Advertisement

Blog

Amazon Bets on Advanced Robotics for Future Fulfillment Centers

Amazon recently held its annual robotic competition, Amazon Robotics Challenge, focusing on picking and packing. This year’s event took place in Japan at the Nagoya International Exhibition Hall, and the company invited 16 teams from all over the world to participate, all expenses paid. A team from the Australian Center for Robotic Vision won with their robot called Cartman, a robotic arm capable of identifying, picking, and stowing a number of different objects.

Team Cartman

Team Cartman

Amazon is growing extremely fast, constantly increasing its market share and entering new markets. The ecommerce giant sells in 15 countries and has fulfillment centers in 13, including seven in Europe and three in Asia. Additionally, the company has an European corporate office in Luxembourg, and call centers in the UK, Germany, Ireland, Italy, and Morocco.

The online giant is becoming one of the largest employers in the United States, with over 90,000 full time employees and over 200,000 temporary workers. These numbers are nothing compared with Walmart, which has over 2.3 million people on its payroll, but Amazon’s workforce is growing at a phenomenal rate every year. That is why the company believes that robotization of its fulfillment centers is the key to sustainable growth.

The drive toward automation doesn’t mean Jeff Bezos is planning to reduce the company’s headcount. It does mean Amazon wants to grow without having to hire more people at the same rate.

Amazon’s race to automate its fulfillment operations started right after the company went public in 1997, but it did not start using robots in its warehouses until 2012, when the company acquired Massachusetts-based Kiva Systems for $775 million.

Kiva Systems invented a revolutionary way of managing vast warehouses by using fleets of mobile robots, or drive units –which look much like your roomba vacuum cleaner–  to sort, organize, and transport inventory. Kiva Systems was renamed Amazon Robotics after the purchase was final.

The Amazon robotics warehouse system looks rudimentary compared with the huge, arm-whirling automated robots that can be seen at some storage facilities or on automaker assembly lines. But the challenge for the eCommerce supply chain is very different from that of a factory assembly line.

Kiva Systems developed a way to use robots to move the racks, or “pods,” on which the products are stored, instead of having them search out and move the individual products themselves. Storage pods can hold hundreds of different types products, but the size and shape of each pod is exactly the same.

Currently Amazon uses more than 100,000 Amazon Robotics drive units in over 25 fulfillment centers worldwide.

While the drive units solve the problem of moving items around the warehouse and can be used to move entire fulfillment centers, they can’t handle the most important tasks for Amazon’s logistics: picking and packing.

Over 50% of American households are members of Amazon Prime, which promises next-day delivery on millions of items. The need for speed and accuracy also contributes to the retailer’s drive to further automate fulfillment operations.

“Commercially viable automated picking in unstructured environments still remains a difficult challenge,” says Amazon Robotics on its website.

That is where the Amazon Robotics Challenge comes in. Being held for the third time this year, Amazon offered $250,000 to the team who could show they could successfully pick and pack different products fast enough to compete with human operators.

The participating teams were tasked to show that their robot software and hardware could recognize objects, grab them, execute tasks, detect errors and recover as needed. The robots were scored by how many items were successfully picked and stowed in the allotted time.

To further complicate the challenge Amazon changed the products to pick and stow during the final round, so teams were not able to reprogram their robots if they had been trained for specific items.

“This year, we made some changes to the Challenge to make it even more difficult and to encourage broader participation from multiple robotics fields – and the response was exciting,” says Joey Durham, contest chairperson and manager of research and advanced development for Amazon Robotics. “The versatility of recognition capabilities in an unstructured environment and the dexterity of grasping mechanisms was truly impressive. What we're most proud of with the Amazon Robotics Challenge is its celebration of robotic community and the venue it's created to share and promote research in a fun and rewarding way. Congratulations to our winners and all of our contestants!”

The Australian Centre for Robotic Vision won the competition using their robot Cartman,  a three-axis Cartesian robot –a robot whose axes of control move in a straight line rather than rotate– developed by the team that grabs items from above, for maximum pick and stow efficiency.

The Aussies developed their own Cartesian robot for the challenge, “we were the only team with a Cartesian robot at the event. Cartman was definitely a large reason for our success,” said team leader Dr Juxi Leitner. “The Amazon challenge is a chance to really advance the Centre’s mission of creating robots that see and understand a task in a real world environment like a warehouse.”

It is still too soon for Amazon to start replacing humans in its picking and packing operations at fulfillment centers. But the Amazon Robotic Challenge shows promising technology which is advancing faster every year. It won’t be long before the retailer’s quest for a fully automated warehouse becomes a reality.

1 comment on “Amazon Bets on Advanced Robotics for Future Fulfillment Centers

  1. ninud
    November 2, 2017

    Walmart is really big and I like their service and especially the walmartone account login. they have improved a lot in recent years in terms of service.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.