Measurement science for 6DOF object pose ground truth

Users of perception systems in industrial manufacturing applications need standardized, third party ground truth procedures to validate system performance before deployment. Many manufacturing robotic applications require parts and assemblies to be perceived, inspected or grasped. These applications need accurate perception of object pose to six degrees of freedom (6DOF) in X, Y, Z position with roll, pitch and yaw. A standardized 6DOF ground truth system should include test procedures, algorithms, artifacts, fixtures, and measurement equipment. Each of them must be openly documented so manufacturers, vendors, and researchers can recreate and apply the procedures. This article reports on efforts to develop an industrial standard for 6DOF pose measurement. It includes the design of test methods using a laser-tracker, an aluminum fixture pose fixture, and a modular, medium density fiberboard (MDF) pose fixture.

[1]  Mili Shah.  Comparing two sets of corresponding six degree of freedom data , 2011, Comput. Vis. Image Underst..

[2]  Joseph A. O'Sullivan,et al.  Pose estimation of known objects during transmission tomographic image reconstruction , 2006, IEEE Transactions on Medical Imaging.

[3]  Hanspeter Pfister,et al.  Fast and automatic object pose estimation for range images on the GPU , 2009, Machine Vision and Applications.

[4]  C. English,et al.  Real-Time Dynamic Pose Estimation Systems in Space : Lessons Learned for System Design and Performance Evaluation , 2011 .

[5]  Jeremy A. Marvel,et al.  Ground truth for evaluating 6 degrees of freedom pose estimation systems , 2012, PerMIS.

[6]  Nancy M. Amato,et al.  A Roadmap for US Robotics - From Internet to Robotics 2020 Edition , 2021, Found. Trends Robotics.

[7]  Henrik I. Christensen,et al.  Robust 3D visual tracking using particle filtering on the SE(3) group , 2011, 2011 IEEE International Conference on Robotics and Automation.

[8]  Michał Maniowski,et al.  Verification of 6DOF platform with wire-based sensors for spatial tracking , 2011 .

[9]  Jeremy A. Marvel,et al.  2011 solutions in perception challenge performance metrics and results , 2012, PerMIS.

[10]  Andrew Zisserman,et al.  Progressive search space reduction for human pose estimation , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Hiroyuki Yamamoto,et al.  A registration evaluation system using an industrial robot , 2006, 2006 IEEE/ACM International Symposium on Mixed and Augmented Reality.

[12]  Supun Samarasekera,et al.  Pose estimation, model refinement, and enhanced visualization using video , 2000, Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662).

[13]  Jeremy A. Marvel,et al.  Coordinate Metrology and the Proposed E57.02 Static Pose Determination Standard | NIST , 2013 .

[14]  Rama Chellappa,et al.  Fast object localization and pose estimation in heavy clutter for robotic bin picking , 2012, Int. J. Robotics Res..

[15]  C. Light,et al.  Establishing a standardized clinical assessment tool of pathologic and prosthetic hand function: normative data, reliability, and validity. , 2002, Archives of physical medicine and rehabilitation.

[16]  Mili Shah,et al.  Methodology for evaluating static six-degree-of-freedom (6DoF) perception systems , 2010, PerMIS.

[17]  Gary R. Bradski,et al.  Fast 3D recognition and pose using the Viewpoint Feature Histogram , 2010, 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems.