Data- and simulation-driven transfer to the Transformer-Cell (C05)

• Subject Area: Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
• Term: since 2024
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 471687386

The goal of this subproject is to enable the transfer of the learned task models from the learning cell (C04) to the kinematics of the Transformer-Cell (C01). For this purpose, a data-driven simulation of the Transformer-Cell will be developed (WP1) and the task models will be refined using interactive reinforcement learning (RL) in the developed simulation (WP2). In WP3, using the interactive RL algorithms of WP2, geometric relations of the motion and force profiles of the task models are applied to the kinematics of the Transformer-Cell. WP4 deals with meta-learning and the use of the knowledge base to transfer new task models to the Transformer-Cell more quickly.

• DFG Programme: Collaborative Research Centres
• Subproject of: SFB 1574: A Circular Factory for the Perpetual Product
• Applicant Institution: Karlsruher Institut für Technologie
• Project Head: Professor Dr. Gerhard Neumann