RESEARCH GROUP Cognitronics & Sensor Systems

VEDLIoT: Very Efficient Deep Learning in IoT

Autonomous vehicles and devices for intelligent homes are becoming increasingly complex. A new system based on machine learning is being designed to make the soft- and hardware used for these applications more robust, powerful, and energy-efficient.

 eProcessor: a European RISC-V processor

The eProcessor project aims to build a new open source processor and deliver the first open source European full-stack ecosystem based on this RISC-V CPU.

Edge computing modules for resilient electronics manufacturing with adaptive process optimization (AdaPEdge)

The project uses federated learning to perform AI-based analysis on data collected by a universal sensor network to monitor the status of manufacturing.

AI on the edge for a secure and autonomous distribution grid control with a high share of renewable energies.

In AI4DG, a decentralized AI platform is being developed to enable secure and autonomous control of distribution grids with a high share of renewable energy.

Trustworthy Electronics - Advanced Security by the Hardware Chain of Trust

Establishment of a verifiable "Chain of Trust" of a trustworthy electronic assembly from component production and commissioning to the end of the operational phase, applicable in critical infrastructures.

Novel design methods for trustworthy electronic circuits

The VE DIVA-IC project aims to strengthen the integrity and identity protection of electronic components in the field of mobility and automation technology.

Neuro-inspired and resource-efficient Hardware-Architectures for plastic SNNs

Design space exploration of hardware architectures (FPGAs and multi-core systems) which allows efficient computation of SNNs and adaption (online-learning).

Climate-neutral Business in OWL (Climate bOWL)

The aim of the project (Climate bOWL) is to develop an intelligent assistance system for the sustainable reduction of greenhouse gas emissions.

Transfer:Square (T2) – Open up new transfer paths and strengthen interactions.

Bielefeld University is designing interaction mechanisms for the digital innovation platform and methods for incentive-based acquisition of new content.

Industry 4.0 ecosystem for the automated deployment of data-driven services (I4.0AutoServ)

The goal of the project I4.0AutoServ is the complete automation of all steps required for data-driven services in an Industry 4.0 ecosystem.

Safe and smart electronic systems for trustworthy electronic products in Industry 4.0 (SiEvEI 4.0)

The SiEvEI 4.0 project applies AI methods to predictively monitor and certify product quality in a distributed, safe electronics manufacturing process.

Development and validation of an AI-based system for autonomous control of smart cellular networks.

In KI-Grid, an AI system is being developed for the autonomous and beneficial control of electrical grids in interaction with renewable energies.

LEGaTO, Low Energy Toolset for Heterogeneous Computing

LEGaTO is focusing on a low energy toolset for heterogeneous computing, consisting of a mature software stack and a cutting-edge heterogeneous hardware substrate.

In the M³S project („Moderne Mensch-Maschine Schnittstelle“ – Modern Human-Machine Interface), state of the art brain-computer Interfaces (BCI) technology will be the basis for the development of more versatile brain-machine interfaces (BMI). In BCIs, brain activity is measured using electroencephalography (EEG) and transformed into machine commands in real time. Therefore, user with physical impairments are able to use a computer.

In the Human Brain Project (HBP) two hardware platforms for the simulation of large-scale spiking neural networks have been developed and made publicly available. The Cognitronics and Sensor Systems group contributes benchmarks for fair comparison of platform performance and efficiency.

The aim of the project was to use interdisciplinary research results to support a proactive design of the working world in the process of digital change.

In this competence center, the working group has led the "Qualification" pillar and conducted practical events for knowledge transfer in the field of AI.

open-mosaic: An open, Modular and Scalable Architecture for Communication-driven Computing

This project targets the development of a highly scalable hardware platform including modular and exchangeable FPGA mircoservers for HPC and Edge computing appliances.

Intelligent data compression as a basic technology for rapid prototyping of automotive applications

Development of an in-vehicle edge server concept and prototypical implementation for demand-oriented data compression in autonomous driving.

NRW Forschungskolleg: Design of flexible working environments - human-centered use of Cyber-Physical Systems in industry 4.0

Projects in an inter- and transdisciplinary environment are to contribute to the design of flexible working environments in which the focus is always on people

Technology Transfer Framework Project (RPTTrans)

The working group contributes to the digital innovation platform and conducts transfer pilots with medium-sized regional companies as case studies.

Machine Learning for Production and its Products

The goal of the ML4Pro² project is to make machine learning (ML) sustainably available for intelligent products and production processes.

Smart sensor network for determining process variables

The aim of the project is to optimize the efficiency of the harvest. A sensor network will monitor the machine processes and interpret them using AI.

High-resolution radar with integrated AI-assisted data processing for cooperative autonomous driving.

The KI-Radar project develops a radar system which is based on an integrated AI-assisted data processing and targets cooperative autonomous driving.