and that the agent does not try to overwork the robotic pickers!

We at EPCC are both proud to see our former students actively engaging to change the world for the better and excited that they want to involve us in bringing forth this change. Plans are already in motion for this collaboration to continue into 2022.


Dr Julien Sindt