AI-DISCO Project Launched as First Module of AI & Microelectronics Research Factory

The Federal Ministry of Research, Technology, and Space has approved €15 million in funding for the AI-DISCO project, the first module of the Research and Innovation Factory AI & Microelectronics. The initiative is designed as a nationwide research structure linking artificial intelligence and microelectronics. The project consortium is led by IHP – Leibniz Institute for Innovative Microelectronics, with participation from Fraunhofer Institute for Photonic Microsystems IPMS.

The project consortium is led by IHP – Leibniz Institute for Innovative Microelectronics. Fraunhofer Institute for Photonic Microsystems IPMS is contributing technical expertise related to AI-based sensor systems and edge computing technologies.

The project focuses on addressing challenges associated with real-time data analysis in applications such as smart cities, critical infrastructure, Industry 4.0, and energy systems. These applications often involve the transfer of large volumes of data between devices and data centres, which can increase energy consumption.

AI-DISCO aims to develop an open and reconfigurable platform that supports energy-efficient distributed AI processing between edge systems and cloud environments. The approach involves deploying intelligent edge nodes that process sensor data directly at the point where it is generated. This method allows local real-time data processing and reduces the amount of information transmitted to cloud systems.

The project builds on earlier developments including the HARMMONAI AI accelerator circuit, an RRAM-based AI accelerator, and a spiking neural network architecture. The technical platform incorporates RISC-V-based edge hardware combined with energy-efficient AI accelerators. Additional methodologies being applied include federated learning, which enables distributed AI model training without centralised data collection, and neuromorphic AI techniques. Application scenarios also include integrated communication and sensing technologies associated with ISAC and 6G systems.

Fraunhofer Institute for Photonic Microsystems IPMS is contributing research related to intelligent sensor systems and neuromorphic computing hardware. The institute’s work within the project includes the development of energy-efficient hardware components and demonstration systems designed for applications in industry, agriculture, and medical technology.

The institute is also working on intelligent sensor systems with low-latency interfaces connected to the edge AI platform. Demonstration systems include a miniaturised NIR spectrometer designed to determine fruit ripeness and an ultrasound eye-tracking system integrated into glasses for use in medical applications and human-machine interaction.

In addition, the institute is developing neuromorphic AI processors based on FeRAM technology. These processors combine computing and memory functions using an in-memory computing approach. The processors are manufactured using 22 nm FDSOI technology and undergo characterisation and optimisation to support energy-efficient edge AI systems.

“With our research into neuromorphic hardware and intelligent sensor integration, we are laying the technological foundation for powerful and energy-efficient AI systems. At the same time, we are making an important contribution to strengthening Lusatia as a forward-looking high-tech location,” explains Sebastian Meyer.