Industrial production is increasingly characterised by the phenomenon of mass customisation: goods are produced in large quantities and yet individualised. Additionally, life cycle is becoming shorter, especially in the consumer goods sector. Consequently, several product generations have to be manufactured on the same production line. This makes the efficient use of industrial automation technologies difficult: these are characterised by manufacturer-specific (sub)components that have proprietary interfaces and can only be programmed and integrated into a functional solution by experts. As a result, developing and commissioning of new automation cells is both time consuming and extremely costly. Even worse, changes to existing solutions are rarely made and are associated with high economic risks. However, this does not fulfil the market requirements mentioned above. Methods of modern software development such as DevOps (including continuous integration – CI and continuous deployment – CD), which enable fast and efficient iteration cycles, are hardly established in automation.

 voraus robotik solves these challenges with a combined approach: the containerised software platform voraus.core runs in real time and orchestrates all subcomponents of an automation solution (operational technology – OT) – including fieldbuses such as EtherCAT. In particular, it can be programmed in high-level language (Python) and encapsulates the underlying components using drivers. This enables high-level language programmers to develop applications for industrial automation, which was due to their missing OT-knowledge hardly possible before. The voraus.core is shipped with an integrated robot library that allows for easy kinematic and dynamic modelling of almost any individual robot (both serial and parallel). Direct and inverse kinematics can be calculated in real time and also model-based feedforward control at torque level is possible. Consequently, task-specific robots can finally be used economically on an industrial scale. Of course, robots from established manufacturers may also be easily integrated. Thanks to a standardised interface to the application layer, an exchange of robots and automation components in general is possible without significant effort.

 At the same time, programming in high-level languages opens up the world of modern software development, including AI-supported co-pilots and simulation. The prerequisite for seamless integration is that programming, testing, simulation, and execution (deployment) for all components of an automation solution are done on the same code base and no intermediate code or similar is generated. Only in that way a direct and seamless transfer to the shop floor (and back again) can be ensured. This unique feature is covered by the development suite voraus.pioneer: numerous widely used and harmonised IT tools for quality assurance (e. g. versioning, static code analysis, and error injection) are available and their capabilities are far superior to the proprietary ones in the OT world (if available at all). The components installed in the automation cell are virtualised and, therefore, can be addressed via standardised interfaces. Thus, almost the entire application can be developed and tested without hardware, saving costs and time. Safeguarding the application in the simulation (in combination with a physics engine) ensures correct operation and also allows for cycle time analyses as well as optimisation. Control code that has been tested and optimised in this way can be deployed with minimized risk and minimal time expenditure, opening the door to cost-effective CI/CD in automation technology.

 We are certain that programming automation solutions in high-level languages while ensuring all production requirements (especially robustness and reliability of OT) is the key to efficient and, therefore, economical reconfiguration of robots and cells. Ultimately, the application takes centre stage and not the solving of manufacturer-specific restrictions.