How can AI benefit industrial laser systems users?
Artificial intelligence has increasingly been making its way into industrial manufacturing over the past few years. With the drive for the factory of the future, or Industry 4.0, it’s difficult to argue that manufacturing is getting increasingly smarter.
Looking specifically at laser processing, AI will enable lasers to ‘become even more efficient, easier to operate and more adaptable’, according to Christian Schmitz, CEO of laser technology at Trumpf.
Indeed, the company has been developing its laser technology for some time to include the adoption of AI solutions.
At the Laser World of Photonics trade show in 2019 it presented a laser system featuring AI that enables it to be operated using voice commands.
Better user communication on laser systems
The system is equipped with a marking laser, and was used to demonstrate some of the potential benefits of AI in laser material processing. For instance, how the operator could instruct a machine to carry out actions such as ‘open/close the door’, ‘start the marking process,’ or ask ‘how many products have been marked today’, by speaking into a microphone. The laser system responds immediately and carries out the instruction.
Schmitz believes that there are numerous advantages with using voice control. Navigating skills gaps, for example, as inexperienced users will not require extensive training to safely operate a machine. They would not need to learn how to navigate what can be a multi-layered menu structure when entering instructions manually via the software interface. It also saves time, allowing operators to keep their hands free to prepare the next part for processing or remove a finished part from the machine.
Moreover, by providing barrier-free access, voice control allows a more diverse range of employees to operate the machine, so a handicapped person, for example, would not be prevented from doing so. Schmitz revealed at the time that the company fully intends to further simplify operations of the laser marking system. For example, advanced sensor systems and image recognition software could help the machine identify a part, select the necessary program, and automatically position the laser at the correct angle before starting the marking process.
A helping hand
It’s not just marking applications that Trumpf is applying AI to. The company’s TruLaser Center 7030 laser cutting machine can cut parts from a metal sheet and remove them automatically using 180 movable pins supported by AI. This repetitive job has, historically, been undertaken by hand because the parts may tilt slightly as they come out of the sheet, and a human can easily manoeuver them into the correct position . In the AI-assisted system however, from below, the pins lift the part from the scrap skeleton, while suction plates hold it in place from above. If a part gets stuck on the first try, the suction plates and pins repeat the process in a slightly different way until they succeed.
The machine sends data about any failed attempts to the cloud, where it is evaluated centrally. TruLaser Center 7030 systems will then receive regular updates, allowing each and every user to benefit from what the algorithms have learned worldwide. This huge pool of machines, all connected to the same central hub, could offer real benefits to users by making their individual machines better over time.
The investment by Trumpf for using AI in its products is not slowing down. It made a further announcement last year that it is extending its partnership with the Fraunhofer Institute for Manufacturing Engineering and Automation IPA to form a research alliance, set to run until 2025. Its goal is to develop AI solutions for connected sheet-metal manufacturing on an industrial scale.
‘Trumpf’s mission is to further extend its AI leadership in sheet-metal fabrication,’ explained Thomas Schneider, managing director of development at Trumpf Machine Tools. ‘To that end, we have already started investing in the kind of future technologies that will drive major efficiency gains in our company and boost our competitiveness.’
AI now enables laser systems to be operated using voice commands. (Image: Trumpf)
The collaboration sees 10 employees spanning both partners involved in the project, which will receive around €2m of funding spread over five years. Professor Thomas Bauernhansl, director of Fraunhofer IPA said: ‘Trumpf has been working with us on connected manufacturing for years because they share our view that Industry 4.0 developments represent a major opportunity. Everything depends on what happens over the next few years – so these are exciting times! We expect the coronavirus pandemic to act as a kind of catalyst: those who are well prepared will be perfectly placed to exploit the huge opportunities that lie ahead. Soon we’ll see whether we have laid the right foundations for the future in our joint projects.'
