AROKI – Adaptive Robot Applications via AI-Based Perception
In the AROKI research project, an adaptive robotic application is being developed with the objective of automating the assembly of galvanic racks through perceptive and tactile sensing. The application incorporates artificial intelligence, which enables the system to adapt to real process conditions, thus facilitating a flexible response to changes.
Objective
The objective of this research project is to leverage advances in the field of artificial intelligence for industrial robotics in order to increase the autonomy and flexibility of automation solutions. End users should be able to generate application-specific synthetic training data and train intelligent agents for robot perception. To achieve this, a framework is being developed that largely automates the steps of synthetic data generation, the selection and parameterization of suitable AI methods and AI training. By providing application-specific planning data, a compromise between realism and randomization of the training data is to be achieved, which enables robust and at the same time precise perception models. The overarching goal is to use improved robot perception and sensor integration to realize new tasks that were previously difficult to automate, such as the loading of galvanic racks.
Main research topics
- Development of a parameterizable simulation environment for training AI models
- Creation of a user interface for the analysis of planning data using generative AI
- Development of a tactile gripper system for forceassisted assembly of deformable pins
Consortium
The consortium consists of three companies (Rohde & Schwarz Messgerätebau GmbH - Location Memmingen, Rohde & Schwarz GmbH & Co. KG - Location Teisnach, KABTEC AG), two associated partners (HEINE Optotechnik GmbH & Co. KG, Siemens AG) and the Institute for Machine Tools and Industrial Management (iwb) of the Technical University of Munich.
Acknowledgments
The Bavarian Ministry of Economic Affairs, Regional Development, and Energy funds the AROKI project, and the VDI/VDE Innovation + Technology GmbH administratively supports the project consortium. The iwb thanks all partners for the funding, the support, and the excellent cooperation in the project.