Industrial robots are characterized by reliable and precise process execution. To ensure the high level of accuracy required for this, the systems must be individually recalibrated at regular intervals. This is costly and time-consuming and means a considerable amount of extra work, especially for SMEs. In addition, more and more inexpensive robot arms are coming onto the market, which potentially have even greater positioning inaccuracies for mechanical reasons.

Only geometric errors can be corrected with currently available calibration methods. However, temperature or load-dependent inaccuracies, for example, can only be compensated for inadequately. Recalibration during operation, which would be important for a sustainable optimization process, is also not feasible.

To close these gaps and use machine learning to develop new software-driven calibration methods for practical use, robotics expert ArtiMinds Robotics, the University of Stuttgart and DHBW Karlsruhe launched the AI project Kirk in April 2020.

The Kirk AI project was launched in April 2020 to develop new software-driven calibration methods using machine learning. © ArtiMinds

Darko Katic, technical contact for the project and Artificial Intelligence team leader at ArtiMinds, explains: "The ability to capture and analyze data automatically reduces the effort for the user and makes it easier for SMEs in particular to build up the necessary expertise to make optimal use of a robot system." The aim is to increase accuracy with software support in order to be able to use robots flexibly for a wide range of applications, to simplify work processes with a solution that is independent of the robot type and manufacturer and to reduce the workload on specialist personnel.

"Deep neural networks (deep learning) form the basis for making the complex relationships between external factors and the time-varying properties of the individual robot controllable and thus increasing positioning accuracy," says AI researcher Prof. Marco Huber from the IFF at the University of Stuttgart.

The Institute for Industrial Manufacturing and Factory Operation (IFF) at the University of Stuttgart and the Robot and Human Motion Lab (RaHM Lab) at the Baden-Württemberg Cooperative State University Karlsruhe are responsible for the basic research in the project. Together with ArtiMinds Robotics as an industrial partner, the results will be transferred to real industrial applications. Finally, the newly developed methods will also be integrated into the Robot Programming Suite (RPS) programming software. The project is scheduled to end in spring 2022. as