Trumpf has also recognised the benefits of collaboration to achieve its goals from the point of view of selecting AI provision partners. The company is now the newest strategic partner and investor to Nnaisense, a provider of artificial neural networks, deep learning, reinforcement learning, artificial evolution and general purpose AI. Jürgen Schmidhuber, chief scientist and co-founder at Nnaisense said: ‘For many years we have been working on many client engineering and research projects and have developed new approaches combining control theory and AI, as well as highly scalable, evolutionary reinforcement learning methods.
'We very much look forward to collaborating with this innovation leader in sheet metal fabrication machinery and industrial lasers. Nnaisense will support Trumpf in its latest laser technologies, applying our AI expertise to create industry-leading solutions.'
Predicting heat conduction in additive manufacturing
Nnaisense works with customers to produce largely bespoke AI solutions. ‘Off-the-shelf AI solutions are often too generic,’ said Schmidhuber, ‘resulting in lost time and investment.’ The backbone of the firm’s solutions is its NNAI engine, which can be customised for three application areas: inspection, modeling and control.
The benefits of these three areas for the user are, said Schmidhuber, that automated inspection can help to ensure efficient quality control. Modelling can help to predict the dynamics of a process, sampling data from the actual process to learn a predictive model bespoke to that application. When it comes to control, intelligent automation allows the sensory-motor loop to be closed in a way that goes beyond traditional control engineering, applying deep reinforcement learning to adapt neural network controllers through safe and efficient interaction with a learned process model. This makes the process much safer for users.
Nnaisense is well versed in bringing AI into laser processing, having collaborated with Electro Optical Systems (EOS) to incorporate intelligent monitoring into its additive manufacturing processes. EOS concentrates on selective laser melting and selective laser sintering. During this process, the distribution of laser energy within the layer is a key factor determining the material properties of the part.
Harald Krauss, from the innovation team at EOS, revealed: ‘One of the most important things you have to consider when thinking about a part’s exposure strategy is heat conduction in the part. The selective laser melting process introduces a lot of heat from the laser beam, and this has to be dissipated through the part and the system. It will depend on the geometry conditions and part shapes in the build process, so you have to adapt your processing strategy there. It is important, to avoid defects, to use monitoring systems to optimise output.’
Nnaisense assisted by developing a deep network model that can accurately predict a heat map based on job parameters. It can detect process anomalies when sensor readings deviate from predicted behaviour, and control laser intensity to avoid defects and optimise material properties.
Florian Trifterer, senior researcher at Nnaisense, said that ‘what is supposed to be produced is defined by build instructions. These are translated into detailed laser path instructions which are executed by the machine, and the observed thermal image is captured from the observed thermal radiation captured from the just-finished layer. It is here that we can see if any heat is uniformly distributed, which could be caused by a number of things, for example, the way the laser moved. The difficulty is being able to differentiate genuine spurious defects from systematic effects.’
This is where AI comes into play. Trifterer said: ‘What if we could predict the systematic effects, based on the building structures?’ This is exactly what Nnaisense did for EOS. ‘We built a custom architecture, designed to allow for occurring physical effects to be modelled,’ said Trifterer. The architecture makes use of convolution and recurrent processing and, as Trifterer explained, it was ‘designed to be able to capture the observed physical effects in principle’.
Following weeks of testing and evaluation, it became apparent that the trial had been a success. Revealed Trifterer: ‘EOS experts agree that the model is able to faithfully predict complex, non-local properties of the optical tomography image. In essence, this means that the inner workings of the network are able to mimic relevant aspects of the physical process. Therefore, it is appropriate to call it a digital twin of the additive manufacturing process.’
Real-time monitoring of the laser cutting process
Returning to AI for laser cutting, MC Machinery, a Mitsubishi subsidiary, has recently released a series of AI-enabled fibre lasers. AI technology is used to monitor the cutting process in real time, with audio and light sensors automatically adjusting parameters to help optimise the machine’s performance. If an incorrect cut is detected, the machine is designed to make the required adjustments to improve or regain the cut. It is also able to optimise the cutting speed in all plates, regardless of plate quality.
