Danu's recycling robot uses computer vision to pick materials from mixed waste

TechCrunch reported on October 9 that Edinburgh startup Danu Robotics has spent six years building H.E.R.O., a pincer-claw robot that uses computer vision to identify and pick material from mixed recycling. Danu says it has letters of interest from two large customers, $500,000 in signed contracts, and a pipeline of more than 200 sales. The company is seeking a $5 million late-seed round.

Danu estimates that a site could gain $485,000 in additional revenue against $160,000 in capital cost and $24,000 in maintenance. Its technology page describes bin-level pick measurement, a conveyor footprint under one metre, and a plug-and-play installation. Glacier and Recycleye are named alternatives in the market, so these economics need to be compared with actual uptime and material prices, not only a model demo.

The figures are company estimates and the sales pipeline is not revenue. Hardware builders should ask for picks per hour, contamination rates, downtime, maintenance records, and a signed customer reference. The interesting software lesson is continuous improvement tied to a physical result, but the result must be measured at each site.