Human-robot teams working together will become more common across industries. For example, humans and automated workers can work together to intelligently navigate a warehouse and select items to ship to fulfill online orders.
Researchers at the University of Missouri are moving closer to this reality by developing a software model designed to make “transport” robots smarter.
A study titled “Collaborative Order Picking with Multiple Pickers and Robots: An Integrated Approach for Order Batching, Sequencing, and Routing of Pickers and Robots” International Journal of Production Economics.
Optimizing human-robot collaboration
Sharan Srinivas is an Assistant Professor in the Department of Industrial and Manufacturing Systems Engineering and Marketing.
“Robot technology already exists,” says Srinivas. “Our goal is to get the most out of this technology through efficient planning. How do you optimize it?” or “How many items should the robot pick on a given tour? Or “In what order should the items be collected on a given robot tour?” also have a similar question. The hardest part is optimizing the collaboration plan between human pickers and robots. ”
Much of the human effort and labor costs in this process come from online order fulfillment. Robotics companies are looking to optimize processes by developing collaborative robots, often called cobots or autonomous mobile robots (AMRs). These bots can operate in a variety of environments, such as warehouses and distribution centers, and are typically equipped with sensors and cameras to aid in navigation. The new model results in faster fulfillment of customer orders by optimizing key decisions or questions about collaborative order picking.
“The robot is so intelligent that if it is directed to a specific location, it will move through the warehouse and not hit workers or other obstacles along the way,” said Srinivas.
No substitute for human workers
Srinivas specializes in data analytics and operations research. According to the professor, AMRs are not designed to replace human workers. Instead, they work together to make the order fulfillment process more efficient. For example, bots can help fulfill orders faster than human workers. At the same time, human workers must pick items from shelves, load them onto robots, and transport them to designated drop-off points within the warehouse.
“One of the drawbacks is that these robots don’t have good grasping ability,” Srinivas said. “But humans are good at grabbing things, so we’re trying to leverage the resource strengths of both human workers and cobots. Instead of one worker going through the entire aisle and picking up multiple items along the way, the robot will come to a human worker who will pick up the items and place them on the robot. Human workers don’t have to strain themselves to move large carts of heavy items throughout the warehouse.”
Srinivas also said future applications of the software could be extended to other locations, such as grocery stores, within three to five years. The robot could fill orders while moving among the masses.
