Space technology transfer

Scanway’s two branches of business

Scanway is one of the few companies in the world operating in two areas: space and industry (for more on our space business, visit: scanway.space). Despite the clear business division between space and industry sectors inside the organization there is a free flow of technology and know-how. The same engineers who build telescopes for nano and micro satellites work on industrial vision systems. Solutions developed for the space industry are used in the wider industry. On the surface, such adaptation may seem very complicated and time-consuming. However, in Scanway’s case, it is extremely simple, as we have already developed certain technologies.

Space technologies brought down to earth

An example of technology used in orbit as well as in industry:

1. Hyperspectral cameras
2. Algorithms using machine learning

Optical solutions providing the possibility of imaging with very high resolution and so-called multispectral imaging, i.e. realized in a much wider range than just the three colors R, G, B, were designed for space and then implemented for industrial projects.

In our portfolio we have cameras that can see up to 200 spectral channels. These are the different bands of electromagnetic radiation that reach us beyond visible light itself. Such capabilities provide a chance to see even minimal deviations from the quality standard in the objects being inspected. This technology is great for the food industry, for example, for detecting contaminants and inclusions in food.

Visualization of the vicinity of the Vistula Lagoon based on data from Landsat 7 satellite (link)

Machine vision is based on arming production systems with instruments that allow taking pictures of products and then automatically analyzing them. Our systems use vision algorithms that enable image processing. Based on the information obtained, we draw specific conclusions, such as diameter, height, surface roughness or color. For this type of analysis, we very often use artificial intelligence algorithms and “deep learning” technologies, i.e. a deeper, more advanced variety of AI, thanks to which, for example, we recognize certain unusual errors.

In industry, there are applications for which it would be impossible to write uniform algorithms, ready to be applied every time – the errors are simply too different, or teaching each variant is not possible. In such a case, we generate a neural network that, after indicating the desired values, will learn on its own to recognize whether the product meets the accepted standards or not. At Scanway, we are constantly using and developing artificial intelligence technologies in both industrial and space projects. In the latter area, we use advanced algorithms, for example, in the SHS product – Spacecraft Health Scanner used for example: to analyze damage to solar panels mounted on a satellite caused by a collision with a space micro-object. Because of the variety of errors that can occur, the system has to make its own analysis based on what it sees.