p3-dualspace-Berlin-Automotive-part-fault-recognition

Automatic component failure analysis

The goal of the failure analysis is to identify the main components in a vehicle which caused a final malfunction.

p3 dualspace- Adversariality-detection-heatmapping

Detection of adversarial attacks in autonomous driving

The so-called adversarial attacks represent a real challenge for the safety of the autonomous driving.

p3 dualspace- Nlp-specialized-grammar, Berlin

Using natural language processing to make technical documentation easier to understand

Technical documentation in the aviation sector is safety critical and must be clear and precise even for non-native English speakers.

Computer Vision

We build image classification and object detection solutions based on deep learning technology. Transfer learning allows us to create robust solutions for our customers’ individual applications.

Anomaly detection

Find anomalies in manufactured parts for quality control or recognise irregular states in production processes. We combine the technology with computer vision experience to avoid manual visual inspections.

Explanation

Explaining why a machine learning model came to a decision is an important tool to make the ML black-box transparent. We use heatmapping methods to provide validation of AI technology.

End-to-end solutions

Our work does not stop with a prototype solution. We support our customers starting with data assessment or data acquisition, discussing feasibility and estimating results to deployment to production.

Data is the oil of digitalization. Contact us to discuss how you can benefit from AI technologies.