Feed Your Friends With Autonomous Chest-Mounted Robot Arms

Enjoy your food in a new, weird way with a robotic third arm with face control

4 min read
Arm-a-dine eating robot
Photo: Exertion Games Lab

Eating food is an experience that tends to be about taste and texture and how the food looks and smells. Our focus goes from what’s going on on the plate to what’s going on in our mouths, without a lot of concern about what happens in between. Eating as a process doesn’t get all that much attention; we tend to treat it as just a chore involving utensils. Which is fine, but are we missing out somehow? The Exertion Games Lab at RMIT University in Australia thinks that the answer to that is yes, and they’re using chest-mounted social feeding robots to prove it.

Arm-A-Dine is our design exploration of a novel two-person playful eating system that focuses on a shared feeding experience. In this experience, all three arms (the person’s own two arms and the “third” arm, the robotic arm) are used for feeding oneself and the other person. The robotic arm (third arm) is attached to the body via a vest. We playfully subverted the functioning of the robotic arm so that its final movements (once it has picked up the food), i.e. whether to feed the wearer or the partner, are guided by the facial expressions of the dining partner. 

If you can’t quite tell how this works from the video, the robot third arms are only partially autonomous. They move up and down by themselves, but there’s no integrated sensing, so they depend on their human hosts to position them over tasty morsels of food pre-grab. The arm also doesn’t take the food all the way to your mouth, instead stopping about 10 centimeters short “for safety reasons.” Depending on what kind of expression your partner has, your arm arm will either give the food to you or to them, meaning that if you want to eat whatever thing you just picked up, you’ll need to somehow convince them to make a frowny face.

Mapping of the partner’s “more positive” facial expression to the feeding of food to the partner (via the wearer’s third arm) we hoped would elicit joy, laughter, and a sense of sharing based on the knowledge of feeding one another that is associated with positive emotions, however, this could also result in the perception of a loss of agency over what one eats. Through to-and-fro ambiguous movements of the third arm in the air (when sensing a “neutral” facial expression of the dining partner), it gave an opportunity to the diners to express their reactions more vividly, as we know that facial expressions become a key element to engage with a partner while eating.

This is all very speculative, so we’re delighted to report that the researchers conducted a user study. Participants were outfitted with their robotic chest arms, and stood in front of a table covered with finger foods including strawberries, chocolates, grapes and cheese, plus cookies and chips and crackers. There wasn’t much in the way of direction—the users were told how the arms worked and allowed to experiment with them a bit, but otherwise the only instruction was to “eat casually with your partner.” Most of the findings were anecdotal in nature (as you’d expect from this sort of research), and here are some of our favorites:

Extra Efforts and Time to Get Food in Mouth Felt Rewarding 
“It pushed me to put an extra effort and attention to the eating process. But when I got the food after twisting, turning and slow movement of the robotic arm, I felt rewarded and satisfied”.

Unpredictable Movements of the Arm Were Enjoyed
“Although I would love perfect arm movement each time but it is too boring. If the arm is too perfect, then there is no chance of anything going wrong or something unexpected to happen and so there is no element of surprise. I think unpredictability of the arm’s movement was great and made the eating experience more playful by increasing the conversation time I had with my partner”.

Paying Attention and Altering Facial Expressions
“I was paying attention to [our] facial expressions: depending on what my [partner] picks, I used to change my expressions and try to feed him the food that he doesn’t like [laughs]”.

Time Gap Between Robotic Arm Movements Facilitated Bonding
“We keep meddling with our phone generally when we are eating. In Arm-A-Dine, we used the time when the robotic arm was doing its movement, to chat about each other’s boyfriend. We have been having some trouble lately [laughs]”.

Feeding Each Other Made Participants Nostalgic 
“I enjoyed the experience a lot, especially the bit where my partner was feeding me as it reminded me of my mother feeding me when I was a child.” 

It’s a little bit weird to think about how having reduced bodily control over eating can actually be a good thing, but it does sound like it makes eating much more interactive and fun. The researchers also suggest that Arm-a-Dine encourages savoring by making eating both slower and stranger, thereby emphasizing the taste of the food. My guess would be that there’s a significant novelty component to the overall experience—while it might be fun to try this once or twice, I could imagine it rapidly becoming frustrating, especially if you’re a picky eater or just very hungry. If anyone wants to run with the concept, though, I can think of one perfect situation for a system like this—a fun and memorable first date. Good luck!

For more details on this, plus a suuuper awkward audience experiment, check out this 10-minute presentation from Rohit Khot at the Exertion Games Lab at CHI PLAY.

“Arm-A-Dine: Towards Understanding the Design of Playful Embodied Eating Experiences,” by Yash Dhanpal Mehta, Rohit Ashok Khot, Rakesh Patibanda, and Floyd Mueller from the Exertion Games Lab at RMIT University in Melbourne, Australia, was presented at CHI PLAY 2018.

[ Exertion Games Lab ]

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman

“I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

This article is part of our special report on AI, “The Great AI Reckoning.”

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓Show less