The July 2022 issue of IEEE Spectrum is here!

Close bar

Using Smiles (and Frowns) to Teach Robots How to Behave

Japanese researchers are using a wireless headband that detects smiles and frowns to coach robots how to do tasks

2 min read
Using Smiles (and Frowns) to Teach Robots How to Behave

Naughty robots can now be tamed with this snazzy smile-detecting device from the University of Tsukuba AI Lab. Anna Gruebler and her colleagues have developed a wireless headband that captures electromyographic (EMG) signals from the side of the face, detecting when you're smiling with delight or frowning with disapproval.

Unlike cameras with smile-detection algorithms, this device can work in low light, while you're walking around, and when you're not looking into your computer's camera. Part of the charm, the researchers say, comes from the discreet headband design that beats traditional face electrodes and wires.

Last year, Gruebler proposed the device to control avatars on Second Life in a hands-free way, as in the explanation video below. More users would approach her avatar, she says, because it was smiling and looked friendly.

Their current version supports smile and frown detection at a success rate of over 97 percent and has been used to train a Nao humanoid robot in real-time, as shown in this video released at the 2011 IEEE-RAS International Conference on Humanoid Robots, in Bled, Slovenia, last week:

The trainer tries to teach the robot her preference: Give the ball or throw it. Although the Nao starts out slow and hesitant, it speeds up after acquiring experience and feedback from the trainer. Their study compared it to using a manual interface: While users made mistakes using a dial, they never confused smiling and frowning -- a natural, intuitive way to interact with a robot.

The main idea, the researchers say, is that it's similar to how parents teach and encourage babies.

The next step is to apply the device to other real-life situations. If you could train a robot with a smile or frown, what would you have it do?

Angelica Lim is a graduate student at the Okuno and Ogata Speech Media Processing Group at Kyoto University, Japan.

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
LightGreen

“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