Abstract
The challenge of getting machines to understand and interact with natural objects is encountered in important areas such as medicine, agriculture, and, in our case, slaughterhouse automation. Recent breakthroughs have enabled the application of Deep Neural Networks (DNN) directly to point clouds, an efficient and natural representation of 3D objects. The potential of these methods has mostly been demonstrated for classification and segmentation tasks involving rigid man-made objects. We present a method, based on the successful PointNet architecture, for learning to regress correct tool placement from human demonstrations, using virtual reality. Our method is applied to a challenging slaughterhouse cutting task, which requires an understanding of the local geometry including the shape, size, and orientation. We propose an intermediate five-Degree of Freedom (DoF) cutting plane representation, a point and a normal vector, which eases the demonstration and learning process. A live experiment is conducted in order to unveil issues and begin to understand the required accuracy. Eleven cuts are rated by an expert, with being rated as acceptable. The error on the test set is subsequently reduced through the addition of more training data and improvements to the DNN. The result is a reduction in the average translation from 1.5 cm to 0.8 cm and the orientation error from 4.59° to 4.48°. The method’s generalization capacity is assessed on a similar task from the slaughterhouse and on the very different public LINEMOD dataset for object pose estimation across view points. In both cases, the method shows promising results. Code, datasets, and other materials are available in Supplementary Materials.
Highlights
The replacement of human muscles and minds across all sectors with repetitive, dangerous, and dirty jobs attracts much interest from both governments and industry
Our pose prediction framework can be considered as two separate systems: a training system for collecting point cloud and tool pose pairs as well as learning a mapping from the point cloud to tool pose; another system for performing inference based on the learned mapping and sending actionable 6-Degree of Freedom (DoF) poses to a robot equipped with the cutting tool
In order to investigate the potential of the method and reveal issues that could guide future improvements, the method was applied on two vastly different public datasets typically used for object pose estimation: LINEMOD [8] and Dex-Net [36]
Summary
The replacement of human muscles and minds across all sectors with repetitive, dangerous, and dirty jobs attracts much interest from both governments and industry. The difficulty of automating these jobs is explained by Moravec’s paradox It observes that machines with relative ease can be made to perform tasks that we perceive as complex or difficult, while great difficulty is associated with automating tasks that, on the surface, seem intuitive and easy [1]. This might stem from the fact that perception and motor skills seem easy because they have been acquired through more than a billion years of evolution, while more abstract skills like math and logic only recently have become relevant. This is manifested by a sensitivity to what should be inconsequential changes in the Sensors 2020, 20, 1563; doi:10.3390/s20061563 www.mdpi.com/journal/sensors
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
