At a London bank where Amy Ma worked as a software developer, recycling bins were emptied into the trash. That experience led her to look more closely at how recyclable materials are sorted—and to build a robot intended to make the work faster and cheaper.
A sorting problem with room for automation
Ma found that recycling depended heavily on people sorting waste by hand to separate different materials. She saw a chance for technology to improve that process, and spent six years developing the idea at Danu Robotics, a company based in Edinburgh, Scotland.
The company’s machine, called H.E.R.O., is now moving toward the market. It is designed to pick items up with a pincer claw rather than the suction system used by some competing robots. Danu’s other key distinction is its use of AI to keep improving the robot’s software.
That combination addresses two parts of the challenge: how the robot handles material and how its software develops over time. The aim is to sort recycling more cheaply and quickly, an outcome that matters to facilities whose revenue depends on the materials they can recover.
The case rests on sorting economics
Danu is entering a market that already includes robotic waste-pickers, including suction-based systems from Glacier and RecycleEye. Ma argues that recycling centers have a direct financial reason to adopt a system that works: material sorted out of the waste stream can generate additional revenue.
Ma estimates a working site could earn $485,000 in additional revenue from adding the robot to its sorting line. She puts the initial investment at $160,000, with maintenance fees of $24,000 each year. Those figures are her estimate of the potential return, rather than reported results from a named operating installation.
The company has early signs of commercial interest. It has letters of interest from two large customers, signed contracts worth $500,000, and more than 200 customers in its sales pipeline. Danu says the opportunity sits within a $20 billion market for recycling sorting across Europe and North America.
These figures suggest a route for Danu to establish itself: show that a robot can recover enough valuable material to justify its purchase and upkeep. The pitch depends on performance at the sorting line as well as the costs of installation and maintenance.
From waste facilities to the source
Danu’s immediate work is to make H.E.R.O. sturdier, smaller, and more versatile. The company has raised $5 million in late-seed funding to maintain commercial momentum as it develops the product.
A smaller, portable machine could broaden where recycling happens. Ma has described using it as a standalone solution at events, shopping centers, hospitals, and airports, so waste could be sorted where it is generated instead of being handled only at dedicated facilities.
That expansion would depend on making the robot practical for settings beyond a recycling center. The device would need to fit the needs of each location while continuing to sort reliably. Ma sees source-level sorting as a way to change how packaging waste is managed.
A commercial test still ahead
Danu’s approach brings together a mechanical choice, AI software, and a business case based on recoverable materials. Its current customer interest and signed contracts offer evidence of demand, while the company’s estimates describe the potential financial upside for a working site.
The next test is whether H.E.R.O. can deliver on that promise as it reaches customers and becomes more durable and adaptable. If it does, a machine designed for sorting lines could eventually serve places that produce recyclable waste in the first place.