At SPS 2024, SICK presented the robot guidance system PALLOC (PALlet content LOCalization), an AI-supported adaptive localization solution for automated robotic depalletizing. It combines a powerful 3D snapshot camera with a factory-installed and pre-trained neural network and new deep learning-based localization algorithms.
The system can position an almost unlimited number of boxes stacked on a pallet and provides position coordinates for reliable and precise robot guidance. The integrated neural network is already trained for different box types and new variants can be added at any time with the easy-to-use AI tool suite.
Automatic palletizing solutions, manual, conventional or robot-assisted – ensuring the smooth transportation of goods from delivery to shipping or further processing often means you can choose from several approaches. The combination of robot guidance systems with smart camera systems, AI and deep learning plays a key role, accelerating logistics and production processes on the one hand and reducing the number of monotonous tasks for employees on the other.
Thanks to the combination of 3D stereo and color imaging and a browser-based user interface, PALLOC from SICK is just such a robot guidance system and one of the most future-proof models on the market, because the neural network app is integrated directly into the camera, while the intelligent standalone system can be seamlessly integrated into the controls of almost any industrial robot and collaborative robot manufacturer via Ethernet TCP/IP.
PALLOC: Depalletizing with robots and AI
The PALLOC robot guidance system identifies the variations of a virtually unlimited number of stacked boxes and pallets and provides the position coordinates of each box to ensure reliable and precise robot control. The 3D snapshot camera of this system solution can be flexibly mounted on the robot arm or above its working area, allowing optimal recognition of even the smallest details, regardless of the distance to the box surface.
It can produce up to 30 full-color images and 3D image pairs per second in high resolution. This 3D data is automatically compared to the color data so that contours, edges and layer depths are measured and recorded precisely and reproducibly. The pre-installed neural network is trained for a wide range of box types with different sizes, colors, designs and printing. If required, SICK’s suite of easy-to-use AI tools can be used. B. SICK Web Service dStudio for training the neural network. New variants can be added at any time.