Maja Matarić Pioneered Socially Assistive Robotics

The robots may have conversations, play video games, and reply to feelings.

In the present day the IEEE Fellow is a professor at USC. She research how robots can assist college students with nervousness and despair bear cognitive behavioral remedy. CBT focuses on altering an individual’s detrimental thought patterns, behaviors, and emotional responses.

For her work, she acquired a 2025 Robotics Medal from MassRobotics, which acknowledges feminine researchers advancing robotics. The Boston-based nonprofit offers robotics startups with a workspace, prototyping services, mentorship, and networking alternatives.

When receiving the award on the ceremony in Boston, Matarić was overcome with pleasure, she says.

“I’ve been very lucky to be honored with a number of awards, which I’m grateful for. However there was one thing very particular about getting the MassRobotics medal, as a result of I knew at the least half the individuals within the room,” she says. “Everybody was simply smiling, and there was an incredible sense of affection.”

Seeing herself as an engineer

Matarić grew up in Belgrade, Serbia. Her father was an engineer, and her mom was a author. After her father died when she was 16, Matarić and her mom moved to the United States.

She credit her father for igniting her curiosity in engineering, and her uncle who labored as an aerospace engineer for introducing her to laptop science.

Matarić says she didn’t contemplate herself an engineer till she joined USC’s college, since she all the time had labored in laptop science.

“Looking back, I’ve all the time been an engineer,” Matarić says. “However I didn’t set out particularly considering of myself as one—which is simply one of many many issues I wish to convey to younger individuals: You don’t all the time should know precisely every part prematurely.”

Maja Matarić and her lab are exploring how socially assistive robots can assist enhance the communication abilities of kids with autism spectrum disorder. National Science Foundation Information

Whereas pursuing her bachelor’s diploma in laptop science on the University of Kansas in Lawrence, she was launched to industrial robotics by means of a textbook. After incomes her diploma in 1987, she had a possibility to proceed her training as a graduate scholar at MIT’s AI Lab (now the Computer Science and Artificial Intelligence Lab). Throughout her first yr, she explored the totally different analysis initiatives being performed by college members, she mentioned in a 2010 oral history performed by the IEEE History Center. She met IEEE Life Fellow Rodney Brooks, who was engaged on novel reactive and behavior-based robotic methods. His work so excited her that she joined his lab and performed her grasp’s thesis underneath his tutelage.

Impressed by the best way animals use landmarks to navigate, Matarić developed Toto, the primary navigating behavior-based robotic. Toto used distributed fashions to map the AI Lab constructing the place Matarić labored and plan its path to totally different rooms. Toto used sonar to detect partitions, doorways, and furnishings, in response to Matarić’s paper, “The Robotics Primer.”

After incomes her grasp’s diploma in AI and robotics in 1990, she continued to work underneath Brooks as a doctoral scholar, pioneering distributed algorithms that allowed a workforce of as much as 20 robots to execute complicated duties in tandem, together with trying to find objects and exploring their atmosphere.

Matarić earned her Ph.D. in AI and robotics in 1994 and joined Brandeis University, in Waltham, Mass., as an assistant professor of laptop science. There she based the Interplay Lab, the place she developed autonomous robots that work collectively to perform duties.

Three years later, she relocated to California and joined USC’s Viterbi School of Engineering as an assistant professor in laptop science and neuroscience.

In 2002 she helped to discovered the Middle for Robotics and Embedded Systems (now the Robotics and Autonomous Systems Center). The RASC focuses on analysis into human-centric and scalable robotic methods and promotes interdisciplinary partnerships throughout USC.

Matarić’s shift in her analysis got here after she gave delivery to her first little one in 1998. When her daughter was a bit older and requested Matarić why she labored with robots, she needed to have the ability to “say one thing higher than ‘I publish lots of analysis papers,’ or ‘it’s well-recognized,’” she says.

“In academia, you may be in a management function and nonetheless do analysis. It’s an exquisite and necessary alternative that lets lecturers be on prime of our subject and in addition practice the subsequent technology of scholars and assist the subsequent technology of school colleagues.”

“Children don’t contemplate these good solutions, they usually’re most likely proper,” she says. “This made me understand I used to be ready to do one thing totally different. And I actually needed the reply to my daughter’s future query to be, ‘Mommy’s robots assist individuals.’”

Matarić and her doctoral scholar David Feil-Seifer introduced a paper defining socially assistive robotics on the 2005 International Conference on Rehabilitation Robotics. It was the one paper that talked about serving to individuals full duties and study abilities by talking with them slightly than by performing bodily jobs, she says.

Feil-Seifer is now a professor of laptop science and engineering on the University of Nevada in Reno.

On the identical time, she based the Interaction Lab at USC and made its focus creating robots that present social, slightly than bodily, help.

“At this level in my profession journey, I’ve matured to a spot the place I don’t need to just do curiosity-driven analysis alone,” she says. “Loads of what my workforce and I do immediately continues to be pushed by curiosity, however it’s answering the query: ‘How can we assist somebody stay a greater life?’”

In 2006 she was promoted to full professor and made the senior affiliate dean for analysis in USC’s Viterbi College of Engineering. In 2012 she grew to become vice dean for analysis.

“In academia, you may be in a management function and nonetheless do analysis,” she says. “It’s an exquisite and necessary alternative that lets lecturers be on prime of our subject and in addition practice the subsequent technology of scholars and assist the subsequent technology of school colleagues.”

