An article in Scientific Reports details a vision-guided robotic system designed to automate the weighing of aero-engine blades, a critical step in blade mass inspection and engine balancing.
Study: Vision-guided robotic system for aero-engine inspection and dynamic balancing. Image Credit: Fasttailwind/Shutterstock.com
The system features a cutting-edge end-effector equipped with a high-precision load cell and imaging sensor, enabling autonomous operation. Tested in industrial environments, the system achieved remarkable accuracy, demonstrating its readiness for integration into existing facilities without requiring significant reconfiguration.
Why Should We Automate Blade Weighing?
Industrial automation has significantly advanced precision tasks across various sectors. Robotic systems have enhanced efficiency in aerospace and automotive industries, excelling in machining, manipulation, and inspection processes. In aerospace, automated inspection has benefited greatly from computer vision techniques for surface assessments and fastener hole inspections. However, tasks involving physical interaction, such as aero-engine blade weighing, present unique challenges.
Weighing aero-engine blades is essential for mass inspection and engine dynamic balancing. This process requires exceptional precision, compounded by the intricate geometry of the blades. Currently, it is performed manually, which is time-intensive and prone to inconsistencies. While prior research has explored robotic manipulation and vision-based systems, they have not adequately addressed the stringent precision demands specific to blade weighing and sorting.
To address this gap, researchers developed an end-to-end vision-guided robotic system tailored for aero-engine blade weighing. The system combines an innovative end-effector with an imaging sensor and a high-precision load cell to enable accurate detection, manipulation, and mass measurement. Extensive testing confirmed its potential to deliver precise and efficient automation for aerospace production lines.
A Closer Look at the System
The proposed robotic system was developed to automate the weighing of aero-engine blades with high precision and efficiency. It utilized an ABB-1600-10 industrial robot paired with an EGP 40-N-N-B Schunk gripper and an S-Beam load cell to ensure precise handling and weighing of blades.
A depth-sensing Intel RealSense D405 camera was employed for blade detection and localization, alongside the YOLOv5s artificial neural network for object detection and an ArUco fiducial marker for 3D positioning. The system followed a sequential workflow: it detected blades, refined their positions, grasped them using the gripper, measured their mass, and then sorted them accordingly.
The design emphasized adaptability, allowing integration into existing industrial setups without requiring substantial environmental modifications. Additionally, the system included mechanisms to handle positional uncertainties and ensure consistent performance across varying conditions.
Results
The system underwent eight experimental trials involving 22 aero-engine blades, testing its performance in simulated industrial conditions. Localization precision demonstrated an aggregate standard deviation of 0.42 millimeters (mm), enabling accurate grasping and mass measurement. Mass measurement results showed a standard deviation of 0.0404 grams (g) and a mean absolute error of 0.0252 g, comparable to a calibrated balance.
A bias compensation mechanism effectively addressed load cell deviations, ensuring reliable measurements even with positional uncertainties of up to 5 mm during blade grasping. The system maintained robust performance in all tested scenarios, achieving a processing time of approximately 25 seconds per blade.
This automated approach streamlined the weighing process by integrating detection, grasping, weighing, and sorting into a seamless workflow. It minimized human error, reduced delays, and logged data in real-time, highlighting its potential for enhancing production efficiency in aerospace settings.
Conclusion
The vision-guided robotic system presented in this study offers a practical solution for automating aero-engine blade weighing. By combining precision-focused end-effectors and advanced imaging technology, the system addresses key challenges in blade mass inspection and engine balancing. Testing confirmed its high accuracy, repeatability, and efficiency, making it a viable option for modern aerospace production lines.
This system not only minimizes reliance on manual processes but also ensures compliance with stringent aerospace standards. With a processing time of just 25 seconds per blade and adaptability to existing setups, it represents a significant advancement for the aerospace industry.
Journal Reference
Ramadan, M., Youssef, A., Ayyad, A., Laith AbuAssi, Hay, O. A., Salah, M., Moyo, B., Yahya Zweiri, & Abdulrahman, Y. (2024). Vision-guided robotic system for aero-engine inspection and dynamic balancing. Scientific Reports, 14(1). DOI:10.1038/s41598-024-80540-w https://www.nature.com/articles/s41598-024-80540-w
Disclaimer: The views expressed here are those of the author expressed in their private capacity and do not necessarily represent the views of AZoM.com Limited T/A AZoNetwork the owner and operator of this website. This disclaimer forms part of the Terms and conditions of use of this website.