Apologies to Matthew McConaughey, but the real, breakout star of "Interstellar" was clearly TARS the sarcastic space robot (voiced by Bill Irwin). A former marine companion bot with angular limbs and acerbic wit, director Christopher Nolan somehow made TARS (and his sister robot, CASE) one of the most fully-formed, anthropomorphized robots in film history -- without even giving him a face.
"I wanted a more realistic approach to what a robot would be," Nolan told the Associated Press. "I didn't even call them robots in the script. I referred to them as 'articulated machines' because I wanted my crew and everybody to stop thinking of your standard idea of a robot. I wanted to have a machine in the film that was like a piece of gear -- very tough, very resilient -- that had been designed for whatever purpose best suited it."
This "purpose" included general companionship, maintaining the Endurance spacecraft while astronauts were sleeping, and at one point carrying Anne Hathaway through a tidal wave. No doubt about it, TARS was incredibly useful -- but was he realistic? In short, unfortunately, the answer is a resounding "no." Here's why:
Our Robots Can't Even Understand The Question
"There's nothing close to TARS right now," said Riverside Robotics Society founder Thomas Messerschmidt. "From a robotics stance, maybe the Honda robot, the Asimo robot, would probably be the closest thing. Not from the block-y rectangular aspect, but just from the mobility and the ability to reach out and interact with things around it."
Indeed, the Asimo robot can do things like walk around and pour a glass of water, but a full-on conversation with a robot -- where it can fully comprehend what you're saying, and have the programming required to respond appropriately -- is still lightyears away.
"Movie robots have the ability to understand the subtle meanings of the spoken word, when someone else might be lying, the situation they’re involved with... [in real life] step one is understanding the question," Messerschmidt continued. "If you say 'How's your mother?' and it understands it to be 'How’s your brother' or just doesn’t understand it at all, then you’re not even there. [The next step is] understanding plain English. You’ve got to take these first steps. And we’re getting there."
To further understand what a momentous task this is -- giving a robot a "human" personality like what we saw with TARS -- Messerschmidt said you should recall conversations you've likely had with Siri, or online chat robots.
"Anybody talking to [a chat bot] for more than a minute would know there’s nothing behind it," he said. "It’s just a bunch of [programmed] questions and answers."
Our Robots Can't Rescue People Through Tidal Waves
Additionally, Messerschmidt said that TARS' physical capabilities, like being flung through a black hole in space, is "quantum leaps" away from where we are with robotics right now.
"Getting a mobile robot to cross a room is a tremendous engineering problem," he explained. "You’ve got to identify where everything is in the room, where the walls are, how big each thing is, and you’ve got to use sensors to calculate these things. If things start moving around, you have to re-calculate. If the cat crosses in front of you, 'Uh-oh!' That’s where these [robotics] schools are at. For the longest time, the hardest thing a robot could do was see a chair leg."
NASA's Robonaut 2 can move around a space station as of fall 2014, and even complete some maintenance tasks, but it certainly can't pick up humans, navigate new worlds, or offer real companionship like TARS can.
"The Robotnaut -- it’s not much of a companion," Messerschmidt explained. "It does a few maintenance things. It’s more experimental than anything."
Our Robots Need A Lot More Money
As seemingly unexciting as our current state of robotics might be, Messerschmidt said that the solution to a more TARS-like state of affairs might not be as far off as we might think -- as long as somebody decides to spend the money.
"Say this movie took place 200 years from now -- I’d give it a chance," he said. "100 years ahead? Maybe. 50 years ahead? I’m a little hesitant to say TARS and CASE would be at that ability 50 years from now. It’s just not there. There’s too much to do, and there’s not enough money being put into it to see that that kind of engineering is going to come forward. It’s all about the money, you know?"
Basically, if you go into cryo-sleep and set your alarm for 200 years from now, TARS might just be there to wake you up with a joke.