CTC GmbH - Development of Solutions for Smart Factory Using Machine Learning Algorithms
Description of the Company
Composite Technology Center (CTC) is a 100% subsidiary of Airbus and located in Stade, Germany. We focus on research, development, production and education in the field of Fiber Composites, lightweight construction technologies, digitalization and automation of production technologies, robotics and additive manufacturing. We work for industries in areas of aerospace, automotive, transportation, mechanical and plant engineering, ship building, railways and wind power - to name a few.
There is a need to implement a vision system with machine learning (ML) algorithms to detect specific objects at the shop floor level and to use this information to automate machine to machine communication for increasing overall production efficiency.
The workers on the shop floor use conventional techniques for quality and progress detection. To automate this feature a use of vision system with ML algorithms running in the background is required which can efficiently distinguish between different workpieces. Moreover, it is necessary to make the results of object detection available to the next machine using a standard industrial protocol. With several ML frameworks available, it is also vital to test a couple of them and then benchmark them on basis of their features and extensibility for use in production of composite parts.
Aims of the Project
Develop or use object detection ML algorithms in different frameworks to detect specific objects. Make the information from this model available to another machine on the shop floor. Explain the advantages and disadvantages of the models and frameworks used as well as suggest how it can be extended to incorporate image inpainting in the system.
Object detection with pre-trained and then customized data sets using multiple ML frameworks Using standard industrial protocol to program interface to communicate the information from ML model to other machines/devices. To research in area of background image inpainting.
Target Group (Students)
Mechatronics, Information Technology, Computer Science