Automate 2022: Teach mobile robots to stop being such fools

Bots don’t care about your feelings, and it bothers the people who work with and around them.

“We’ve created algorithms that make robots greedy and jerky,” said Melonee Wise, vice president of robotic automation at Zebra Technologies. And jerkily, she does not speak of abrupt and discordant movements. She talks about machines acting like over-groomed kids in 80s movies with their collars up. “They see everything as a goal or an obstacle.”

Speaking during a panel discussion at the Automate show in Detroit on Wednesday, Wise explained that the robots broke free from their cages, first with collaborative robots (cobots) who could work safely alongside people and more recently with the explosion of Autonomous Mobile Robots (AMR) that roam factories performing various tasks.

This robotic evolution was evident on the show floor. Ten years ago, Automate was dominated by 6-axis stationary robots, conveyor systems, and other pieces of hardware infrastructure automation. Immediately before COVID, cobots were the hot items with startups and established leaders showing such models. This year, almost everything was on wheels, with AMRs in kiosks at:

3M ABB Doosan
Fanuc Flex Qube Gideon
Kuka Venue MiR
Nashi Otto engines SEW Eurodrive
Staubli Yaskawa

The problem, as Wise said, is that robots do exactly what they were programmed to do. It was fine when they were locked up and not interacting with people. Now that they’re free, people need to feel comfortable around them, so robots need etiquette lessons.

For over a year, Wise’s team has been rewriting algorithms to focus on social signals. Some things are minor – Americans tend to walk on the right side of sidewalks and paths and pass slower moving people on their left side. The robots flipped a coin each time they chose where to drive and which side to pass. The result: people didn’t like working near them because they felt rude or out of place.

Other behavior cues are more complicated, such as knowing that the robot can continue to move towards a person in its path, as long as that person is moving in a predictable way.

Ignoring social prompts isn’t just about making bots less scary. It’s a matter of productivity. Most of the time, when a mobile robot encounters an obstacle, it slows down or stops to avoid collisions, then it thinks about what to do next and chooses a new path to its destination. In one video, Wise showed that a person simply crossing in front of a robot’s path caused it to stop for half a minute and then take a circuitous route to avoid where the person had been – a person who didn’t was then nowhere near the robot. Thus, each of these interactions slows down the machines.

“Getting rid of those weird navigation artifacts is key to robot speed,” Wise explained, adding that in one test, making robots a little more aware of human behavior increased average movement speed by 51%.

Liam Mongovan, sales manager at Mobile Industrial Robots (MiR), said workers should expect to see more AMRs in the coming months as many companies turn to such systems to address staff shortages. Mobile robots can usually be deployed quickly, making them an attractive short-term solution.

“It’s a quick fix to the three Ds – dirty, boring or dangerous,” Mongovan said. With businesses having such a hard time finding people, many are choosing AMRs where they once used forklift drivers or people sorting boxes for delivery.

“We’re talking hours a day. How many hours of work can you shift from people to robots? It becomes the number of people you can redeploy to your factory to do more useful work,” he added.

Owners of several robotics companies said the large number of AMRs came as no surprise to them at this year’s show, the same way the large number of cobots came as no surprise at IMTS 2018 or Automate 2019. Even before the pandemic, companies were seeing success stories of manufacturers using AMRs to automate shipping tools or supplies to workers on assembly lines, getting pallets of goods to shipping docks, and performing simple production tasks such as drilling holes in parts.

Wise said the direction for the industry is clear: more companies will offer more mobile options, putting more workers in the paths of moving robots. In this case, current security algorithms and protocols will hamper productivity, hence his group’s focus on human-robot interactions. On the positive side, his research projects improved robotic performance without radically redesigning the hardware. On the other hand, teaching a robot workshop manners is really difficult.

“It took a year of research with engineers and navigation algorithm experts to show these improvements,” Wise said. “So it’s not simple or easy, but it doesn’t involve redesigning everything else or adding expensive new components.”

Casey J. Nelson