AiVibeRobot Repeatability and Accuracy
Repeatability is how closely a robot returns to the same taught position over repeated cycles, while absolute accuracy is how closely it reaches a position commanded in coordinates. Industrial robots are typically far more repeatable than accurate.
ISO 9283 defines performance criteria and test methods for industrial manipulators, including pose accuracy, pose repeatability, path accuracy and path repeatability, measured under specified loads, speeds and test positions. Repeatability is mainly affected by gear backlash, encoder resolution, control quality and thermal effects. Absolute accuracy also depends on how well the kinematic model matches the real robot's link lengths, offsets, deflections and mounting.
Taught-point applications such as pick and place rely on repeatability, because each position was recorded on the real robot. Offline programming, vision guidance and CAD-based paths command positions in coordinates, so they rely on absolute accuracy and often need calibrated robots.
Datasheet repeatability values are typically measured under ISO 9283 conditions and may not reflect performance at full speed, with a heavy tool or before the robot has warmed up. Path accuracy matters for welding and dispensing. Laser trackers and similar instruments are used to measure performance and to calibrate robots for higher absolute accuracy.
Key points
- Repeatability means returning to taught points; accuracy means reaching commanded coordinates
- ISO 9283 defines the performance criteria and test methods
- Offline programming and vision guidance depend on absolute accuracy
- Calibration improves absolute accuracy
Where AiVibe comes in
AiVibe designs and manufactures the AiAmbA AI Factory, whose edge devices and AI agents let people talk to robot controllers in plain language. Robotics perception is an AiAmbA use case, and robot safety functions follow ISO 10218 and never depend on the AI layer.