This summer, National Geographic Studios presents the giant screen ROBOTS. Narrated by Simon Pegg, who voices the robot guide, RoboThespian, it takes viewers on an exploration of a dozen humanoid robots from around the world.
The film covers the DARPA Robotics Challenge, illustrating state of the art of robotic engineering, and addressing the vast potential and current limitations of this young and fast-moving branch of scientific development.
Its director, Mike Slee, spoke to Blooloop about the film, and about its subject, the fascinating, fast-evolving science of robotics.
According to Slee, the choice of RoboThespian as narrator and commentator was a case of necessity being the mother of invention: they were seeking a thread to bind all the elements – and all the robots – together in a film with a relatively low budget for an IMAX. One very seductive idea was that they should actually start from scratch and make the film about building a robot.
Slee says, “… that seemed like a good idea, but scientifically it’s sort of misleading, because…
"…what we wanted to build, obviously, was a robot that doesn’t exist, which is a robot that can do everything.”
For Slee, and therefore for the viewers, part of the story of the film and of the science of robotics, is the fact that we still don’t have a robot that is really like a human being. We have various robots around the world that can do some of the things that a human can do – sometimes even better – but mainly not as well. But, there’s no one robot that joins them all together.
“That was one of the key things I wanted to get across in the movie without disappointing the science fiction fans, because, of course, everybody loves C3PO and the idea that there is a real android out there that can be the ultimate humanoid robot. Unfortunately there isn’t – but look how fascinating all these attempts are.”
Now in its third incarnation, with six years of continuous development under its robotic belt, RoboThespian is a life-sized, interactive, humanoid robot that was designed for human interaction in a public environment. Slee and Jini Durr, producer and co-writer, came across the charismatic android in a convention centre in Atlanta, and the idea of using him as a host and narrator was born.
For the film, he was voiced by Simon Pegg, known not only for the hugely popular films Shaun of the Dead, Hot Fuzz and The World’s End (among others) – but for his supporting roles in Star Trek and Mission Impossible.
Slee says, “Having RoboThesp as the host seemed like the best way forward. He is an actor robot, so we don’t even analyse him and what he can do…"
Simon Pegg was brilliant. He jumped at the idea straight away, saying ‘This is great – I can play a robot!’
"He knew what we needed, and delivered just superbly.”
“We had 40 minutes – we hardly scratched the surface of what I think will be one of the fastest-growing science and technology areas in the world. In the space of making the film, every single robot that we filmed, by the time we cut it into the movie, had improved. What we filmed, basically, was already out of date – they’d already made a Version X of that humanoid robot, which shows you how quickly the science and technology is moving.”
Slee wanted the film to provide a snapshot of how far robots have come in the last few decades. Paradoxically, this only serves to highlight how very far they have to go.
.jpg)
At the beginning of the process, Slee knew, he says, nothing about robots. He spent six months reading and meeting and going to robot conferences, and as his acquaintance with robots grew, so did his fascination and enthusiasm – something which comes across in the film, an utterly compelling glimpse into the future, that leaves the viewer desperate to learn more about these machines that, while falling short of emulating humanity’s complexity, evoke an involuntarily emotional response.
The film's appeal is far wider and further-reaching than the makers had envisaged.
The film has been conceived as an entertaining but educational schools film: NatGeo is supporting it with teacher packs, and the 8-9 year-olds at whom it is aimed will be spellbound. But, it transpired on initial screenings, its appeal is far wider and further-reaching than the makers had envisaged: it opens a door for everyone on something quite incredible – the crossover between reality and science fiction.
In the process of demonstrating how incredibly difficult it is for a robot to be made to function like a human, the viewer is brought face to face with the intricacy of humanity: as Shakespeare put it, ‘What a piece of work is Man…’
Slee says, “What this film taught us, and in fact what robotics is teaching the robot scientists, is how unbelievably complex the human body and mind are… This wasn’t in the film, but, for instance, you use many hundreds of muscles and sensors just to stand still. You’re constantly adjusting. For a robot to maintain a posture against the wind or against any movement that there might be, has got to involve a very sophisticated set of monitoring and reactions.”
“In the end… this film is wonderful because it has the ability to sing the praises of the current stage of robotics while at the same time going ‘You humans are amazing!’”
The film’s featured robots include, among many others, ‘ASIMO’ (below right), Honda’s famous humanoid which can jump and run up to 5 mph; ‘ATLAS’, a 6-foot, 330-pound search-and-rescue robot which navigates rough terrain (above); ‘HERB THE BUTLER’, a household help which clears the table and does the dishes; ‘ROBONAUT’: NASA’s space handyman that helps .jpg)
astronauts aboard the International Space Station.
At one point, RoboThespian says, ‘We humanoids can go as far as you humans can take us.’
But might they, one day, go further?
To the post-Asimov generations there is inevitably a tiny, exhilarating suspicion that as each of the obstacles in the way of a true human replica is removed, we are paving the way towards our own destruction.
Slee says, “The robot scientists are very wary of saying it’s a frightening prospect, the future of robots. But, of course, you can imagine a frightening future very easily, with robots ruling the world.”
Underpinning the film is the inexorable tide of science, leading us – possibly – closer and closer to the stuff of science fiction – and to something that has caused speculation among real-world scientists: the Singularity. As Slee says, “It’s the point, basically, where you can’t tell the difference between the human and the machine. The theory with the Singularity varies – how far away we are from it. Some say it’s only 30-50 years away; some say it’s always going to be unattainable. You’ll get closer and closer to creating a humanoid robot that you can’t tell is a machine, but there will be always a point at which it fails to be believable. It’ll be interesting when it gets closer, though."
"Looking in the eyes of a humanoid robot is a very strange feeling.”
