Stereo Vision Feedback for High-Accuracy Sample Positioning Using a Robotic Arm at ESS
(2026)Department of Automatic Control
- Abstract
- At the European Spallation Source (ESS), the Beamline for European materials Engineering Research (BEER) instrument aims to position samples within the neutron beam using a robotic arm system, given the high flexibility and large end-effector workspace provided by robotic arms. However, precise and accurate sample positioning is limited by the pose accuracy capabilities of industrial robotic arms, particularly in the presence of heavy payloads. The accuracy requirement of the sample positioning is directly related to the resolution of the instrument’s imaging as the sample is moved instead of the detectors. This master’s thesis presents the development of a high-accuracy sample-positioning system for ESS that is robust to variations in... (More)
- At the European Spallation Source (ESS), the Beamline for European materials Engineering Research (BEER) instrument aims to position samples within the neutron beam using a robotic arm system, given the high flexibility and large end-effector workspace provided by robotic arms. However, precise and accurate sample positioning is limited by the pose accuracy capabilities of industrial robotic arms, particularly in the presence of heavy payloads. The accuracy requirement of the sample positioning is directly related to the resolution of the instrument’s imaging as the sample is moved instead of the detectors. This master’s thesis presents the development of a high-accuracy sample-positioning system for ESS that is robust to variations in sample mass distribution.
The system consists of the following components: a Stäubli TX60 industrial robot arm; two monochrome Atlas ATL120S-MC cameras equipped with Schneider‑Kreuznach TOURMALINE F2.8/40 mm 1.3″ lenses; a high-precision ceramic ChArUco marker; an external stereo-vision-based pose estimation method solving the 2-PnP problem; and communication and control implemented in ROS.
The proposed system introduces an external feedback control loop around the robot control system to compensate for uncertainties in geometric positioning of the robot and sample. Robot end-effector pose estimation is achieved using a highaccuracy fiducial marker in conjunction with a stereo vision system. The system was evaluated using a laser tracker. The experiments demonstrated a reduction in median positioning deviation across different payloads from 175 μm to 12 μm, and an improvement in end-effector pose accuracy from 70 μm to 35 μm. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/student-papers/record/9251310
- author
- Andersson, Gabriel and Sjöstedt, Pontus
- supervisor
- organization
- year
- 2026
- type
- H3 - Professional qualifications (4 Years - )
- subject
- report number
- TFRT-6324
- other publication id
- 0280-5316
- language
- English
- id
- 9251310
- date added to LUP
- 2026-09-18 11:44:37
- date last changed
- 2026-10-01 16:03:37
@misc{9251310,
abstract = {{At the European Spallation Source (ESS), the Beamline for European materials Engineering Research (BEER) instrument aims to position samples within the neutron beam using a robotic arm system, given the high flexibility and large end-effector workspace provided by robotic arms. However, precise and accurate sample positioning is limited by the pose accuracy capabilities of industrial robotic arms, particularly in the presence of heavy payloads. The accuracy requirement of the sample positioning is directly related to the resolution of the instrument’s imaging as the sample is moved instead of the detectors. This master’s thesis presents the development of a high-accuracy sample-positioning system for ESS that is robust to variations in sample mass distribution.
The system consists of the following components: a Stäubli TX60 industrial robot arm; two monochrome Atlas ATL120S-MC cameras equipped with Schneider‑Kreuznach TOURMALINE F2.8/40 mm 1.3″ lenses; a high-precision ceramic ChArUco marker; an external stereo-vision-based pose estimation method solving the 2-PnP problem; and communication and control implemented in ROS.
The proposed system introduces an external feedback control loop around the robot control system to compensate for uncertainties in geometric positioning of the robot and sample. Robot end-effector pose estimation is achieved using a highaccuracy fiducial marker in conjunction with a stereo vision system. The system was evaluated using a laser tracker. The experiments demonstrated a reduction in median positioning deviation across different payloads from 175 μm to 12 μm, and an improvement in end-effector pose accuracy from 70 μm to 35 μm.}},
author = {{Andersson, Gabriel and Sjöstedt, Pontus}},
language = {{eng}},
note = {{Student Paper}},
title = {{Stereo Vision Feedback for High-Accuracy Sample Positioning Using a Robotic Arm at ESS}},
year = {{2026}},
}