91²Ö¿â

In 2011, the Finnish company ZenRobotics was first on the waste management scene with their robotic waste sorter. Their system used a combination of  computer vision, machine learning, artificial intelligence (AI) to run synchronized robotic arms to sort and pick recycled materials from moving conveyor belts. The company broke ground with recycling stations that were tied together in a neural network and the AI learned, from real-time feeds providing data from metal sensors, 3D laser cameras and spectroscopic cameras, to select and sort the right items from the belt.

Then came Spain. Sadako Technologies has been working on an AI-infused garbage sorting systems on a smaller scale than Zen Robotics. The company also uses proprietary algorithms that connect to a multi-layer neural network system which lives in the cloud.

Now, there’s Clarke. Clarke is a pilot program in Denver, Colorado that’s been learning to identify a wide variety of food and beverage cartons so it can grab and separate them from the rest of the recycling. The pilot is a collaboration by the Carton Council and two Denver-based companies, AMP Robotics and Alpine 91²Ö¿â & Recycling. The recycling robot is named after the British science fiction author and futurist Sir Arthur Charles Clarke.

According to Matanya Horowitz, founder, AMP Robotics, what’s novel about Clarke, is that the system uses an off-shelf robot that’s actually been in use for two decades in other industries.

“The differentiator in Clarke is in its computer vision system,” said Horowitz. “The AMP Cortex inside the robot utilizes advances in deep learning that first emerged in 2012 and which allowed robots to understand the environment around them. Other examples of these software capabilities are present today in self-driving cars and Facebook’s ability to recognize faces in photos.”

Since it was installed in late 2016, Clarke has been using AI to learn as it sorts. Clarke has constantly improved on its recycling skills, grabbing approximately 60 cartons per minute with near perfect accuracy. Clarke is always learning as it sorts and grabs in order to increase the recognition of a multitude of food and beverage cartons such as milk cartons, coconut water water and even wine containers. The Clarke system has the ability to sort at super-human speeds, diverting material for re-processing that might otherwise end up in landfill.

To read the full story, visit .

Sponsor