Radalytica Robotic Imaging System

Compared to common X-ray imaging technologies, such as films or flat panels, we use Advacam’s new-generation X-ray detectors, which are also used by US NASA in space at the international ISS station for their unique properties. These imaging detectors are characterized by high resolution, an almost unlimited range of grey levels, and high sensitivity.
The advantage of using these advanced detectors is the fact that we are able to use up to half the energy of X-rays compared to existing technologies, thanks to their sensitivity. This implies higher image resolution, but also lower demands on shielding against leakage of X-rays. This simplifies, reduces the costs, and lightens the construction of shielding chambers.
An x-ray image of a dry rose demonstrates the unprecedented quality of x-ray imaging. Unlike a standard x-ray device, which is unable to display a dry rose in a vase, an x-ray image of a robotic system using Advacam detectors in high resolution and contrast shows both all the delicate parts of the head and even the stem inside the bottle, at the same time!
Something that HAS NOT BEEN POSSIBLE so far!

Radalytica a.s., in cooperation with SONOTEC GmbH, integrates air-coupled ultrasonic probes. They are particularly effective for detecting delaminations in composite materials that are virtually invisible by X-rays. Air-coupled ultrasound will also find further use in material, aviation, and other research fields.
Different imaging methods have different uses depending on the required application. In many cases, it is appropriate to use more than one method at a time to get a better overview of the sample.
A good example is a damaged composite wing of an aircraft after hitting a foreign object. The X-ray image reveals the structure of the composite and its possible damage by fine cracks at the point of impact. However, delamination in the surrounding area cannot be detected by X-ray. Ultrasound is nevertheless suitable for the detection of delamination, but it does not allow high-resolution imaging of fine cracks. The solution is a combination of these methods.
The robotic system has a record of the coordinate system and thus it is possible to accurately combine the image output of individual methods and get a better overview of the situation and the overall damage.

The Robotic Imaging System is equipped with laser distance measurement as standard. It is used by the system to map the shape of the sample before measuring. This information is used to navigate the robots around the sample to avoid collisions.
It can also be used to automatically control the trajectory of movement of the robots so that, for example, they maintain a constant distance from the surface and the angle to it.
This is illustrated on a curved honeycomb sample, where all honeycomb cells are displayed perpendicularly. This makes it easy to check, for example, the sufficiency of glue, bubbles, etc.

Robots also allow measuring 3D images using computed tomography or laminography. These are commonly used methods in X-ray inspection, but with limited applicability to large objects.
Conventional computed tomography is limited by the size and shape of the sample (part). The sample must be placed in a chamber where the rotary table rotates it in one axis and the system records images from different angles. Then a 3D model is created from the acquired images.
This operation is relatively lengthy, even through considerable limits on the size and shape of the sample. Robots exceed this limit and also allow to measure 3D images in the selected area of a large object.

We use the most advanced technology of artificial neural networks for automatic image evaluation. In technology, neural networks are often referred to as artificial neural networks (NN) or neural networks to distinguish them from the biological neural networks over which they are modeled.
The main idea of NN is that the human brain is the most complex and intelligent “computer” that exists. By modeling the information processing used by the brain, it is possible to approach human intelligence in image processing.
We use this advanced technology to set up a system for the automatic evaluation of OK and NOK parts. This system can learn to evaluate parts with almost 100% reliability and eliminates the factor of human error caused by negligence, fatigue, insufficient qualifications, etc.

Real-time imaging with a 3D mouse allows full control over the position and viewing angle of the X-ray image. The X-ray image of the given area of the sample is displayed in real-time on the screen. Simple manual control using live view creates the perfect tool for locating defects in the inspected structure in 3D. Immediate feedback between manual control of the direction of view and the X-ray image is what makes the system a very intuitive tool.
Thanks to the connection of the image with our actions, the brain is able to create a very clear idea of the 3D structure. Therefore, inspection with robots is faster, less demanding on data processing compared to CT, and can be applied to the selected areas of larger objects.
The great benefit is also the possibility of remote robot control independently of the user’s location.