At the Hannover Messe 2026, Beckhoff showcased its vision for Physical AI, merging artificial intelligence with traditional machine control. The highlight was a demonstration where a large language model (LLM) directly controlled the modular industrial robot Atro through natural language commands.
Using the TwinCAT CoAgent for Operations and the Model Context Protocol (MCP), the control system acts as an intelligent agent, translating human speech into machine instructions. This enables intuitive human-robot collaboration, with the agent handling trajectory planning and diagnostics.
Product Manager Dr. Fabian Bause explained that standardised interfaces allow LLMs to influence real motion sequences seamlessly. The demo illustrates how future manufacturing can leverage AI for flexible, conversational programming of industrial equipment.