Company
Solwr
Application
Logistics
Order fulfillment
Piece picking
Features
Mobile robot
Warehouse Management System (WMS)
High-definition 3D machine vision
3D camera
Zivid 2+ L110 - industrial 3D camera
Published
23 September 2024
The quickly evolving world of food and beverage logistics presents unique challenges for warehouse automation. Despite advances in automation, one significant hurdle remains: reliably picking and handling a wide range of items, from heavy boxes to fragile, transparent bottles, all while ensuring minimal disruption to operations. Warehouses need a solution that can seamlessly integrate into existing systems, handle diverse packaging and items, and maintain the high-speed performance required in modern warehouses.
Grab™ combines cutting-edge robotics and Zivid’s high-resolution 3D vision system to create a seamless end-to-end automation process. The Zivid 2+ 3D camera provides precise point clouds, enabling Grab™ to detect and handle challenging items, such as transparent or disorganized products. This data feeds into Grab’s motion planning system, ensuring accurate picking and placement of diverse objects weighing up to 30kg.
The deployment of Grab™ in warehouses has demonstrated its ability to handle a wide variety of items efficiently and accurately. With a picking rate of 100 items per hour and an error rate of less than 1%, the system ensures smooth operations, reducing the need for human intervention. This solution has increased productivity and improved safety by minimizing manual lifting tasks.
Zivid's ability to generate high-resolution point clouds of transparent items is invaluable as we previously required additional sensors to handle such objects.”
Rasmus Nes Tjørstad, Product Owner of Grab™ at Solwr
In today’s fast-paced logistics landscape, Grab™ was developed to address a critical gap in warehouse automation. Inspired by founder Torbjørn Korgen’s experience as a logistics manager, where manual picking proved time-consuming and labor-intensive, Grab™ was created to fully automate the picking process. By leveraging advanced robotics and 3D vision technology, Grab™ is designed to handle diverse and complex items—from heavy boxes to fragile, transparent packaging—while offering scalability and flexibility for various warehouse operations. The goal is to streamline the picking process and reduce the reliance on human labor, making warehouses more efficient and cost-effective.
Growing labor shortages, especially exacerbated by the COVID-19 pandemic, highlighted the need for solutions that could reduce the strain on human workers while maintaining high levels of productivity. By focusing on a combination of precise vision systems and robotic flexibility, Grab™ meets the demands of modern warehouses, allowing for efficient order fulfillment in industries like food and beverage distribution.
One of the key challenges in developing Grab™ was designing a robot capable of picking a wide variety of items, from heavy boxes to delicate, transparent containers. The warehouse environment is dynamic, with disorganized shelves and diverse packaging types, making it difficult for traditional robots to function effectively. The robot needed to accurately identify and grasp items of various shapes, sizes, and textures—especially fragile or wrapped objects—without damaging them. Integration with Warehouse Management Systems (WMS) and seamless operation in confined spaces added to the complexity.
To meet these demands, the vision system had to be extremely precise. It needed to capture objects of different materials, including difficult-to-detect items like translucent bottles and glossy surfaces, without requiring multiple capture settings. These items had to be identified not just for their pickability, but also for accurate placement onto the robot’s load carrier. Ensuring this level of precision was crucial, as improper handling could lead to damage or errors in the picking process.
Another challenge was designing a universal gripper, capable of handling a broad range of objects. It needed to be gentle enough to pick up soft, plastic-wrapped items while still robust enough to manage heavy, rigid objects up to 30kg. The gripper had to work seamlessly in cramped spaces, reaching into shelves where items might be stacked unevenly. Through continuous development and testing, Grab™ has successfully integrated advanced grippers and Zivid’s 3D vision system, enabling it to operate effectively in these challenging conditions, even without pre-programmed CAD models of the items.
"The standout feature of the Zivid 3D is its excellent resolution and accuracy in producing point clouds. This significantly enhances the precision and robustness of our localization operations, allowing us to detect more items per image capture with high confidence.”
Rasmus Nes Tjørstad
Product Owner of Grab™ at Solwr
To meet the challenges of automating the picking process in dynamic warehouse environments, Solwr chose to integrate Zivid’s advanced 3D vision system. The Zivid 2+ L110 3D camera provides true-to-life point clouds with exceptional resolution and accuracy, allowing Grab™ to detect and localize a wide range of items with high precision. One of the standout features of the Zivid camera is its ability to capture transparent and reflective objects, a crucial capability given the variety of packaging materials present in food and beverage warehouses.
The Zivid 2+ L110 3D camera captures a wide range of items on Grab™
The unique 2D + 3D data offered by Zivid streamlines the vision processing, giving Grab™ the ability to not only detect the items but also accurately determine their pickability and proper placement on the load carrier. This precision allows Grab™ to handle objects even in disorganized and confined spaces, with minimal risk of errors.
For handling diverse items, Solwr paired Zivid’s vision system with a custom-designed gripper capable of picking up objects weighing up to 30kg, ranging from sturdy boxes to soft, fragile packages. This gripper, combined with Zivid’s high-resolution point clouds, enables Grab™ to execute accurate picks even in cluttered storage conditions without the need for pre-programmed CAD models.
Seamless integration with WMS ensures that Grab™ operates efficiently, receiving real-time pick orders and updating the system with progress, all while using Zivid’s vision capabilities to provide the required accuracy and performance. With this solution, Solwr was able to create a highly reliable, scalable mobile picking robot that excels in diverse warehouse environments.
The Grab™ mobile picking system is now one of the most advanced solutions in warehouse automation, offering unmatched flexibility and precision in handling diverse items. Thanks to the integration of Zivid’s 3D vision system, Grab™ has achieved exceptional situational awareness, allowing it to accurately identify, pick, and place items ranging from rigid boxes to fragile, transparent containers. The high-resolution point clouds generated by the Zivid camera ensure precise detection and localization, even in disorganized environments.
This cutting-edge vision technology, combined with Grab’s versatile gripper and seamless WMS integration, has revolutionized the pick-and-place process in warehouses. The ability to efficiently handle items up to 30kg in confined spaces while minimizing errors has led to significant improvements in productivity and cost efficiency.
As a result, Grab™ not only improves operational performance, but also enhances worker health and safety by reducing manual lifting. This advanced system, with its innovative capabilities, positions Grab™ as a leading solution in the future of warehouse automation, capable of scaling to meet the demands of global logistics operations.
Solwr is a Norwegian company specializing in logistics technology, combining software and automation to solve complex challenges in the trade industry. Formed from the merger of Driw and Currence Robotics, Solwr offers innovative solutions that integrate ERP systems with robotics.
Zivid brings
Gjerdrums
N-0484, Oslo
Norway