Mood-Detecting Sensor Could Help Machines Respond to Emotions

Smart homes could play music to fit your mood

2 min read
mood-detecting sensor
Photo: Jason Dorfman/MIT CSAIL

Emotions can be detected remotely using a device that emits wireless signals to help it measure heartbeat and breathing, say researchers at MIT's Computer Science and Artificial Intelligence Laboratory.

The new device, named “EQ-Radio,” is 87 percent accurate at detecting whether a person is excited, happy, angry or sad—all without on-body sensors or facial-recognition software.

“We picture EQ-Radio being used in entertainment, consumer behavior, and healthcare,” says the study’s lead researcher, Mingmin Zhao. “For example,” says Zhao, a graduate student, “smart homes could use information about your emotions to adjust the music or even suggest that you get some fresh air if you've been sad for a few days.” Zhao adds that remote emotion monitoring could eventually be used to diagnose or track conditions like depression and anxiety.”

Zhao and study co-authors Dina Katabi and Fadel Adib will present their work in October at the Association of Computing Machinery's International Conference on Mobile Computing and Networking (MobiCom) in New York.

Existing technologies for detecting emotion require on-body sensors or audiovisual cues, but there are downsides to both these methods. For example, on-body sensors such as chest bands and ECG monitors are inconvenient to wear and become less accurate if they shift position over time. Systems that rely on audiovisual cues require people to face cameras, and can miss subtle facial expressions.

Instead, EQ-Radio emits radio signals that reflect off a person's body and back to the device. Its algorithms can detect individual heartbeats from these radio echoes with an accuracy comparable to on-body ECG monitors.

In experiments, the researchers then had EQ-Radio scan 30 volunteers, each seated about one meter away from the gadget. They were listening to music, looking at photos, or watching videos to help them recall memories that made them feel either happy, excited, sad, angry, or neutral. The system collected more than 130,000 heartbeats.

The device examined small differences in heartbeat length that indicated these emotions. It accurately pegged the person’s state of mind  72.3 percent of the time, even if it had not previously analyzed him or her.

“This work demonstrates that wireless technologies can capture meaningful information about human behavior that is invisible to the naked eye,” Zhao says.

Though EQ-Radio achieved a level of accuracy comparable to methods using on-body sensors, Microsoft's vision-based Emotion API, which focuses on facial expressions, was right only about 40 percent of the time. Emotion API figured out when a person is in a neutral emotional state about as well as EQ-Radio did. This is likely because noting the absence of emotion on someone’s face is usually easier to do than labeling whether it’s a happy, excited, sad, or angry face. The MIT team, which tested both systems, noted that Emotion API was more accurate with positive emotions than negative ones, likely because positive emotions typically have more visible features such as smiling, whereas negative emotions are visually closer to a neutral state.

“We view this work as the next step in trying to develop computers that can understand us better at an emotional level and potentially interact with us similarly to how we interact with other human beings,” Zhao says.

The scientists noted that the implications of this work may extend beyond emotion recognition. For instance, wireless heartbeat analysis could be used for non-invasive health monitoring and diagnosis of conditions such as arrhythmia.

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

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

"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.

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