Picking a Conveyor Clean by an Autonomously Learning Robot
This addresses industrial waste sorting automation, but it appears incremental as it builds on existing methods with learning enhancements.
The researchers tackled the problem of autonomous waste sorting by developing a robot that learns to pick objects from a conveyor belt with minimal human intervention, achieving a success rate of depositing 70 out of 80 objects from a difficult pile into the correct chute.
We present a research picking prototype related to our company's industrial waste sorting application. The goal of the prototype is to be as autonomous as possible and it both calibrates itself and improves its picking with minimal human intervention. The system learns to pick objects better based on a feedback sensor in its gripper and uses machine learning to choosing the best proposal from a random sample produced by simple hard-coded geometric models. We show experimentally the system improving its picking autonomously by measuring the pick success rate as function of time. We also show how this system can pick a conveyor belt clean, depositing 70 out of 80 objects in a difficult to manipulate pile of novel objects into the correct chute. We discuss potential improvements and next steps in this direction.