Baxter has been designed to react to situations and adapt to changing environments. Baxter will move slower when it senses humans within a narrow range of itself and if it did bump into someone it would not cause damage. The robot also has the ability to sort products based on colour, size and weight, and is being used in a range of commercial operations in the States, including packing boxes, sorting out parts and doing light assembly work. KanDO Innovation's business development manager Ash Taylor said the ease of programming Baxter was a big advantage for users.
"With traditional automation solutions, we are used to spending $100,000-plus on the factory reconfiguration and the safety cages necessary to keep the robot away from humans, so with this cost taken away and no factory reconfiguration required, this drives a whole new alternative," Taylor said. "Also with the ease of ongoing programming and software changes being delivered through a much friendlier interface this lowers the technical barrier for workers to interface with Baxter."
Taylor said the company had been talking to a few initial customers and would begin marketing and looking for more clients from next year.
Helpful droids with heads in the cloud
A vaguely humanoid robot served juice to a researcher on a hospital bed. The robot then uploaded its memory of the experience to a system of cloud servers, essentially a shared global brain. When the next robot came along, it had downloaded the memory and knew how to find the juice and get to the bed.
The phenomenon of robots teaching one another, called transfer learning, could prove increasingly useful as more people rely on robots for medical care and other services.
A robot facing a row of unfamiliar objects could locate the one it needs, check with the cloud about the best strategy for grasping it, and pick it up even if it had not been trained to do so directly, says Gajan Mohanarajah, who worked with the juice-serving robots while pursuing a PhD at Swiss university ETH Zurich.
He has spent more than four years working to develop the technology as part of RoboEarth, a project by academics at six European universities funded by the EU.
The initial RoboEarth project officially ended with the juice experiment this year, but spinoff groups are working to close the gaps in the development of a multipurpose transfer-learning system.
Robots using it would have to be able to communicate in the same language and maintain consistent wireless connectivity. And what if hackers could compromise these systems full of sensitive data?
"The security aspect has to be dealt with very delicately, in terms of building a public network," says Mohanarajah, adding that a logical first step would be a series of private networks operated by individual companies or government agencies.
Moritz Tenorth, a researcher at the Institute for Artificial Intelligence at the University of Bremen in Germany, helped to develop RoboEarth's common machine language and is now working on a related project called RoboHow.
His goal is to build a system that can take information from the web, convert it into a format that robots can understand, and upload it to a memory bank in the cloud. The challenge is to include all the steps a person might not have to think about: most recipes for pancakes don't come with instructions on operating a stove.
If RoboHow succeeds, a robot could eventually be able to access the internet and find directions for itself. For now, humans curate the data.
"It's very difficult for a robot to assess the quality of this information. It could be a fictitious text or a piece of malicious code," says Tenorth. "I'm not sure we want to hook up robots to the open web right now. I think their heads would explode."
- Bloomberg
Machine age
• Baxter the robot will be available in New Zealand from next year
• The robot is expected to cost around 30 per cent less than other robots
• Baxter stands at 1.8m tall and has arms and a LCD screen face which can display different emotions
