The $10 million ANA Avatar XPRIZE is a global competition focused on the development of a robotic avatar system that transfers an operator’s senses, actions, and presence to a remote location in real time, leading to a more connected world. The registration for the competition closes on Sept. 30, 2019.
Robotics Business Review recently interviewed Jacki Morie, technical advisor for the Avatar XPRIZE, to get her perspective on how robotic avatars could change the world. Morie spent 13 years as a senior research scientist at UCS’s Institute for Creative Technologies (ICT), where she created novel VR tele-health and educational activities. She is also the founder and chief scientist at All These Worlds LLC, and served as a member of the DARPA Information, Science, and Technology think tank.
Robots provide physical form to avatars
Q: The terms avatar and robotics can mean many different things. How do you define the term ‘robotic avatar’?
Morie: The term avatar has been used for quite a number of decades now to mean a digital form of a person that they inhabit and use in different digital situations. It has been beyond our capabilities to have avatars that have a physicality to them – until recently.
Now that robotics has gotten more sophisticated, and we’re able to do more work with humanoid types of robots, as opposed to more industrial types of robots, we can finally look at the term avatar as something that not only is digital, but also that has a physical aspect to it. By having a physical construct that people could actually teleport into or avatar into or inhabit in some way, the use of that term suddenly takes on a much broader, much greater meaning.
Q: How far away are we from having robotic avatars?
Morie: We can do it in general terms right now, but what we need is the sophistication and the subtlety to really make someone who is using a physical robotic avatar feel like they are really there. The challenge is not just putting your sense and presence into that physical robot, but also having what the robot is experiencing, touching, smelling, the wind in its face; how does that get back to the person who is getting into that avatar, the operator?
Those are the challenges today – and there are huge gaps – how do we take the physical experience the robot is getting and take that back into the person who is in a distant location.
Business cases for avatars
Q: One side of this discussion revolves around the technology, but then there’s the business side. Is there a business case for robotic avatars?
Morie: Certainly for elder care, being able to pop into an avatar in an assisted living facility or your grandmother’s house and say goodnight and give her a little hug, or being able to move people from a bed to a wheelchair [through] an assistive avatar. There are lots of different situations where companionship, that connection, that physical being there with someone else is going to be important.
Beyond that, if most houses have an avatar like they have a vacuum cleaner or a Roomba robot, the avatar could be a general-purpose robot that can make person-to-person connections. For example, an expert can “avatar” into the robot and do repair work on your dishwasher, or check your electrical system to see if there’s a problem. Whatever it might be, instead of traveling to your house, experts could teleport into the physical avatar and perform things that way.
Q: So in the scenario you’re describing the robotic avatar doesn’t necessarily have to look like the actual plumber, the robot has a more generic physical appearance?
Morie: That’s one possible future and one possible kind of robotics, but if we extrapolate 50 to 100 years out, the robot could have some sort of scanning mechanism and could adjust some type of synthetic skin to look like that person. If you’re tucking in your grandmother every night, you would want it to look like you, but for more practical kinds of functions, you don’t necessarily need it to.
Q: What are some other use cases?
Morie: There’s a whole range of possible business cases. Engineers or doctors being able to avatar into a remote village where people need their expertise. Imagine if you have been a mountain climber in your youth, and now you’re older and you can’t do that anymore. But you can get into that able-bodied mountain climber avatar that allows you to experience climbing the mountain the way you did before. Or, if someone is totally bedridden, they can get into an avatar and go off on a hike.
This will eventually give us capabilities we might not have had, [such as] going into the ocean, even going into space. In a lot of ways, the fact that this is not a human body allows us to do more than we’ve done before. Imagine if robotic avatars are as ubiquitous as scooters [on] street corners. If people can use them easily, we might see many more uses and much broader acceptance.
Q: Many people are fearful of new technologies like autonomous vehicles. We’ve also noticed several articles discussing whether AI is going to be used for good or evil. What are your thoughts on this?
Morie: Certainly, any new technology is going bring out fears in people who are not familiar with it. I tend to look at a future in which AI is much more for good. In fact, I just came from the AI for Good Global Summit in Geneva that was attended by hundreds of people.
AI is such a broad range of different kinds of programs, functions and applications that it’s really hard to put a label on it that says it’s good or not good, because it really depends on the application. AI is a number of different tools, it’s not just one tool, and I think people don’t quite understand that.
Q: What about fears that robotic avatars could be taken over by a hacker?
Morie: I’m sure when they’re ubiquitous, those things will be an issue, but no more of an issue than they are now for your own computer. It’s the same type of vulnerabilities, and vulnerabilities can be dealt with. It’s not something we can ignore, but I think we certainly know how to deal with it.
The bigger concern is making sure the person you’re interacting with is safe. It’s a bit like those circular saws that can sense moisture content on a finger and stop. We’re going to need the same kind of failsafe mechanisms on these robotic avatars. We might need them to be strong enough to pick up a beam off of somebody who is in a disaster situation, but you don’t want to hug your grandmother with the same kind of strength you’re using to pick up a beam. All these considerations we will have to look at going forward, but we will have decades of experience in balancing out those issues.
Q: How much will a robotic avatar cost?
Morie: You can get humanoid-form robots today for anywhere from $2,000 to $1 million, but if they are mass produced, the cost will come down to $2,000 to $2,500, with the high end at $10,000. But it’s not just the robot, it’s the entire system. That’s where the technology has to be developed.
Today, you’ve got to put on a couple hundred dollars worth of VR gear. Hopefully that gets less intrusive, with maybe just motion detection cameras in your living room that can translate your movements to the avatar, does what it’s expected and has safety measures and ethics built it. It’s not far-fetched that robotic systems coming down to $2,500 is certainly within realm of where we might be in five years.
Q: Are you optimistic about the future of robotic avatars?
Morie: I’m optimistic about everything. I truly believe that we are trying to make technology that makes our lives better and improves our situations as humans on this planet. If more of us believe that, then maybe that’s the way it will turn out.