AIA

AI in Automation provide Applications and Limitations of AI Methods in Automation Systems:

Foundations of LLMs and agentic AI: from automata to autonomous systems, degrees of autonomy.
Prompt engineering and structured outputs: prompt patterns, few-shot and chain-of-thought prompting.
Tool calling and function calling: tool design, argument validation, natural-language-to-SQL.
Retrieval-Augmented Generation (RAG) and vector databases: embeddings, chunking, similarity search, advanced retrievers.
Agentic design patterns: reflection, ReAct, planning and replanning.
Stateful and multi-agent systems: LangGraph, communication and orchestration patterns.
Integration with digital twins and industrial interfaces: Model Context Protocol (MCP), OPC UA, Asset Administration Shell.

In-person lecture with integrated tutorial
Lectured in English on Thursdays from 12:15 p.m. to 1:45 p.m. in Buildung A5 1 - Room I (-1.22)
First lecture on Thursday, 22.10.2026, 12:15 p.m. to 1:45 p.m.
Exercise on Monday from 2:15 p.m. to 3:45 p.m. (every two weeks)
First exercise on Monday, 26.10.2026, 2:15 p.m. to 3:45 p.m.
We use Microsoft Teams for organization, scheduling practice sessions, and distributing materials. To participate in the event in the 2026/2027 winter semester, please join the team [166167] WiSe26/27 AI in Automation in Microsoft Teams using the code uhnfs6x