Robots are universally fascinating, partly because of the identity that science fiction has given them – suddenly, robots such as CHIMP, the humanoid that can ‘see’ by using laser light and sending out pulses which bounce back, and the crawling, walking HRP-2, designed to study locomotion, are being created that are eerily like the ones that have been featured in books and movies for decades. Another factor is the fact that we can’t help responding on an emotional level to a humanoid machine. On an intellectual level, we know it’s just a machine, programmed with a number of responses. Viscerally, however, there’s recognition of sorts.
Slee says: “That is a very curious side of robotics, and it’s actually why we chose the humanoids.
You’re designed to interact with other humans, and therefore you interact with robots when they look vaguely human. It is quite strange, staring into the face of what is just a programmed and programmable machine, but it stares back at you, and you get that slightly unnerving feeling.
The androids are the freakiest, actually – the ones that they’ve tried to make look human. The closer they get, the weirder it feels.”
Slee feels the universal fascination with humanoid robots has a lot to do with the idea of making versions of ourselves:
“It’s the Frankenstein thing, almost. In fact, they are looking at organic materials for data processing in robots, and they are looking at ways of mimicking synapses and the processes in the human brain using organic computers, but it will always be something that we sort of made from scratch – that we built on the workbench.
But, for it to stand up and look at you and talk to you or ask you questions…that’s when it does get a little unnerving.”
Many people find the androids designed to resemble human children particularly eerie – they have the quality of animated dolls.
Slee makes an interesting point about this.
“There’s a cultural difference between the East and the West. In the West, the general feeling about the child-like androids is that creating machines that look and act like children is strange –it has that spooky side to it. Whereas in Japan and Korea, where the real massive robot industry is, their argument – and it’s a very valid one – is that the science of robot humanoids is in its infancy, so therefore they will make these robots look like infants. Then people understand how much they don’t know, or how they might fall over: the logic is pretty sound.”
“To the Western eye and Western sensibilities, it seems strange to be making child machines."
"We make Terminators, I guess…”
He goes on:
“At the moment, each country has its own kind of look and its own way of approaching the subject. In Japan, for example, they’re looking at humanoid robots working with autism with disabilities where they need to have continuous stimulation and for repetitive tasks to be performed. But, rather than it being just a technical thing, putting a face or giving a personality to the robot allows people to engage emotionally with it. It’s pseudo-emotion: it’s only one way, but it doesn’t really matter.”
Slee feels there is a great potential for the further development of therapeutic robots.
“I’m not so keen on robot soldiers. I guess if it’s just one robot soldier fighting another robot soldier that’s OK, but not robots against people. Obviously there’s a big military programme developing robot soldiers, which we didn’t touch on: you’d need a much bigger film to touch on the implications of that. That’s why I did the rescue story. Here’s a machine that needs, in this instance, to go into a nuclear plant – a highly toxic or radio-active environment – and turn off a valve. It’s an example of a machine that had to be designed like a human being, because the nuclear plant is designed to be operated by humans… because the world is designed for human beings."
Another example of this is the robots that can operate in space – without, of course, needing space-suits.
“I think as we move off this planet and start colonising other parts of the solar system robots will have a big impact, because they’re ideal.”
Slee says: “There’s Robonaut that NASA has built, and they’ve also got a prototype called Valkyrie, which is a full humanoid robot designed to be astronauts’ support.”
"Robots are not fluffy like us."
The film hints at where we may go in the future, and what we might be able to do, and because the machines we’re making to help with human tasks take this necessarily humanoid form, it makes it all the more tangible.
Slee says: “If we did do another film, it would look at the idea of going off earth; off this planet – and what the robots will be able to do for us. And also deep down inside the planet – going places that, again, we can’t go.
The key phrase up until now has always been ‘dull, dirty, dangerous’ – that’s what robots are for. But then we’re starting to get smarter and more sophisticated robots, and maybe one day they’ll tell us they don’t just want to do the dull, dirty and dangerous: maybe they want to paint or compose an opera.”
He adds, “One of the big take-aways that I had after talking to all the scientists was that – they said the reason human beings are so brilliant is because they compromise. There are better ways to all of the things that we do – but nothing can do all of the things that we can do, because we are able to make this compromise. And, they were trying to work out how to build a robot that could emulate that. How do you build a machine that can compromise? Every step you take, you’re making loads of judgements, and you’re managing your weight and your balance in an extremely sophisticated way, and the compromise is, ‘as long as I don’t fall over, I’m OK.’
A robot has precision. I wanted Asimo, the (Honda) robot to go up and down the stairs and they said, well, we will bring the stairs. I said, no, I’d like him to go up some nice-looking stairs, maybe at an old university, and they went: No, no, no. He will only walk up the stairs if they’re our stairs in our studio. Which goes to show the limitations. If you ever asked Asimo to run up the rickety stairs of a mediaeval castle, he couldn’t. No chance.
That kind of sums up the state of robotics. The human makes compromises, all of the time, whereas the machine, really, it’s digital – it’s binary; black and white. They’re not fluffy, like us.
What I’d like to do is to go back to our original concept and bring together the best at the time in the ten main disciplines, and say to them, what happens if we combine all of your thinking into one machine?"
Images © National Geographic Studios except Rollin Justin © DLR-German Aerospace Center, HRP-2 © Kawada Industries, ASIMO © Honda America, Herb the Butler © Carnegie Mellon University and Simon Pegg © David Willis.
Film credits: Music is by Mark Korven. Editors are Harry Miller, A.C.E. and Jim Ruxin, A.C.E. Director of Photography is Sean Macleod Phillips, A.S.C. Executive Producers are Brooke Runnette and Lisa Truitt; it is produced by Jini Dürr, and written by Richard Panek, Mike Slee and Jini Dürr.
