Learn more about the use case:
Research Project: Software Engineering for AI, Robotics/Autonomous Systems
AI-based robot calibration (KIRK)
What drove the decision to develop an AI-based use case?
Multi-axis, collaborative industrial robot arms are the key to flexible and cost-effective automation. A common obstacle here is the accuracy that the robot can achieve. Until now, high-precision applications have required complex calibration procedures and costly mechanics. However, such calibrations have to be repeated after a few years, for example due to wear and tear, and are highly dependent on external influences such as temperature. The relationships are strongly nonlinear and difficult to model.
The goal of the project is to increase the accuracy of the robot cost-effectively via software. Therefore, stepwise error compensation and calibration methods based on deep artificial neural networks will be developed and put into practice.
What were your expectations or requirements for the AI application?
The technology to be developed must be easy to use and quick to set up. In particular, it must be usable by users in the factory after brief instruction.
Which partner did you rely on with which technology for the development?
The project is conducted in collaboration with the Institute of 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. We rely on deep neural networks to learn the complex relationships from external factors as well as the time-varying characteristics of the individual robot to increase positioning accuracy.
The use of AI solutions must also be accompanied by a change in competences. How have you dealt with this challenge in respect of your employees?
Both ArtiMinds and the consortium partners have many years of experience in the field, so no change was necessary.
In what timeframe did they implement the AI solution in your company?
The project is designed for the period 01.04.2020 - 31.03.2022.
A conclusion to your AI solution:
The clear advantage of the learning-based approach in this project is that subtle interrelationships that are difficult to identify and even more difficult to model manually can be learned well. The ability to capture and process data in an automated fashion reduces user effort and makes it easier for SMEs to build the necessary competency to use the system. This fits perfectly into the philosophy of ArtiMinds. After all, the goal of the ArtiMinds Robot Programming Suite (RPS) is to reduce the qualifications required for robot programming. This is particularly important for (producing) SMEs, as they often do not want to or cannot work directly with AI algorithms. The project explicitly addresses this challenge and represents a superiority of the solution approach with respect to alternative approaches.
One final piece of advice/tip to other entrepreneurs looking to apply AI:
AI methods form new opportunities and make it possible to tackle projects that cannot be implemented economically and robustly using classic means. The technology has now also reached the stage where it can be safely used in demanding real-world applications. Investing in employee training and cultivating strategic partnerships in this area is certainly worthwhile.
More info at: