Autonomous Robotic Inspection System for Drill Holes Tilt: Feasibility and Development by Advanced Simulation and Real Testing
Nour M. Morsi, Mario Ortí Mata, Colin Harrison, David Semple
- Year
- 2023
- Citations
- 2
Abstract
Industry 4.0 and forthcoming 5.0 introduce process methods that primarily rely on robotic systems to succeed. Robotic systems provide accuracy in inspection tasks over current methods. However, in the aerospace industry the required accuracy is beyond what a standard robotic arm can provide. This paper aims to introduce a simple finite state-controlled robotic system using image processing to guide the motion and evaluate the tilt of a series of drill holes using one of them as reference hole. The system is developed via simulation in RoboDK and then validated in both a simulated and a real setup, using an ABB GoFa robot. The inspection experiment is performed by aligning the camera optical axis to the first drill hole axis. By taking it as a reference, the rest of the holes are measured and compared with the reference one, and to the ground-truth results. Drill hole detection is refined using a proposed circle optimisation algorithm. Results demonstrate the validity of the approach with an error below 0.5 degrees.
Keywords
Related papers
Statistical Learning Theory
Yuhai Wu, Vladimir Vapnik
1999
Artificial intelligence: a modern approach
1995
Fractional Differential Equations
Igor Podlubný
2025
Applied Nonlinear Control
Jean-Jacques Slotine, Weiping Li
1991