AIMS 5.0: AI-Supported Automation for the Industry 5.0 Factory
The EU research project AIMS 5.0 (Artificial Intelligence in Manufacturing leading to Sustainability and Industry 5.0) is one of the key European initiatives aimed at strengthening digital sovereignty in semiconductor manufacturing. Between 2023 and 2026, 53 partners from 12 countries will develop new AI-enabled components, software platforms, and automation solutions for productive, resilient, and sustainable industrial production.
The project is coordinated by Infineon Technologies, with a significant contribution from its Dresden site, among others. Technologies are being validated directly in real factory environments across multiple use cases—ranging from AI-supported process optimization to autonomous logistics in cleanrooms. Fabmatics plays a key role in the areas of “Smart Transport & Storage Concepts for Future AI-based Automation Systems” and “AI-based Control & Virtual Commissioning for Mobile Systems”: In addition to new storage solutions such as the FlatStocker 3.0 and digital twins for automation systems, the project’s contribution specifically includes the development of a next-generation mobile robot system—the HERO Scout—featuring AI-enabled control software and auto-teaching capabilities for applications such as reticle logistics.
Reticle Logistics as a Key Process in Lithography
Lithography is one of the most cost-intensive and time-critical areas in semiconductor manufacturing. Here, photomasks (reticles) are used to project ultra-fine micro- and nanostructures onto the wafer—often across several dozen layers. Any delay or supply failure at the exposure systems directly impacts throughput, OEE (Overall Equipment Effectiveness), and the yield of the entire production process.
Therefore, mask logistics must operate with a high degree of stability and synchronization in parallel with wafer WIP logistics. Manual handling of reticle pods and wafer cassettes carries risks: errors, wait times, and ergonomically unsound workstations negatively impact both productivity and staff. This is precisely where the use case within the AIMS 5.0 framework comes into play: autonomous, high-precision reticle transport via an AMR that can be flexibly integrated into existing cleanroom infrastructures.
HERO Scout: Compact AMR for precise reticle and wafer handling
The HERO Scout is an Autonomous Mobile Robot (AMR) developed specifically for cleanrooms in the semiconductor industry, featuring an integrated 6-axis cobot arm. Its design meets clear practical requirements—defined jointly by Infineon Dresden and Fabmatics: high functionality, performance, maintainability, cost-efficiency, and maximum flexibility in lithography environments.
Key technical features:
- Compact footprint
Ultra-small footprint (approx. 66 × 56 cm) for use in narrow, winding cleanroom sections
- Long reach & payload
Cobot reach of approximately 200 to 1,700 mm, designed for elevated load ports, with a payload of up to 20 kg
- Active, camera-guided gripper
Adaptive gripper jaws for reticle pods and open cassettes, camera-integrated position detection, and 6D offset correction. Payloads can be securely gripped even if load ports have been manually repositioned afterward.
- 3D Vision & Safety Features
3D cameras detect obstacles (e.g., chairs, tables, displays, keyboards), as well as negative obstacles such as missing floor panels. Safety features with sensitive collision detection enable safe operation in areas with high foot traffic.
- High Driving Dynamics with Vibration Control
Travel speeds of up to 1.2 m/s, optimized vibration damping even on perforated raised floors.
- Inductive Charging & Auto-Teaching
On-the-fly charging via an integrated charging coil, as well as rapid map training and roadmap learning through SLAM-based navigation and auto-teaching functions.
On the software side, Fabmatics relies on a newly developed, AI-capable local control architecture that allows the HERO Scout to operate as an intelligent unit within the fab. At the same time, Fabmatics’ proprietary fleet manager was enhanced with AI capabilities to efficiently coordinate a fleet of AMRs in the future—including dynamic route planning and task distribution.
Real-world testing at Infineon’s 200 mm fab in Dresden
To validate the HERO Scout, the prototype was deployed in Infineon’s 200 mm lithography production facility in Dresden under real cleanroom conditions. The tests covered key application scenarios:
- Transport and handling of reticle pods and wafer cassettes
- Navigation through narrow and complex cleanroom environments
- Interaction with dynamic obstacles and pedestrian traffic
- Operation in extremely confined spaces
- Evaluation of driving stability at high speeds
The tests were conducted using dummy material to simulate real-world conditions without disrupting production operations.
Across all scenarios, the HERO Scout demonstrated high reliability in navigation, even in narrow and dynamic cleanroom areas. Its camera-guided gripping system proved robust against shifts in the load ports and enabled precise handling without the need for re-teaching. In addition, the tests confirmed safe interaction with the environment, including pedestrian traffic and variable obstacles. Driving stability, particularly on perforated raised floors, met the requirements, and vibration measurements remained well below the relevant limits. Furthermore, specific areas for optimization were identified, which will be incorporated into further development without compromising the robot’s fundamental operational readiness.
Significance for AIMS 5.0 and Automation in Semiconductor Factories
With the successful real-world test of the HERO Scout at Infineon Dresden, Fabmatics delivers a concrete, industry-relevant demonstrator within AIMS 5.0:
- Automated reticle transport as a critical lever for increasing OEE in lithography
- Scalable AMR platform with AI-enabled control architecture and fleet option
- Seamless integration into existing fab layouts—without fundamental infrastructure modifications
Looking ahead, the HERO Scout opens up the possibility for semiconductor fabs to gradually transition reticle and wafer logistics—from manual to autonomous processes—all the way to fully digitized, end-to-end automated material flows. In combination with other AIMS 5.0 results, such as digital twins and AI-enabled fleet control, this creates a key building block for the intelligent, connected, and sustainable factory of the future—made in Europe.
About AIMS5.0
The AIMS5.0 project is supported by the Chips Joint Undertaking and its members, including additional funding from national funding agencies of the participating countries under Grant Agreement No. 101112089.
About Fabmatics
Fabmatics GmbH, headquartered in Dresden, is a leading specialist in the automation of material flows and handling processes in the semiconductor industry and other high-tech production environments. Since its founding in 1991, the company has successfully implemented customized automation solutions and has established a strong market position worldwide, particularly in the modernization of 200-millimeter fabs.
Its portfolio comprises a modular range of hardware and software solutions for handling, transport, and storage processes, as well as systems for product identification and localization. These include, among other things, mobile robots, robotic cells, cleanroom storage systems, RFID technologies, and purge systems that increase efficiency, speed, and process reliability in production. Fabmatics combines all expertise in material flow automation under cleanroom conditions from a single source.
With approximately 300 employees and a subsidiary in the U.S., Fabmatics operates globally. Since 2023, the company has been part of the SCIO Automation Group, which operates in eight countries and provides intelligent automation solutions for leading industrial companies. Fabmatics is also an active member of international and regional industry networks such as SEMI, Silicon Saxony, and the AND Automation Network Dresden.
– – – – –
Related Links
👉 www.fabmatics.com
Photo: Fabmatics
