Robotic systems in manufacturing applications commonly assume known object geometry and appearance. This simplifies the task for the 3D perception algorithms and allows the manipulation to be more deterministic. However, those approaches are not easily transferable to the agricultural and food domains due to the variability and deformability of natural food. We demonstrate an approach applied to poultry products that allows picking up a whole chicken from an unordered bin using a suction cup gripper, estimating its pose using a Deep Learning approach, and placing it in a canonical orientation where it can be further processed. Our robotic system was experimentally evaluated and is able to generalize to object variations and achieves high accuracy on bin picking and pose estimation tasks in a real-world environment.
Individual manipulators are limited by their vertical total load capacity. This places a fundamental limit on the weight of loads that a single manipulator can move. Cooperative manipulation with two arms has the potential to increase the net weight capacity of the overall system. However, it is critical that proper load sharing takes place between the two arms. In this work, we outline a method that utilizes mechanical intelligence in the form of a whiffletree. This system enables load sharing that is robust to position deviations between the two arms. The whiffletree utilizes pneumatic tool changers which enable autonomous attachment/detachment. We outline the overall design of a whiffletree for dual-arm manipulation. We also illustrate how this type of mechanical intelligence can greatly simplify cooperative control. Lastly, we use physical experiments to illustrate enhanced load capacity. Specifically, we show how two UR5 manipulators can re-position a 7 kg load. This load would exceed the weight capacity of a single arm, and we show that the average forces on each arm remain below this level and are relatively evenly distributed.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.