Vision-based Quality Assurance System for automated wind turbine blade repairs (drone-based / UAV)
Description
The company is developing robotic wind turbine blade repair tools mounted on a UAV as payload.
The Quality Assurance System (QAS) must be a part of the ROS2 application, which must record and analyse the video stream of the processed wind turbine blade.
This internship position will be responsible for the design, implementation, integration, and validation of the QAS.
The mandatory QAS results are the following:
- Recorded raw video stream
- Soft real-time blade damage classification (two classes)
The additionalQAS results are the following:
- After-processing blade damage classification/segmentation (five classes)
- Soft real-time grinding result quality classification
- After-processing grinding issues classification/segmentation
- Soft real-time coating result quality classification
- After-processing coating issues classification/segmentation
- Soft real-time speed of theUAV movement detection
- Soft real-time blade tip detection
- Soft real-time blade central line detection
- Soft real-time blade roll-angel classification (two classes)
- Presenting the real-time feed into the remote controller for the operator
The soft real-time result must be produced on the device without communication with cloud-based (or other external services). The after-processing results might be produced in cooperation with cloud-based services. It’s preferable to have zero-shot models for each result.
#software #ROS #ROS2 #AI #C++
Skills
Sign in to read and join the discussion.