Key points
- Repeatability and accuracy answer different questions; a robot can be very repeatable and less accurate.
- Universal Robots publishes ±0.03 mm pose repeatability for UR5e and cites ISO 9283.
- ABB publishes 0.02 mm repeatability for GoFa and, separately, 0.1 mm absolute accuracy with its Ultra Accuracy feature.
This explainer uses Universal Robots’ and ABB’s published specifications as worked examples and identifies ISO 9283 by its published title. It does not reproduce the standard’s test method.
Two different questions
Pose repeatability describes how closely a robot returns to the same pose when the same motion is repeated under the same conditions. Pose accuracy describes how closely the robot reaches a pose that has been commanded numerically, for example from a CAD model or an offline program.
The distinction matters because the two numbers can differ by a large factor. A point taught by jogging the arm to a fixture and recording it relies mainly on repeatability. A point calculated from a drawing relies on accuracy: the robot’s model of its own geometry has to match the real arm.
How manufacturers publish it
ABB’s own figures show the gap between the two measures. The same product family is published at 0.02 mm repeatability and at 0.1 mm absolute position accuracy with an accuracy-improving feature, a fivefold difference. The Universal Robots UR5e datasheet cited here publishes repeatability only.
| Robot | Published figure | Measure | Document |
|---|---|---|---|
| UR5e | ±0.03 mm | Pose repeatability, ISO 9283 | Universal Robots datasheet, September 2026 |
| GoFa 5 | 0.02 mm | Pose repeatability (standard not named) | ABB portfolio, 2023 |
| GoFa 5 | 0.05 mm | Position repeatability (RP) | ABB datasheet, 2021 |
| GoFa family | 0.1 mm | Absolute position accuracy with Ultra Accuracy | ABB GoFa product page |
Read the conditions behind the number
ISO 9283, titled “Manipulating industrial robots — Performance criteria and related test methods”, is the standard Universal Robots cites for UR5e repeatability. ISO lists the 1998 edition as last reviewed and confirmed in 2021.
A repeatability figure is measured under defined test conditions. Payload, speed, temperature and the part of the workspace used can all change what an installed cell achieves, so the datasheet figure is a starting point rather than a guarantee for your process.
When one manufacturer names the standard and another does not, treat the comparison as indicative. Check the document date as well: ABB’s 2021 and 2023 documents publish different GoFa repeatability figures.
What to ask for your application
Decide whether your points will be taught or calculated. For taught points, compare repeatability at your payload and speed. For offline-programmed or vision-guided points, ask the manufacturer for absolute accuracy data and the calibration options that support it, then validate the finished cell against your own tolerance.
Primary sources
- ISO 9283:1998 — Manipulating industrial robots — Performance criteria and related test methodsInternational Organization for Standardization · accessed 8 October 2026
- UR5e technical specification (online datasheet)Universal Robots · accessed 8 October 2026
- ABB collaborative robotics portfolioABB · accessed 8 October 2026
- GoFa CRB 15000 datasheet (Spanish edition)ABB · accessed 8 October 2026
- GoFa collaborative robotsABB Robotics · accessed 8 October 2026