The Economist of March 29th – April 4th 2014 contains a special 14-page report on The rise of the robots. The feature contains an appealing review on the relationship between people and the technology, apart from providing an overview of the development of robots, their current military, business and care uses, as well as a whole chapter on the impact of robots on the labour market which deserves a separate article on its own. Right now I want to look at the concept of cloud robotics.
One of the assumptions put forward by the authors is the fact that as of today, we have no artificial intelligence (AI) so all robots that we produce are “dumber than a doornail”. This is unlikely to change any time soon. The robots that are produced or used now are capable of following instructions, moving trolleys in hospitals or detecting tone of voice. They are, however, incapable of any more complex tasks, such as face recognition, or for that matter telling a tree from a house from a human being. Even with the best optical sensors on the market, the authors argue, a robot is still largely able just to capture the image but can’t really analyse it, not to mention acting upon it.
Of course, the world is full of people who dream of developing AI and some of them even work towards it. However, a more achievable solution may be coming from a technology that we’re pretty familiar with: the cloud.
The EU has been investing in RoboEarth, “a worldwide web for robots” for four years now. The aim of the project is to develop “a World Wide Web for robots: a giant network and database repository where robots can share information and learn from each other about their behavior and their environment”. In other words, by using cloud robotics, a robot could connect to RoboEarth to use vast amounts of data (big, big data) to analyse a captured image against the repositories, leading to it being able to tell that there’s a house, a tree and a human being in danger in the image.
Much further in the report, when discussing domestic service robots, the authors bring up the example of robots operating trolleys at American hospitals. Tug robots operate within their clouds where they can connect with a staffed help desk that deals with queries from robots: “if one gets stuck or lost, a remote operator can look through its eyes, check its logs and sort things out before the hospital concerned even becomes aware that anything is wrong”.
After investigating cloud robotics a bit, I started wondering about the future where the cloud we, humans, know and use, becomes combined with the robots’ cloud. I keep a lot of information in the cloud and personal clouds are more and more common. Let’s imagine that I buy a robot and I hook him (yes, it would definitely be a he) up to my cloud. Tens of scenarios unfold in my head, from remote presence (through my robot) to running complicated analyses on my lifetime photos. If a robot had access to all my digital artefacts and I had access to his capabilities, would that be the ultimate superhuman?