Editorial: “Is telepresence the next big thing?“
See more in our gallery: “Telepresence today: How you can live by remote control“
“Devon. I told you, go and brush your teeth!” It’s Thursday morning and 7-year-old Devon Carrow-Sperduti is meant to be starting school in 5 minutes. His mum is getting impatient. Devon is like any child his age – in all but one respect.
Advertisement
When Devon gets to Winchester Elementary School in West Seneca, New York, he chats with his friends between his classes, he sometimes gets told off by his teachers for not paying attention and, occasionally, he bumps into walls. It is an inevitable consequence of attending school as a robot.
Devon has allergies that prevent him from physically being at school. Instead, he stays at home and logs into a two-wheeled, 1.5-metre-tall Segway-like robot called VGo, which is waiting in the school grounds. He navigates between classrooms by peering through a camera, and talks with classmates and teachers via a real-time video screen displaying his face. “He’s treated the same as everyone else,” says his mum. “If he doesn’t turn up to class, they ask why.”
Devon isn’t the only one routinely transporting himself to another location like this. He joins surgeons, soldiers and an increasing number of other workers who are turning to an army of surrogates often hundreds of kilometres away. These virtual travellers can hold down nine-to-five jobs, fight wars and perform life-saving operations.
This year, you’ll be able to buy one for the same price as a laptop, and eventually they will be controlled by thought alone and will transmit a sense of touch back to their pilot. It means our senses will become immersed in another location like never before. Researchers, legal experts and ethicists are realising that the way this technology will be used over the next decade and beyond is not only going to affect the way we live and work, it is also going to disrupt economies, challenge laws and may even transform social norms.
Giving people the ability to step into machines in distant places is an idea that can be traced back to at least the early 1980s, when , one of the pioneers of robotics, . He used it to refer to the suite of technologies that allow a person to feel as if they are present at a place other than their true location. In his futuristic vision, these robotic systems would pave the way for a “remote-controlled economy” and would transform society.
Although Minsky’s vision hasn’t turned out exactly as he imagined, these technologies have nonetheless come on in leaps and bounds in the internet age. The first wave of telepresence simply allowed people to peer at remote places through a camera, and to appear at the other end, too. One of the first such systems introduced in a workplace was at . In their Media Space project, having cameras in labs in Palo Alto, California, and Portland, Oregon, meant people could collaborate on video screens almost 1000 kilometres apart. This ability is now commonplace, thanks to real-time video apps like Skype or FaceTime.
Subsequent generations of telepresence have added to that experience by allowing people to move around and to manipulate objects. Some of the first organisations to embrace this widely have been hospitals and the military – and it has changed medicine and warfare.
If you’re in the US and need part of your prostate removed, for example, it is likely you will experience the skills of a telepresent surgeon using a robotic manipulator – 90 per cent of these operations are now performed this way (). Scalpels have even been wielded across oceans: in 2001, surgeons in New York removed the gall bladder of a woman in Strasbourg, France, more than 6000 kilometres away.
Meanwhile, soldiers today routinely control aerial drones and robots for surveillance, bomb disposal and even attacks. In fact, their use is now so common in the US army that some commentators argue that remote-controlled warfare could come to be seen as .
In the past few years, telepresence devices designed for less hostile purposes have been gaining an increasing presence in the mass market. VGo Communications in Nashua, New Hampshire, and Anybots in Mountain View, California, for example, sell roving telepresence robots to the general public. These can be moved around from any computer with a web camera, using a browser.
This market is young and relatively small, so it is difficult to get a broad view of where you might spot one of these surrogates. But so far, the signs suggest that people have been using them to visit family and friends, tour buildings like museums, work remotely with distant colleagues or, for doctors and nurses, to check on patients from afar.
VGo claims its customers range from hospitals to corporations like Nokia and Ericsson. Meanwhile, Anybots says it has shipped about 70 bots since launching in 2010, to companies such as a Mexican manufacturer whose employees get virtual visits from out-of-town managers to inspect their work. It has also supplied bots for patients at Stanford Hospital and Clinics in Palo Alto. The doctors use them to treat children who have to be kept in isolation. Other robots are placed back at the child’s home so that they can attend birthdays or family gatherings. “The kid can control his own fate,” says Tim Lenihan of Anybots. “If he wants to talk to his mum he can log into his Anybot and walk into her bedroom and tell her ‘I’m lonely, come visit’.”
These bots are pricey, though. A VGo costs $6000, and Anybots start at $9700. That may be why a company called Double Robotics in Miami, Florida, is tipped to make waves this year. Next month, it is due to launch a roving telepresence robot – with an iPad for a head – for the same price as a high-end laptop. Double’s first production run sold out, and the company is reported to have secured almost $2 million of pre-orders for its 2013 run. Its customers reportedly include numerous US universities and Fortune 500 companies.
Of course, past failures of “telecommuting” to catch on widely suggest that these devices won’t completely replace face-to-face meetings or communal working spaces. Still, they do provide us with a way of communicating that wasn’t possible before. And the broader progression of the underlying technology suggests it is only going to become easier and easier for people to control machines from afar in the coming years.
The next wave of telepresence under development in laboratories suggests the technology will become significantly more immersive. For example, a team led by at University College London (UCL) has built a surrogate robot whose actions mirror a person’s body movements. Hold out your arm for a handshake, and the robot’s arm follows suit. The bot isn’t mobile, but you can greet people, gesture and manipulate objects. It is currently controlled by a bodysuit but, in principle, motion capture like Kinect could be employed to do the same job down the line, “and at a fraction of the cost”, says Slater.
