By March 7, 2018 0 Comments Read More →

Universal Robots for bin-picking and packaging

180307_UniversalOnline sales show no sign of slowing and the growth is driving demand for fulfilment operations to pack and ship all those ordered goods. The work is repetitive, however, and requires high accuracy to reduce costly errors or returns. That means that jobs in picking and packaging items for shipping aren’t particularly desirable, especially in tight labour markets.

These tasks are ideal for automation, but there are challenges. While repetitive, packaging applications are also highly variable, requiring the ability to recognize and pick a wide range of items, in different sizes, weights, and materials. And human workers are typically still needed for upstream and downstream processes that are often performed right next to the packaging operation.

With that combination, collaborative robots fit the bill perfectly. Cobots offer low upfront costs, flexible implementation, and easy programming for developers who are rapidly creating innovative new systems to optimize bin-picking and packaging. And with no need for safety guarding (after risk assessment), they are ideal for collaborative operations next to human workers.

RightHand Robotics, based in Somerville, Massachusetts, has created a complete solution that includes a Universal Robots (UR) robot arm, computer-vision and machine-learning systems, as well as an innovative gripper that can pick up a wide range of objects in an unstructured environment such as a bin or tote. Co-founder Leif Jentoft refers to this flexible grasping as “the holy grail” in robotics research: piece-picking that delivers “The Three ‘R’s” – range of items, rate of picking, and reliability of the customer getting what they ordered.”

It’s not reasonable for a warehouse to have 3D models of all the items in the warehouse. The real challenge is to enable robots to pick things that it’s never seen before, just as a person can.

RightHand Robotics team member Nadia Cheng explains that robots have traditionally been used in manufacturing to perform the same task millions of times. In the supply chain industry, however, with millions of different items in the warehouse, the robot might be doing each task just once, but for a million different things. “It’s not reasonable for a warehouse to have 3D models of all the items in the warehouse. The real challenge is to enable robots to pick things that it’s never seen before, just as a person can.”

The RightHand Robotics system can run on any robot arm, but Cheng says that many customers prefer UR robot arms because they are collaborative and safe to work side-by-side with humans in warehouse environments. “To have collaborative arms that are as easy to work with as the UR arms gives us a very accessible solution,” Cheng says. She adds, “I was blown away by how easy it was to use, and how elegant the whole process was.”

Fulfilment productivity increased by 8-10x

The RightHand Robotics technology has also been incorporated into multiple other logistics systems, such as the Autofulfillment SPrint System from Accutech Packaging in Foxboro, Massachusetts. The system automates the outbound bagging, labeling, sorting, and shipping of merchandise such as health and beauty aids, electronics and office products.

Mike Keneally, vice president of Accutech Packaging, explains the advantages of automating these processes for e-commerce retailers or pharmacies: “They are manually opening bags, loading product into it, printing a pack slip, putting it in the bag, and then applying a shipping label. All of that is very time-consuming. On average, a packer can operate at 65 to 85 packages an hour. With our technology, we are able to increase the productivity of an individual packer to something like 600 packages an hour.”

That’s an 8x to nearly 10x improvement in productivity, which is critical to meet consumer demands for fast product fulfilment. The RightHand Robotics system combines a fingered gripper with a suction cup along with a UR robot arm to provide a system that is reliable and fast to meet the 600 packages-per-hour goal.

Accutech’s Keneally says that the market demand for this system is phenomenal. “Everything is moving online, and so they’re looking for better ways of packaging their items, particularly when it is so difficult to find many people to work for them,” Keneally states. “Adding that robotic technology is significant to enable our customers to really fulfil their customers’ requirements.”

Cobot sorting addresses labour shortage

Eurosort – based in Amsterdam and with North American operations in Baltimore, Maryland – is also using the RightHand Robotics system with its Bomb-Bay sorter (also known as a flatsorter or split-tray sorter). The system is used for high-throughput applications for lightweight items under 25 pounds, such as apparel, books, pharmaceuticals, polybags, and more. It can sort from 7,000 items to 30,000 items an hour.

Scott Eisenberg, key account manager for Eurosort, is very familiar with the challenge of filling material-handling jobs, and how collaborative automation can help. “In the whole material-handling industry, there’s a labour problem – either a labour shortage or quality of labour – and having robotic induction helps with that,” he says. That’s especially true when warehouses are located far from areas where workers live, and during peak seasons when it can be difficult to find and train workers.

The Rightpick system, inducting onto the EuroSort Split-Tray sorter, is designed to work alongside people. That makes the collaborative robot arm important. Eisenberg explains, “With a lot of other arms, you have to have a big cage, you have to kind of keep them away from the rest of the inductors. By using the UR arm, we can stand right next to it.” Human workers can be inducting along with the robot, and not have to worry about their safety. “It’s a really great collaborative effort between robot and person,” he states.

Visit the Universal Robots website for more information

See all stories for Universal Robots

Disclaimer: Robotics Update is not responsible for the content of submitted or externally produced articles and images. Click here to email us about any errors or omissions contained within this article