data science & machine learning.

The vast domain of AI is rapidly entering the high-tech industry. We focus on data science & machine learning (ML), referring to a system/machine that, without explicit programming, can learn from its environment and therefore may adapt its behaviour. In mechatronic system development, building on our domain knowledge and user understanding, we can apply ML to increase functionality, improve performance, automate activity and reduce cost. We do not engage in exploratory data analysis or algorithm development per se, but rather specialize in state-of-the-art ML implementation in hardware and software systems.


  • machine vision and time series & control
  • deep learning, reinforcement learning, neural networks


  • data acquisition, transformation, augmentation, statistics, visualization
  • classification, segmentation, prediction, object detection, anomaly detection

filtering data, training algorithms

In ML implementation, we have two focus areas. In the field of machine vision, we have a 20-year ML track record. The upcoming area of time series & control allows us to apply ML, for example, to systems that are equipped with multiple sensors, for combining and filtering data. Among our specialties are preparing data for efficient training of ML algorithms and using synthetic data for the training. This is cheaper than using a dedicated physical test set-up, and faster than real-life training, where it takes a long time before the rare events have been trained.


Irene van Eerden.

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum molestie varius arcu, semper tincidunt nulla scelerisque id. Pellentesque arcu odio, ultricies vel mattis eget, tempus at enim. In vulputate massa molestie, hendrerit ex ut, efficitur libero. Suspendisse nec malesuada mauris. Phasellus dolor metus, condimentum ut arcu sit amet, porttitor accumsan.

addressing expectations and reservations

We are aware of the expectations and the reservations that surround ML, and AI in general. We will stay away from the hype, in each case reviewing the need for ML critically; when a classical data analysis solution is sufficient, we will adopt it. On the other hand, we acknowledge the non-explainable character of ML. In the medical domain, for example, we take into account the concerns about the autonomy of professionals and the safety of (patient-related) procedures. Wherever ML is accepted by the regulatory body and the customer, we can exploit the major benefits.




all expertises



industrial design




optics & vision

quality & regulatory compliance

software architecture & design

system architecture & design

Please fill in the form below to download this file. Please note the download will be sent by email.