Loading...
Thumbnail Image
Publication

Robotic antenna metrology range calibration, uncertainty evaluation, and scan generation to improve performance and flexibility

Moser, Benjamin L.
Citations
Altmetric:
Editor
Date
Date Issued
2023
Date Submitted
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Robotic positioning systems have recently been introduced in state-of-the-art antenna ranges for use during antenna metrology, a process with rigorous performance requirements to produce accurate field strength and polarization measurements. However, existing implementations rely on expensive and view-limited positional feedback systems, contain poorly defined pose (end effector position and rotation) uncertainties with impacts to final scan quality, and fail to fully utilize the motion capabilities of the robots by directly replicate the scanning motions of traditional antenna positioning systems with few degrees-of-freedom. These current constraints on robotic antenna ranges limit their adoption and flexibility. To enhance robotic antenna range performance and flexibility, we developed kinematics software and algorithms oriented toward high-DoF, multi-robot antenna range cells performing relative positioning tasks. We demonstrate the ability to reproduce the transforms of canonical scans, although antenna scanning with the implementation of these methods has not yet been evaluated due to current equipment limitations. The purpose of this work was to (1) establish and experimentally verify calibration approaches for the manipulator types found in current robotic antenna ranges, (2) characterize the post-calibration pose uncertainty of these manipulators, and (3) implement a framework for efficient antenna scan generation that incorporates secondary objectives. Through two experimental studies on a high-precision 7 DoF hybrid robotic configuration for high-frequency (80 GHz- 500 GHz) antenna scanning, we demonstrated improvements in positioning accuracy for complex manipulators and generated post-calibration pose uncertainty estimates with a parameterized measurement uncertainty fit. Pose error reductions of 46.7 % and 67.2 % from the un-calibrated experimental manipulator model were found in the first and second study, respectively. Jacobians and uncertainty models were incorporated into an inverse kinematics pipeline to generate a variety of canonical scan types using full manipulator motion and simulate scan motion. This approach permits larger scan geometries to be performed when compared to an existing implementation, typified by spherical scan geometry expanded from a maximum radius of 1 m to a 4 m. Secondary objectives are minimized in a hierarchical manner, with joint excursion reduced by an average of 34.6 %. The produced scans offered improved performance for maximum scan geometry and joint excursion when compared to existing implementation. These studies collectively advance the state-of-the-art of antenna metrology by enhancing multi-arm robotic positioning systems at the National Institute of Standards and Technology (NIST) and supports NIST's mission to provide traceable antenna calibration services for all sectors in the United States.
Associated Publications
Rights
Copyright of the original work is retained by the author.
Embedded videos