M-AIR | Multimodal AI for adaptive intralogistics and robotics

The M-AIR network brings together companies and research institutions that are jointly exploring new approaches for AI-based applications in production, logistics, and industry.
The following topics show possible project ideas that have emerged from previous exchange formats and workshops and could be further developed or explored in greater depth in the future.

• Voice-controlled mission definition for mobile robots
  The aim is to formulate complex logistics tasks more simply and translate them automatically into precise driving commands.
• Intelligent route planning for autonomous vehicles
  By evaluating sensor data, plans, and user instructions, vehicles could navigate more flexibly and safely in the future.
• Adaptive bin picking systems with multimodal AI
  The combination of computer vision, robotics, and language models could significantly expand automation in warehouse and production environments.
• AI-supported gripping point determination in robotics
  With the help of improved gripping decisions, robots could also act reliably with unknown objects in the future.
• Generative planning of industrial plant layouts
  Use of generative AI models for the automatic creation and evaluation of factory or logistics layouts.
• Digital remodeling of existing plants
  Conversion of 2D plans or scans into 3D models for efficient use in modernization or simulation.
• Automated safety assessment of technical facilities
  Automated assessment of risks and safety requirements.
• AI-supported management of product compliance
  Use of artificial intelligence to comply with regulatory requirements in global supply chains.