And other companies are making progress as well.
For decades, training a robot to be more human has puzzled engineers, unable to replicate the ability to grasp and move objects. But now the gains in artificial intelligence technology, cameras and engineering are bearing fruit, allowing robots to see objects of different shapes and sizes and adjust their grip accordingly.
The technology, say computer scientists, has finally become reliable enough that companies have found it viable to deploy.
“This moment is a turning point,” said Kris Hauser, a robotics expert and computer science professor at the University of Illinois at Urbana-Champaign. “They are capable enough at this point.”
But there is also a controversial debate. Critics worry that robots will take people’s jobs, although advocates say they will simply do different jobs. Others noted that more robots could result in higher worker injury rates, or result in tighter human surveillance to make sure they hit targets.
Beth Gutelius, a professor of economic development at the University of Illinois at Chicago, says that the way companies are releasing these robots is without much testing or attention to worker safety.
“Don’t we all want these things to work better for more people?” he said.
Amazon founder Jeff Bezos owns The Washington Post.
Robots have been on the scene for years, but it’s been a slog for scientists to get them to replicate tasks as well as humans — especially when it comes to hands. Amazon has Kiva robots, which look like Roombas and move packages around the factory floor, but still require people to pack and sort them.
Elon Musk famously said he would automate Tesla manufacturing, but people still need to work on the assembly line at the factory in Fremont, California. He also recently unveiled Tesla’s Optimus humanoid robot prototype, which aims to revolutionize physical work.
Google recently unveiled robots that use artificial intelligence to help people with everyday tasks. Some robots have even learned how to cook fries.
Despite the advances, the most difficult challenge for researchers is teaching robots to adjust their grips to different sizes and shapes, said Ken Goldberg, a professor in industrial engineering at the University of California at Berkeley.
But in the last decade, things have started to change, he said. 3D camera technology, inspired by Microsoft’s Kinect motion sensing camera, has gotten better at viewing images. Deep learning, a field of artificial intelligence that uses algorithms modeled on the brain, allows computers to analyze large amounts of images. Researchers are beginning to better understand the physics of grasping objects, and that includes robotic suction cups and pickers.
The result: modern-day robotic machines that often look like long arms. Their vision is fueled by software that uses machine learning algorithms to analyze how objects look to teach robots how to hold objects. Suction cups or claws adjust pressure and control with a finesse that people take for granted.
Amazon in particular is chasing the technology, industry experts said. As one of the world’s largest retailers, plagued by high turnover rates and promises to deliver packages quickly, it made strong financial sense to try to automate warehouse processes as much as possible.
In 2012, the company acquired mobile robotics company Kiva for $775 million in cash. In 2014, the company announced a “picking challenge,” challenging scientists to build robots that could pick a variety of items, ranging from Sharpies to Oreo cookie packages, from a mobile shelf.
Last month, Amazon unveiled its picking-and-sorting robot called Sparrow, a long robotic arm that can handle items before placing them in boxes. It was researched and developed in Massachusetts and in operation at an Amazon facility in Dallas, officials said. It can sort roughly 65 percent of the products in its inventory, according to company officials, but plans to expand nationwide have not been set.
The robot fits into a broader automation strategy, according to Amazon. If mastered, Sparrow can pick products after they are offloaded from trucks and before they are packed and placed on mobile shelving. Once boxed, Amazon’s robotic system, called Robin, can sort them to their destination. Cardinal, another robotic machine, can place it on a waiting cart, before being loaded onto a truck.
Amazon has always said that more machines will allow people to find better jobs. The robots “do some repetitive tasks within our operations, freeing up our employees to work on other tasks that are more attractive,” said Xavier Van Chau, a company spokesman. .
In March, postal delivery giant Pitney Bowes signed a $23 million deal with Ambi Robotics to use the company’s picking and sorting robots to help sort packages of different shapes, sizes on and packaging materials. In August, FedEx agreed to buy $200 million in warehouse robotics from Berkshire Gray to perform similar tasks. A few months before that, it launched an AI-fueled mail sorting robot in China.
Although much of the technology began to appear several years ago, it takes time to ensure that these systems can reduce errors to less than 1 percent, Hauser said, which is important for the bottom lines of the company.
“Every mistake is expensive,” he added. “But now, [robots] is at a point where we can really show: ‘Hey, this is going to be as reliable as your conveyor belt.’”
The revenue earned by companies that make pick-and-order robots is skyrocketing, said Ash Sharma, a robotics and warehouse industry expert at Interact Analysis, a market research firm.
The research firm estimates that companies that make these products will make $365 million this year. Next year, it is estimated to exceed $640 million. This is a jump from almost $200 million last year and $50 million in 2020 these companies earned in revenue, data forecasts show.
A big reason is the lack of workers, he said.
Gutelius, of the University of Illinois at Chicago, said that although the technology has proven interesting, it has risks. With more robots on warehouse floors, the workers with them have to work at a faster pace, risking more injuries.
The Washington Post reported that Amazon’s warehouses could be more dangerous than those of rivals. Experts say adding robots to the process will increase injuries.
Van Chau said machines that perform repetitive tasks can help workers. “We can take some of that strain out of the employees,” he said.
But Gutelius said companies that claim these robots can help should be scrutinized, saying the solutions are too easy to implement.
“It’s a classic ‘move fast and break things,'” he said. “And in this case, I think ‘breaking things,’ it’s going to be people.”