Analysis in socially assistive robotics

One of many longest analysis initiatives Matarić has led at her Interplay Lab is exploring how socially assistive robots can assist enhance the communication abilities of kids with autism spectrum disorder. ASD is a lifelong neurological situation that impacts the best way individuals work together with others, and the best way they study. Youngsters with ASD typically wrestle with social behaviors reminiscent of studying nonverbal cues, enjoying with others, and making eye contact.

Matarić and her workforce developed a robotic, Bandit, that may play video games with a toddler and provides the teen phrases of affirmation. Bandit is 56 centimeters tall and has a humanlike head, torso, and arms. Its head can pan and tilt. The robotic makes use of two FireWire cameras as its eyes, and it has a movable mouth and eyebrows, permitting it to exhibit quite a lot of facial expressions, in response to the IEEE Spectrum’s robots guide. Its torso is hooked up to a wheeled base.

The research confirmed that when interacting with Bandit, kids with ASD exhibited social behaviors that have been out of the peculiar for them, reminiscent of initiating play and imitating the robotic.

Matarić and her workforce additionally studied how the robotic may function a social and cognitive help for elderly individuals and stroke sufferers. Bandit was programmed to instruct and inspire customers to carry out day by day motion workout routines reminiscent of seated aerobics.

Maja Matarić and doctoral scholar Amy O’Connell testing Blossom, which is getting used to check the way it can help college students with nervousness or despair.College of Southern California

Through the years, Matarić’s lab developed different robots together with Kiwi and Blossom. Kiwi, which appeared like an owl, helped kids with ASD study social and cognitive abilities, helped inspire aged individuals residing alone to be extra bodily energetic, and mediated discussions amongst members of the family. Blossom, initially developed at Cornell, was tailored by the Interplay Lab to make it inexpensive and personalizable for people. The robotic is getting used to check the way it can help college students with nervousness or despair to apply cognitive behavioral remedy.

Matarić’s line of analysis started when she discovered that giant language mannequin (LLM) chatbots have been being promoted to assist individuals with mental health struggles, she mentioned in an episode of the AMA Medical News podcast.

“It’s usually not straightforward to get [an appointment with a] therapist, or there may not be insurance coverage protection,” she mentioned. “These, mixed with the charges of tension and despair, created an actual want.”

That made the chatbot concept interesting, she says, however she was to see in the event that they have been efficient in contrast with a pleasant robotic reminiscent of Blossom.

Matarić and her workforce used the identical LLMs to energy CBT apply with a chatbot and with Blossom. They ran a two-week research within the USC dorms, the place college students have been randomly assigned to finish CBT workout routines day by day with both a chatbot or the robotic. Contributors stuffed out a scientific evaluation to measure their psychiatric misery earlier than and after every session.

The research confirmed that college students who interacted with the robotic skilled a major lower of their psychological state, Matarić mentioned within the podcast, and college students who interacted with the chatbot didn’t.

“Becoming a member of an [IEEE] society has an influence, and it may be private. That’s why I like to recommend my college students be part of the group—as a result of it’s necessary to get on the market and get linked.”

She and her workforce additionally reviewed transcripts of conversations between the scholars and the robotic to judge how properly the LLM responded to the contributors. They discovered the robotic was simpler than the chatbot, although each have been utilizing the identical mannequin.

Based mostly on these findings, in 2024 Matarić acquired a grant from the U.S. National Institute of Mental Health to conduct a six-week scientific trial to discover how efficient a socially assistive robotic may very well be at delivering CBT apply. The trial, at the moment underway, additionally is predicted to check how Blossom may be personalised to adapt to every person’s preferences and progress, together with the best way the robotic strikes, which workout routines it recommends, and what suggestions it offers.

Through the trial, the 120 college students collaborating are carrying Fitbits to check their physiologic responses. The contributors fill out a scientific evaluation to measure their psychiatric misery earlier than and after every session.

Knowledge together with the contributors’ emotions of regarding the robotic, intrinsic motivation, engagement, and adherence might be assessed by the analysis workforce, Matarić says.

She says she’s pleased with the graduate college students engaged on this undertaking, and seeing them develop as engineers is likely one of the most rewarding components of working in academia.

“Engineers usually don’t anticipate having to work with human research contributors and needing to know psychology along with the hardcore engineering,” she says. “So the scholars who select to do that analysis are simply great, caring individuals.”

Discovering a neighborhood at IEEE

Matarić joined IEEE as a graduate scholar in 1992, the yr she printed her first paper in IEEE Transactions on Robotics and Automation. The paper, “Integration of Representation Into Goal-Driven Behavior-Based Robots,” described her work on Toto.

As a member of the IEEE Robotics and Automation Society, she says she has gained a neighborhood of like-minded individuals. She enjoys attending conferences together with the IEEE International Conference on Robotics and Automation, the IEEE/RSJ International Conference on Intelligent Robots and Systems, and the ACM/IEEE International Conference on Human-Robot Interaction, which is closest to her subject of analysis.

Matarić credit IEEE Life Fellow George Bekey, the founding editor in chief of the IEEE Transactions on Robotics, for recruiting her for the USC engineering college place. He knew of her work by means of her graduate advisor Brooks, who printed a paper within the journal that launched reactive management and the subsumption structure, which grew to become the muse of a brand new approach to management robots. It’s his most cited paper. Bekey, who was editor in chief on the time, helped information Brooks by means of the difficult assessment course of. Matarić joined Brooks’s lab at MIT two years after its publication, and her work on Toto constructed on that basis.

“Becoming a member of a society has an influence, and it may be private,” she says. “That’s why I like to recommend my college students be part of the group—as a result of it’s necessary to get on the market and get linked.”