The new series, known as the GX-F Advanced series, was designed by Mitsubishi engineers with major Mitsubishi components. The firm says it is one of the only laser systems in the industry with a single source for service and support.
The GX-F Advanced series is MC Machinery's new series of AI-enabled fibre lasers. (Image: MC Machinery)
To reduce setup time, the zoom head delivers flexibility by automatically changing the beam size, shape and focal point for each material, with the ability to process plates with a wide range of thicknesses. Piercing time is reduced by as much as 60 per cent, making it possible to pierce 25mm-thick mild steel in 0.8 seconds.
The AI nozzle monitor uses a camera system to monitor nozzle life, and the nozzle changer automatically replaces defective nozzles to support continuous processing.
An additional and particularly timely benefit to increased efficiency for the end-user, is the reduced need for operator input. Shane Herendeen, North American sales manager for fabrication at the company, said: ‘Power lies in what a fibre laser can do, not the kilowatt it has. With the manufacturing industry suffering from a shortage of experienced workers, the new fibre lasers were designed to help minimise the need for operator input while maximising quality and productivity. The AI technology means that the machine is easy to use for operators of all skill levels.’
By integrating advanced gas reduction technology, the new fibre lasers are able to offer more power but use 77 per cent less nitrogen. What’s more, it incorporates augmented reality technology, which allows the display of overhead 3D images of the system without distortion.
AI is able to monitor cutting processes in real time and simplifies use for workers of all skillsets. (Image: MC Machinery)
This, in turn, allows the user to easily place and nest parts to reduce setup time. Herendeen said: ‘Not only does the GX-F Advanced Series require much less operator input because of its AI technology, it uses less nitrogen to lower operating costs and maximise profitability. This is truly a game-changer for metalworking companies of all sizes.’
The GX-F Advanced series is designed to be easily integrated with a wide range of automation systems, including material storage, loading, removal and part sorting. Additional features include user-friendly, smartphone-like controls; real-time tracking of electric and assist gas consumption; real time on-site and remote monitoring of the cutting process and remote diagnostics and predictive maintenance.
AI-powered lasers to remove weeds from fields
AI-powered laser technology is also being used in industrial agriculture, with a recent project combining partners from eight EU countries to develop technology to damage weed growth, with the goal of replacing chemical and mechanical weed removal methods.
The WeLaser project, funded by Horizon 2020, features research institutions, companies, and nongovernmental agricultural organisations. They are developing a movable, high-power, thulium-doped fibre laser and scanner, and testing their effectiveness on selected crops over the next three years.
The idea is to damage the growth centre of weeds in a sustainable way, as an alternative to heavy chemicals or manual/mechanical weeding.
In order to selectively target the weeds, scientists at the Laser Zentrum Hannover (LZH) are developing an image processing system that uses AI to distinguish them from crops while recognising the position of their growth centre. Target co-ordinates will then be used to control a robust, multi-row scanner system that directs a laser beam at the growth centre of the weeds.
Carbon Robotics’ autonomous weed elimination robot combines AI and laser technology to identify, target and eliminate weeds. (Image: Carbon Robotics)
For the field, the systems will be installed on an autonomous vehicle. They will be co-ordinated via a smart controller that uses the Internet of Things and cloud computing techniques to manage and deploy agricultural data.
The LZH is also developing concepts to ensure laser safety for everyone involved, such as farmers and machine operators.
The partners want to test the prototype on sugar beet, corn, and winter cereal crops. It is forecasted to be available at the end of the project in 2023, and then be further developed for commercialisation.
Seattle-based Carbon Robotics has recently unveiled an autonomous weed elimination robot similar to that under development by WeLaser. One of its robots is capable of weeding 15 to 20 acres per day using AI-assisted laser technology, and could replace the need to deploy several hand-weeding crews. The firm has already sold out of the new systems for 2021.