True embodiment
Immersion needn’t stop there. Early this year, a student called Tirosh Shapira controlled a robot using only his thoughts – he was at Bar-Ilan University in Israel, inside an fMRI brain scanner, and the robot was in France. The scanner tracked changes in blood flow to his brain’s primary motor cortex, and then an algorithm distinguished when he was thinking about moving his left hand, right hand or legs. “True embodiment goes far beyond classical telepresence, by making you feel that the thing you are embodying is part of you,” says of the CNRS-AIST Joint Robotics Laboratory in Tsukuba, Japan, part of the team involved in (91av, 7 July, p 9).
Sure enough, Shapira found it a peculiar experience. “It was mind-blowing,” he says. “I really felt like I was there. When the guys in France surprised me by placing a mirror in front of the robot I was like ‘oh I’m so cute, I have blue eyes’, not ‘that robot is cute’.”
Eventually, surrogate bodies will be able to impart a greater number of sensations back into our own flesh and bone. For example, Massimo Bergamasco, Antonio Frisoli and their team at the in Pisa, Italy, hope to simulate the sensation of movement when embodying a remote bot – even when you are not moving yourself. All it takes is a little vibration over specific muscles, which stimulate sensory receptors that are then interpreted by the brain as movement in the associated joint. Next year, the team intend to include this illusion in the experience of controlling a robot using thought via fMRI.
Given the technology’s overall trajectory, then, many researchers have begun to explore the looming economic, legal and social impacts. If the history of communication is a guide, telepresence devices will inevitably be used in ways that their makers hadn’t initially intended. “There are a lot of people who have spent a lot of money on this tech and haven’t always thought about the impacts,” says , a lawyer and deputy director of the Centre for Law and the Environment at UCL, who, along with Slater and others, . So what might be the consequences of it becoming easier for everybody to move about remotely?
For a start, telepresence could disrupt labour markets. One plausible scenario for the technology is to allow low-wage foreign workers to be employed for jobs that were impossible until now. After all, it has happened before – more than a decade ago, improved internet speeds and coverage meant nations like India became prime targets for Western companies to outsource online and telecommunication services at lower cost.
Remote-control world
Consider how a retail business like Home Depot or Tesco might use telepresent workers. It could stop employing as many local assistants to do jobs like directing customers to products in-store, or potentially even operating machinery, and hand those tasks to employees overseas instead. “One remote worker could be responsible for 10 stores and 30 robots,” says at the MIT Sloan School of Management, who has also been investigating the impact of telepresence technology. “I’d be very surprised if in 10 years, 10 per cent of that kind of work wasn’t being performed by remote workers. From a technical and economic perspective, it looks pretty inevitable.”
Or take the implications of medicine continuing on its path towards remote procedures. It is bound to trigger legal and regulatory headaches if it spurs a new wave of medical tourism, for example. “What happens when a dentist in Cuba offers cheaper procedures through teleoperation to people in England?” asks Purdy. “Where is the service taking place, and who regulates it?” If something goes wrong during a procedure, or if an unqualified doctor practises remotely, for example, it is unclear which court or medical board would be responsible for investigation or punishment.
Indeed, no country has specific legislation covering the use of telepresence, and most law enforcement would be ill-prepared for the way it would allow anybody to transmit physical actions across international boundaries. Purdy predicts more legal problems when it comes to issues like criminal responsibility. “If I throw a punch in England and it hits someone in another country, is the offence committed here or there, and which country’s law should take precedence?” Or what if a controller accidentally knocked somebody over, for instance, causing them to have a heart attack? Was it the controller’s fault or the manufacturer’s? Could a connection problem or malfunction be blamed? “I’m not sure the police are ready for this sort of thing,” says Purdy. As far as crime goes, the ways in which nations are attempting to harmonise their laws on cybercrime, via international treaties, might offer some precedent. “There’s going to have to be much more collaboration across different jurisdictions,” he adds.
“If I throw a punch and it hits someone in another country, is the offence committed here or there?”
There’s also the issue of deception – already rife online. Although a real-time video screen on a robot would verify identity, not all devices have this feature. Sometimes you can’t tell who is controlling a surrogate. “What happens when a person hacks into this technology and takes over?” asks Purdy. “Who would know that it had happened, and how could you prove that it wasn’t you?”
Indeed, the issue of verifying somebody’s identify is just one of many social interactions that researchers think could be challenged by the new technology. For a start, many people tend to treat robots very differently to humans – often with less respect. The lines are blurred when a human is at the helm.
In one of the first studies to look at in the workforce, Min Kyung Lee at Carnegie Mellon University in Pittsburgh, Pennsylvania, and of Willow Garage, a robotics company in Menlo Park, California, found that local workers and their remote colleagues using telepresence bots fostered closer relationships with each other than when they used phone calls or Skype to communicate. Yet the norms of social etiquette were tested. Some of the office-based people in the study used their colleague’s robot as a footstool, for example. The controllers, meanwhile, reported feeling uncomfortable when their telepresent self’s body space was invaded (91av, 26 March 2011, p 25). These findings suggest that people will have to find ways of interacting via telepresence that avoid upset or offence.
The makers of telepresence technology ultimately aim to fully immerse our senses in a location far from our own. And this may inevitably raise the question of how we anchor ourselves in reality. When we can walk, talk and work in a distant land while our body resides at home, where do we exist at that moment in time? In the world that holds your body, or the one that holds your mind?
I ask Devon a similar question: when he uses VGo, does he feel more like he’s at school or at home? He answers with the matter-of-fact simplicity of a 7-year-old. “Oh yeah, I’m definitely at school,” he says, before running off to brush his teeth.
