A recycling robot is in our sights thanks to MIT’s Computer Science and Artificial Intelligence Lab (CSAIL). The bot, dubbed RoCycle, has soft grippers that can pick up objects from a conveyer belt and identify the object based on touch.
The tactile sensors and sensorized gripper can detect the difference between paper, metal, and plastic.
“Although environmental and sustainability concerns have made it crucial to scale up recycling operations, object sorting remains a critical bottleneck for recycling scalability,” the researchers write in their paper.
The current method of sorting recyclables comes with numerous challenges. Today, many large recycling centers use magnets to pull out metals and air filters that sort paper from heavy plastics. Ultimately, most of this is done by hand, which can be dangerous work.
The researchers say the “failure to properly sort materials for recycling leads to waste; in the United States, 25 percent of all recycled materials are so contaminated they must be sent to landfills.”
RoCycle would eliminate the labor-intensive process and enable autonomous recycling.
“This classifier works over a variety of objects,” the researchers write, “including those that would fool a purely vision-based system.” Instead of relying on vision, the system can differentiate a metal object from one that looks like metal by sensing its conductivity.
The video below shows how the recycling robot works its magic, and the authors say the “materials classifier has 85 percent accuracy with a stationary gripper and 63 percent accuracy in a simulated recycling pipeline.”
But, this recycling robot still has a lot to learn and improve upon.
“Though RoCycle is already pretty impressive, as the video shows, it could do better. The research team plans to meld RoCycle’s sense of touch with video input from cameras to improve sorting accuracy,” Amanda Kooser says in CNET.