Projects

A central focus of the AI Innovation Center is direct cooperation with industrial companies: Feasibility studies (Quick Checks) and projects to develop initial prototypes (Exploring Projects) of AI applications. In addition, consortium projects and innovation networks are aimed at cross-company exchange.

Following the successful completion of the pilot phase, the KIRR Real project was integrated into and expanded within the AI Advancement Center. This expands the cooperation offerings to include regulatory analyses (Legal Quick Checks). The results of the Legal Quick Checks from the pilot phase are available on the KIRR Real Homepage documented. Legal Quick Checks from the AI Advancement Center will be published on this page.

Development and innovation

Logistics and merchandise management

Marketing and communication

Production and quality management

Administration and organization

Legal Analysis

ML-supported prediction of screw straightening quality

An ML model predicts the change in deflection of bolts during the straightening process using measurement data and process parameters, enabling automation and quality assurance. Classification, in particular, is highly successful, predicting with high accuracy whether the bolts will be within tolerance after pressing.

MRK application for equipping machines

As part of the Quick Check, a concept for equipping machines for punching screw heads has been developed as a collaborative application. The application is designed as a mobile robot platform and by means of a bin-picking solution.

NeuraConnect: Outsourcing compute-intensive operations to a 5G edge cloud

Outsourcing of compute-intensive operations to a 5G edge cloud.

Optical quality control of surfaces

The defects in the thermal deformation of aircraft landing gear are detected by the SFS algorithm. The results can be further processed with AI algorithms and material defects can be visualized with suitable lighting.

Packing planning shipping carton

Automatic model-free specification of packing patterns for shipping cartons.
Automation of the packaging process for various objects.

Peg-Men: Flexible assembly under disturbance variables

Flexible assembly of sensitive valve needles is optimized using deep reinforcement learning and physics simulation. The aim is to automate repetitive processes in order to increase efficiency and precision.

Perception for automated handling of dimensionally stable components (AutoLab)

Automated handling of dimensionally unstable components optimizes assembly processes through AI-supported perception, handling and inspection to increase efficiency and quality in production.

Planning Bot

The aim is to identify AI applications to improve production program planning. Employees receive training to develop use cases and support implementation.

Potential for the use of AI in electronic manufacturing services (EMS)

Efficiency in the customer order process is increased by an AI prototype that supports the automatic checking of Bills of Materials in the EMS sector. Inconsistencies and errors are detected and marked using sample data, which significantly reduces manual effort.

Predictive services

Machine learning is used to detect errors in the cutting process at an early stage. Data from log files and service reports enable reliable prediction and minimization of machine failures.