iRobot CEO decries the slow advance of robotics

Lack of funding, too much star-gazing are hindering the robotics industry, says Colin Angle

The CEO of one of the biggest and arguably the best-known robotics companies in the world is disappointed at the progress being made in the industry, which he says is nowhere near where he thought it would be today.

"We are just about none of the way to where we need to be to create a robot industry that is understood as a robot industry and not an interesting adjunct to another industry," said

Colin Angle iRobot

Colin Angle, CEO and co-founder of iRobot

Colin Angle, CEO and co-founder of iRobot. "Where I thought we'd be? No way."

iRobot builds robots for the military, but is better known for its Roomba vacuum cleaner.

Face it, a lot of people grew up watching Star Wars, Blade Runner and Battlestar Galactica -- and, of course, The Jetsons with its robo-maid.

But in 2014 there are no robots masquerading as humans and no C-3POs or R2-D2-like companions. There definitely are no robotic maids folding our laundry, making dinner and getting the kids off to school.

So what happened to the future everyone was imagining?

"Here we are. iRobot, is just entering its twenty-fifth year," Angle said. "I think I would have expected a lot more diversity of robots in the marketplace that I could go buy that would do some real stuff. Twenty-five years has passed since we started and we've accomplished..., I won't say nothing. We're starting to see the next generation of manufacturing robots, but that's still relatively early. That's not very blue sky."

He added that he considers it a win that Bedford, Mass.-based iRobot has changed the vacuum industry, noting that last year robotic vacuums captured 18% of all dollars spent on vacuums worldwide.

"If 25 years ago, you'd said, 'Do you think you'll change the vacuuming industry?' I might not have said, yes," Angle said in an interview after the RoboBusines conference in Boston last week. "At least we have that in the surprise win column. The rest of it is either not surprising or disappointing."

To be sure, there are a lot of robots working in the world.

Large industrial robots have been working in manufacturing plants for decades. Robotic arms on board the International Space Station capture visiting spacecraft and unload cargo. Robotic rovers are exploring Mars, while autonomous cars are being tested on city streets and the military is investigating the use of weaponized autonomous robots.

While great strides are being made in robotics, few think the machines are where they could be by now.

For instance, teams from the likes of MIT, Virginia Tech and Worcester Polytechnic Institute are readying their 6-foot-tall humanoid robots to compete in the finals of the DARPA robotics challenge next June. Roboticists are creating the software needed to get the robots to climb stairs and ladders, open doors and even drive cars.

But in the last DARPA challenge, the robots' movements were slow and halting. Some fell over, unable to get back up. The autonomy was more about getting the machines to balance themselves than to carry out complex tasks on their own.

It's not all bad news, though, said Dan Olds, an analyst with The Gabriel Consulting Group. It's all about perspective and adjusting expectations.

"Many of those who decry the lack of advances in robotics are overlooking the massive number of instances where sophisticated robotics are being used," he added. "While they've been looking for a robot butler, they've missed seeing that the Navy is close to having an automated seaborne battle group or the built-in robot that makes sure their car doesn't smash into the car ahead of them or drive off the road."

Angle and other roboticists, though, feel we could be further along in building robots that make life easier and safer, or act as assistants and companions.

The problem with progress is three-fold, according to Angle, who worked on a miniature design for NASA's Mars rovers and holds a bachelor's degree in electrical engineering and a master's degree in computer science, both from MIT.

The robotics industry simply isn't getting the influx of funding it needs. Researchers are getting too hung up on the cool technology, instead of focusing on robotics that can be put to work. And the robotics industry simply needs a big win to get the attention, momentum and financial backing it needs to explode.

Pam Henderson, a former assistant professor at Carnegie Mellon University and co-founder and CEO of consulting firm NewEdge, Inc., agreed, adding that too many roboticists are in love with their own ideas.

"Some of you have one idea you are in love with and you are going to ride that thing until it goes to market or it kills you," Henderson said during a keynote talk at RoboBusiness. "Robotics is not the industry. Appliances are the industry. Home health care is the industry. The need isn't for a robot.... What's the opportunity? No opportunity? Don't build. Be pretty crisp on the applications before you do the development."

That's a lesson iRobot and its executives had to learn.

Last year, the company generated $487 million in revenue and employed more than 500 people. But iRobot wouldn't have succeeded if Angle and his partners hadn't focused on the need, instead of the robot.

"We could have built a robot that could climb stairs because it's super cool, but we would have gone out of business," said Angle. "Hundreds of millions of dollars have been wasted building cool demos instead of useful products in the robot space."

He pointed to Sony's Aibo, a series of robotic pets officially launched in 1999 and discontinued in 2006.

"It cost them huge amounts of money every year to continue to advance the robot," said Angle. "They may have created valuable marketing and PR events to enhance the reputation of the builder, but they didn't advance the robot industry."

Michael Gennert, director of robotics engineering at WPI, said it's easy for researchers to get obsessed with an idea or a technology but, at least at his university, educators are trying to get students to look at the bigger picture.

"It's a rookie mistake. A lot of college graduates are kind of in love with their technology and don't think if their product has a compelling value proposition," Gennert said. "You have to understand the opportunity. It's not enough just to be cool. You have to be useful as well."

To propel students in that direction, WPI requires its robotics engineering students to take a course in entrepreneurship.

However, even if someone has a great idea to develop a useful robot, they still need to get the funding to create it. That's an expensive proposition at a time when funding is more of a trickle than a flood.

"It takes $20 million or more to build a legitimate robot. It's a much bigger check to write to get to play," said Angle, who noted that iRobot spends $40 million a year on Roomba and related research. "It's expensive to do what we're doing.... If it takes $20 million to build a legitimate robot, where do you find that as a startup? You need someone to take such a big bet."

To get that kind of funding, the industry needs a big, high-profile win to entice venture capitalists and companies to open up their wallets and invest. That means someone needs to develop that next great robot that makes splashy headlines -- or Google needs to add to the list of robotic companies it's acquiring.

"iRobot has a market capital around $1 billion and that makes us the 35-pound gorilla of the robotic space," said Angle. "I'd love to be the 350-pound gorilla. It's just not good enough yet. We need more big wins. Rod [Brooks, founder and CTO of Rethink Robotics] can have a good success with Rethink and that could have a significant impact. It helped that Google bought a lot of companies last year. But the track record is not yet where it needs to be."

However, Angle isn't pessimistic. Advances may not be coming as fast as he'd, like but he says they are coming.

Google is working on its self-driving car and the company, along with Amazon, is working on delivery drones. The U.S. Navy is testing those autonomous boats.

"The Internet of Things, which is an eminently investable term right now, is in many ways closely overlapping with robotics," said Angle. "It's going to be a very cool next five years as we see more economic success for robotic entrepreneurs, who may have to call themselves Internet of Things entrepreneurs.

"Things are absolutely changing," he added. "It'll still be maddeningly slow but it's accelerating."

Copyright © 2014 IDG Communications, Inc.

7 inconvenient truths about the hybrid work trend
Shop Tech Products at Amazon