|
listen to this article  |
Maza Matarić is a 2024-2025 ACM Athena Instructor. | Source: ACM Awards
Maya Mataric has been researching robotics in academia for nearly 30 years, pioneering the field of socially assistive robotics and conducting basic research on multi-robot coordination and human-robot interaction.
Matarić’s early work was the first to demonstrate that action-based systems (BBSs) could be given representations and had the expressive power to plan and learn. Her lab’s Toto system was the first BBS to learn online maps and optimize actions. She later pioneered distributed algorithms for scalable control of robot teams and swarms, allowing robot teams to collaborate on tasks.
Now Matarich Chan Soon-Shiong is Distinguished Professor of Computer Science at the University of Southern California and Chief Scientist at Google DeepMind. Much of her recent research remains in the field of socially assistive robotics (SAR), where she seeks to create engaging and trustworthy machines that foster long-term human-robot bonds.
Matarić said the robot could be programmed to help children with autism communicate and socialize, encourage stroke patients to perform rehabilitation exercises, and more. SAR could also help Alzheimer’s patients recognize and enjoy their favorite songs, motivate older people to stay physically active, and more.
With this work she The ACM Athena Lecturer Award celebrates female researchers who have made fundamental contributions to computer science. ACM every year Honoring Distinguished Female Computer Scientists as Athena Instructors
To learn more about her work and her hopes for the robotics industry as a whole, robot report We spoke with Mataric about SAR, recent industry breakthroughs, and how the industry can learn from academia.
What is a social assistance robot?
“Unlike all other robotic technologies where robots perform physical tasks, here we are using robots as social companions for people with specific health, education, learning or rehabilitation needs,” Matarić said. Robot Report.
SAR aims to play an important role in society, and these unique robots will become more necessary in the future. Since January 2020, 400,000 nursing home and assisted living workers have quit, and 10,000 people are turning 65 every day. Robots offer a practical solution to this type of labor shortage, Matarić said.
“Never in my work has I said that robots will be better than humans,” she said. “If we lived in a world where people were properly trained and properly compensated for caring for others, we wouldn’t need robots.”
Mataric also emphasized that SAR is designed to be human-centric. “Social assistance robots are a form of robots that don’t lose their humanity, because they help humans do more. Unlike automation, which automates tasks, social assistance robots empower people to do things for themselves,” she said.
It may seem like people have a hard time being vulnerable with social robots, but Matarić said he is more interested in how roboticists can leverage the trust people place in them.
“It’s actually not that difficult to get people to trust machines. And that could be potentially dangerous, right?” She said. “But people will always trust machines. “We are constantly feeding data to machines.”
“The next question is, ‘Can this help me not only gain your trust, but actually maintain your trust and add real value to your life?’” he said. Mataric. “And by value, I don’t mean money, I mean ‘Can this actually help me improve my health, well-being, education, etc.?'”
Matarić says generative AI is a key innovation in SAR.
One of the biggest recent innovations in socially assistive robotics is, unsurprisingly, generative AI.
“There is no doubt that there is enormous potential,” Matarić said. “There is also a huge amount of investment being made, which means great things can be done in both research and industry.”
“Two years ago, when our robots were talking to people, and everyone’s robots were talking to people, they were using conversation trees,” she said. “Everything the robot said had to be scripted in advance.”
“Now we can use language models to enable robots to have truly natural conversations with users.” Mataric said. “It’s an amazing, enabling, foundational feature that gives us the ability to naturally engage with and support our users, so now we can combine it with all the other features we’ve developed to truly understand what each user needs. ..and we can provide support to achieve that goal.”
Register now.
Academia and industry must work together to make progress.
While doing matariić spent most of his career in academia. She is passionate about how industry can learn from academia and the possibilities for collaboration between the two fields. She said she believes her academics’ slower approach could benefit organizations seeking to deploy robots commercially.
“First, we don’t rush. If we were to enter into competition, there would be a possibility that we would compromise some values, so we don’t rush.” Matariić said. “Second, we talk to our user community a lot, much more than most people do. We pride ourselves on collecting data and conducting research almost exclusively outside of the lab, in nursing homes, people’s homes, families with autistic children, nursing homes, stroke rehabilitation centers. Not in convenient places, but in places that are real.”
But she acknowledged that there is a disconnect between academia and the robotics industry today. “I always think the commercial incentives are the problem,” she said. “They often don’t fit the population we’re trying to serve. So what’s often developed is something that has a lucrative market.”
“We worked with children with cerebral palsy and it was really interesting,” Matari said.ić said. “There has been a lot of transfer from stroke patients and other populations we have worked with, but that is not a population that gets enough attention because it is not a market. So if the ecosystem of innovation is driven solely by market forces, there will be problems. “
Nonetheless, she believes that the students she works with are likely to be the future of the industry, and that they are learning important lessons about robotics in the real world while working in the lab. She places a special emphasis on testing robots with real people in real-world scenarios.
“By actually being in a physical location, you learn a lot about what people want and need,” she said. “What actually happens in the real world is very humbling. “It makes you slow down because it’s hard.”
“But I am very, very grateful for my students and their values, and I am very proud of them because they are willing to go out and collect data that is hard to get and really learn and be humble.” Matari said.Seed.
