First paint, then mill Machining tool paths can now be drawn using AI

Related Vendors

Fraunhofer-Institut für Produktionstechnologie IPT

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

Wuxi InfiMotion Technology Co., Ltd.

SIMTOS (Seoul International Machine Tool Show)

Planning the movement paths of milling cutters & Co. in CAM systems is not trivial. With an AI-supported option, this can now be done using pen gestures.

The paths of cutting tools used in CNC machine tools are planned today using CAM software. However, many CAM parameters must be taken into account, as researchers from the Fraunhofer Institute for Production Technology (IPT) say. These parameters dictate to the CAM system how the exact tool path is generated before the tool path is handed over to the machine control. But the more complex a component is, the more extensive but also error-prone is the description of the tool path in this way.

Less effort in CAM programming

Therefore, good CAM programming requires a high level of expertise. Because only very trained and experienced experts can quickly and correctly estimate the result without countless laborious iterations. Frequently, many different parameters would have to be tried out in several passes until the tool travels the optimal path. The IPT team and its industry partners want to reduce this effort in the future, which is happening within the framework of the "CAMStylus" project. The aim was to make the planning of tool paths in CAM systems more intuitive through Virtual Reality (VR). The VR application, in combination with specially developed hardware, now allows the desired tool paths to be described three-dimensionally using gesture control. The system then calculates the CAM parameters from the registered gestures, which are used in the CAM system to calculate the tool path, the researchers explain.

Define the tool path by pen and hand

In the first phase of the project, the Fraunhofer IPT researchers developed the VR environment and programmed suitable software to track gestures that describe the path movements. The combination of hardware and software makes it possible to capture the movements of so-called tracking pens, which were also developed in the project. In order to derive the right information for the planned tool path from the motion tracking data, the team relies on an AI application (artificial intelligence) based on neural networks. These were trained for this task using specially designed geometric bodies. A particular challenge in software development was to continually align the CAM data with the captured data from the tracking system. Because only then can a reliable data basis be accessed for later calculations, as the researchers emphasize.