What we do
Challenges
Consumers increasingly expect manufacturing products to be free of defects which sets high standards for defects identification and repair in the production process. However, associated working processes are physically and cognitively demanding for workers and executed in a potentially hazardous environment. Currently, these operations require skilled and experienced workers to localize and classify defects in a short time, and to rework defects using grinding tools with a considerable physical effort.
Objectives
MAGICIAN activities aim at:
- developing robotic solutions to classify and rework defects from semi-finished products autonomously before the finalization of product aesthetics
- designing robotic solutions in a modular way to make them applicable for various purposes and needs
- improving safety conditions and reducing physical strain on workers while maintaining accuracy in operations
- applying a human-centred approach to the development of robotic solutions, by learning from human expertise and fostering trust in AI technologies
Solutions
MAGICIAN will develop two modular robotic solutions, a sensing robot for defect analysis (SR) and a cleaning robot for reworking defects (CR) which will use AI modules to perform associated operations. Data needed for these AI modules will be gathered by learning from workers operating on semi-finished products.
The MAGICIAN solutions will:
- contribute to a safer working environment as they enable workers to avoid tasks which are physically demanding and risk-prone
- allow workers to shift their focus to other tasks, including controlling the areas treated by the robot
- offer workers a support system for decision-making
MAGICIAN approach
The MAGICIAN project places the human at the center of activities in the development of robotic solutions and is guided by the ambition to design modular and generalisable solutions, allowing end-users to flexibly adapt the MAGICIAN solutions to their respective purposes and needs.
The robotic solutions will be tested in an automotive use-case and extended by engaging additional use cases through Open Calls to prove their general use for different applications.
Robotic solutions
For the development of the sensing robot (SR) and the cleaning robot (CR) modules, MAGICIAN partners will gather data from workers operating on semi-worked products to train the robots how to:
- identify and classify defects (SR)
- use the grinding tool to subsequently remove the defects (CR)
The robotic modules can each be used as independent solution, thus enabling the integration of external services and components, and the execution of certain tasks by human workers while others are executed by a robot.
The SR and CR will rely on the software services of a common robotic platform including both, the basic services needed to implement the respective module individually, and a common group of services allowing for the combined operation of the SR and CR.
Human-centred approach
MAGICIAN will apply a human-centred design strategy to shape the progress of automation and human-robot collaboration in manufacturing towards an emphasis on trust/comfort, empathy, ethics and an understanding of the overall impacts on humans and the working environment.
Input from all key stakeholders will be collected and inform the development process starting from the system design, down to the different interfaces. The aim is to understand the human experience, motivation and behaviour.
Another focus will be on integrating the potential of human-robot collaboration for upskilling workers, e.g. for instructing, supervising and/or engaging with robots.
MAGICIAN thereby strives to create technology solutions that all stakeholders can perceive as easy to use, reliable, safe and trustworthy, ultimately fostering a sustainable transition towards human-robot collaboration in manufacturing.
Automotive Use Case
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.
Use Case 2
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.
Use Case 3
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.
Use Case 4
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.
Use Case 5
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.
Use Case 6
The MAGICIAN solutions will be tested and piloted in a real operational environment provided by the automotive manufacturer Tofaş. The system will be integrated by Altinay.
Both, the sensing robot identifying and classifying defects, and the cleaning robot repairing defects, will be coupled with human operators during the testing with the worker either adding defect repair or defect identification to the overall process.
Testing, assessment and improvements will follow an iterative cycle.
The demonstrator will not only be evaluated in terms of robot performance, but also in terms of whether the technology developed succeeds in facilitating acceptability, interaction, and trust experienced by workers and end-users in their direct use of the system.