Robotics Two Worlds Perfectly Connected With Each Other

Related Vendors

SICK AG

FAULHABER Drive Systems

Taiwan Machine Tool & Accessory Builders' Association (TMBA)

Wuxi InfiMotion Technology Co., Ltd.

A revolution is currently taking place in industrial automation: the connection of the real and virtual world. The AI-supported robot guidance system Palloc from Sick is an exemplary representation of this.

Palloc (PALlet content LOCalisation), as the name suggests, is Sick's intelligent solution for one of the major challenges of today's economic cycles: millions of fully loaded pallets that need to be unloaded quickly and error-free at the destination, preferably around the clock. What is increasingly lacking is suitable personnel, keyword: skills shortage. And on the robot side, often the integration of AI. Therefore, a forward-looking automation solution with high adaptability thanks to AI is sought.

Advantage of Virtual Worlds

Palloc closes this gap by being able to integrate into the real shop floor of companies via plug and play. Additionally, the industrial metaverse is currently emerging at the boundary between virtuality and reality, opening up entirely new paths in automation. One of the first areas undoubtedly benefiting particularly from this is robotics, with its very high demand for adaptable AI solutions. With the example of Palloc, Sick can already demonstrate today the possibilities that arise from leveraging virtual worlds. Initially, to virtually represent and extend this and other robotics solutions in customer applications, but also prospectively to virtually commission corresponding applications in advance. Industrial robotics expert Peter Pokrandt and Jan Jarvis, as head of virtualization, are pioneering work at Sick with their respective teams and in close collaboration with the involved product developers in both worlds.

In reality, Palloc is already an ideal combination of Sick's Visionary-S 3D snapshot camera with a pre-installed and pre-trained neural network. The result is a uniquely compact set for robot guidance that operates as a so-called end-of-arm solution in the range of seconds. Mounted on the robot arm or alternatively stationary above the pallet, the system quickly recognizes a virtually unlimited variety of different carton variants on the top loading level of the pallet. It identifies the contour and forwards the position coordinates with height information for gripping and transferring to the robot control. "Unlimited number" because the neural network, which is already factory-trained to a wide range of carton spectrum, can have any number of additional carton variants added via a user-friendly AI tool suite. Thus, Palloc, which thanks to an integrated computing unit also does not require a separate industrial PC, is one of the most future-proof localization and robot guidance systems of its kind.

Simulate Concrete Applications

To maintain and expand the lead, the metaverse comes into play. Its virtual 3D environment is precisely designed for the development, testing, and training of AI. With the now possible simulations in the virtual environment, specific applications can already be validated before being deployed in the real world. To continue to offer its customers innovative solutions in the future, Sick collaborates with Nvidia, one of the world's leading providers of AI computing, and utilizes their simulation technologies such as the Isaac Sim platform based on Nvidia Omniverse. In the case of Palloc with its integrated deep learning, this means that the functionality of a solution can already be represented and, if necessary, validated as a simulation in application-related environments. In the future, building on this, images and examples can be trained, and assessment procedures developed even before practical implementation. "The metaverse is thus the fastest and most sustainable way for us to integrate sensors like Palloc into new customer applications," explains Pokrandt.

The possibilities go even further: customers receive Sick's solutions presented visually appealing in the later factory environment, for example at trade shows. And with specific interest, it would also be possible to verify with these means whether a specific robot model performs its tasks correctly in a concrete, simulated warehouse with Palloc, even before it takes place in the real world. Any necessary parameter adjustments can be made easily and directly, so that later the depalletizing goes as desired. For Jarvis, working with the metaverse is mainly about avoiding certain paths and uncertainties in reality, because you can simply try it out virtually: "Who wants to take detours in the real world when you can lay the right tracks straightaway, especially when the detours cost a lot of time and money."

And the journey is not yet over. Whoever can depalletize can also palletize. For this, Palloc is currently being prepared in the lab through virtualization. The so-called mix palletizing is of a completely different caliber with significantly higher accuracy and speed requirements. But Pokrandt sees Sick on a good path: “What looks like Tetris with real boxes on the monitor—also thanks to Palloc—enormous added value for our customers.”

*Peter Pokrandt, ASF Industrial Robotics / Integrated Automation, Sick AG; Jan Jarvis, Head of Virtualization/Think Tank, Sick AG

Machine Authentication

Machine Operator Authentication - Why knowing "who" matters