Work

Traceable Metrological Digital Twin for Robot-Based 3D Inspection: Calibration, Uncertainty Propagation, and Cross-Validation

Year: 2026
Type: article
Abstract: Digital Twins (DTs) are increasingly employed in manufacturing to support simulation, monitoring, and control tasks, but their use in dimensional metrology requires explicit traceability and uncertainty information. This work develops a traceable DT for robot-based 3D inspection by integrating an industrial articulated robot with a laser-line triangulation sensor. The methodology combines robot kinematic identification, optical-sensor characterization, sensor-to-robot referencing, and a machine-readable representation of the calibrated metrological state. Uncertainties associated with the calibrated parameters are propagated through the complete measurement chain using a Monte Carlo method aligned with the Guide to the Expression of Uncertainty in Measurement (GUM). Validation combines a CMM-characterized four-sphere artifact and a geo-metrically complex part. For the complex part, two complementary cross-validation routes are considered: dense-surface comparison against a calibrated Triple Scan reference and feature-level comparison against CMM/VCMM reference values and task-specific uncertainties over 67 evaluated characteristics. The results support the use of the DT to generate uncertainty-enriched point clouds while also showing that feature-dependent systematic differences are not fully represented by global point-cloud statistics. The current validation is restricted to the environmental conditions and short-term stability of the experimental campaign; thermal drift and long-term stability are not included in the present model. The proposed framework therefore provides a basis for traceable, uncertainty-aware robot-based inspection and for subsequent industrial qualification under variable operating conditions.
Source: Metrology
Language: English
Open Access status: gold
Estimated APC paid: $1,207
OpenAlex ID: W7236427807
FWCI:
Cites:
Cited by:
Related to: 10