Computer scientists unlock new methods for the laser material processing industry

A novel approach from Saarbrücken, Germany, promises to overhaul a whole range of laser material processing techniques. Researchers at the Max Planck Institute for Informatics have developed an artificial intelligence-based method for this purpose that significantly speeds up some of the intermediate steps in production. There is an industry collaboration with TRUMPF. The high-tech company offers manufacturing solutions in the fields of machine tools and laser technology. On top of that, the team is now being funded by the EXIST startup program of the German Federal Ministry of Economics and Climate Action (BMWK) to bring their development to market maturity.

The new approach is based on a so-called evolutionary exploration algorithm combined with a customized sorting method. "With this, we have developed the first computational, automated solution for a whole range of highly complex problems of the laser material processing industry, that were until now still solved 'manually' by trial and error" explains Vahid Babaei, head of the research group Artificial Intelligence aided Design and Manufacturing at the Saarbrücken Max Planck Institute for Informatics at the Saarland Informatics Campus.

An example for such a production method is the so-called "Color Laser Marking," a process in which metallic surfaces are colored by means of heating them with a laser. Heat-induced colors are extremely difficult to predict, as they depend on numerous process parameters—for example, the type of metal being heated, the speed at which it is heated, the heating power, or how long a single laser pulse lasts.

"If one wanted to heat metal to a certain color, the only solution until now was to manually edit the corresponding parameters until the desired color tone was achieved. This very inaccurate and inefficient approach prevented industrial adoption of color laser marking and also similar production methods, that depend on optimizing the interplay of a large set of laser parameters," adds Vahid Babaei.

Together with his colleagues Sebastian Cucerca (research engineer at the Max Planck Institute for Informatics) Piotr Didyk (Professor at University of Lugano) and Hans-Peter Seidel (Scientific Director at the Max Planck Institute for Informatics) Vahid Babaei has developed an algorithm which automates and streamlines this process of parameter exploration to such an extent, that it makes a whole set of laser material processing methods feasible for industrial use. With regard to Color Laser Marking, the new approach makes it possible for example to "print" saturated, high-resolution, colored images on metal without using any dyes, only through heating the material.

The method uses a so-called evolutionary exploration algorithm in combination with a customized sorting method. The exploration algorithm repeats the same steps until it finds the best possible result. First, it starts with the laser making markings based on randomly selected parameters. The algorithm then measures the properties (here color and resolution) of these markings and then, based on these, calculates the next set of parameters for the laser to use to mark.

This is where the customized sorting algorithm comes in. It sorts the best-performing process parameters of the previously marked colors according to various metrics, such as resolution or color saturation. The evolutionary algorithm, in turn, uses these sorting results to generate a new generation of colors that contains the best characteristics of the parent generation. This iterative process runs until there is no significant improvement in the result and thus the best possible outcome has been found.

This new approach to these types of optimization problems is currently unique in the world and is highly scalable to different types of lasers and substrates and also other properties than color. "We strongly believe that color marking is just the tip of the iceberg and our algorithm can accelerate many different processes dealing with surface activation through lasers, like for example changing the haptics of a material," says Vahid Babaei.

To bring this development to market maturity, a team led by Vahid Babaei and composed of Azadeh Asadi, Sebastian Cucerca, and Mojtaba Bemana (all from Max Plank Institute for Informatics) is now being funded by the EXIST start-up program of the German Federal Ministry of Economic Affairs and Climate Action (BMWK).

In addition, the researchers have already been able to win one of the world's leading suppliers of machine tools for flexible sheet metal processing and for industrial lasers, TRUMPF, based in Ditzingen near Stuttgart, as an industrial partner. Paul Stumpf, business development manager at TRUMPF, says, "We believe this technology can have a major impact on laser marking industry and beyond. The technology treats physical laser-material interaction side of the process as a black box and takes advantage of the power of data and artificial intelligence algorithms."

Provided by Max-Planck-Institut für Informatik