Robots may be able to produce products at a rapid speed, detect safety issues or other risks typically with precision and accuracy but it seems some everyday human tasks still aren’t mastered by robotkind.

Those tasks could include things we do every day like preparing coffee or folding clean laundry.

University of California, Berkeley researchers have created Blue, a human-friendly, affordable robot that may someday be able to help people with those mundane household tasks so humans have time for more important things.

Blue was created using recent artificial intelligence and deep reinforcement learning advances making it able to master human tasks while staying affordable and safe for home use.

Pieter Abbeel, professor of electrical engineering and computer sciences at UC Berkeley, postdoctoral research fellow Stephen McKinley and graduate student David Gealy are the brains behind Blue.

"AI has done a lot for existing robots, but we wanted to design a robot that is right for AI," Abbeel said. "Existing robots are too expensive, not safe around humans and similarly not safe around themselves — if they learn through trial and error, they will easily break themselves. We wanted to create a new robot that is right for the AI age rather than for the high-precision, sub-millimeter, factory automation age."

Abbeel has pioneered deep reinforcement learning algorithms that help robots learn by trial and error or by being guided by a human like a puppet. He developed these algorithms using robots built by outside companies, which market them for tens of thousands of dollars.

Blue costs less than $5,000 to manufacture. The robot is made with durable plastic parts and high-performance motors that power its two arms. Each arm is about the size of an average bodybuilder’s and is sensitive to outside triggers like a hand pushing it away. The arms also can range from stiff and traditionally robotic to very flexible, almost limp like a real human arm.

The traditionally rigid robot is useful in industrial settings for doing the same task over and over. But most of us would agree that the traditional home, especially one with children or pets, is anything but rigid with repetitive tasks. Blue was designed to be able to handle these outside factors by varying the amount of force it exerts based on the environment it is in.

"We've often described these industrial robots as moving statues," Gealy said. "They are very rigid, meant to go from point A to point B and back to point A perfectly. But if you command them to go a centimeter past a table or a wall, they are going to smash into the wall and lock up, break themselves or break the wall. Nothing good."

Blue is able to continually hold up 2 kilograms of weight with the arms fully extended. However, unlike traditional robots, Blue is "thermally limited," McKinley said.

Just like a human being, Blue can exert force in a quick burst of much more than 2 kilograms but it needs time to rest or cool down. Just like how a person can lift a heavy box to move it across the room but likely could not take that same box on a three-mile walk.

"Essentially, we can get more out of a weaker robot," Gealy said. "And a weaker robot is just safer. The strongest robot is most dangerous. We wanted to design the weakest robot that could still do really useful stuff."

The team is currently working on 10 arms to distribute to early adopters as they continue to investigate Blue’s capabilities. If they find that Blue has the potential for manufacturing on a larger scale, that may happen through the UC Berkeley spinoff Berkeley Open Arms.