Squirming towards bio-inspired artificial intelligence | MIT News

Juncal Arbelaiz Mugica hails from Spain, where octopus is a common menu item. However, Arbelaiz appreciates octopuses and similar creatures in a different way, with his research on the theory of soft robotics.

More than half of an octopus’ nerves are distributed among its eight arms, each of which has some degree of autonomy. This distributed information sensing and processing system has intrigued Arbelaiz, who is investigating how to design decentralized intelligence for man-made systems with integrated sensing and computation. At MIT, Arbelaiz is an applied mathematics student who has been working on the fundamentals of optimal distributed control and estimation for the past few weeks before completing her doctorate this fall.

She draws inspiration from the biological intelligence of invertebrates such as octopus and jellyfish, with the ultimate goal of designing new control strategies for flexible “soft” robots that could be used in tight or delicate environments, such as a surgical tool or for research and research. -rescue missions.

“The flexibility of soft robots allows them to dynamically adapt to different environments. Think of worms, snakes or jellyfish and compare their movement and adaptive abilities to those of vertebrate animals,” says Arbelaiz. an interesting expression of embodied intelligence – the absence of a rigid skeleton gives advantages to certain applications and helps deal with uncertainty more effectively in the real world. But this extra sweetness also brings new theoretical challenges to the system.

In the biological world, the “controller” is usually associated with the brain and the central nervous system – it creates motor commands for muscles to perform movement. Jellyfish and some other soft organisms do not have a centralized nerve center, or brain. Inspired by this observation, she is now working on a theory that software robotic systems could be controlled using decentralized sharing of sensory information.

“When sensing and actuation are distributed throughout the robot body and on-board computational capabilities are limited, it can be difficult to implement centralized intelligence,” she says. “So we need these kinds of decentralized patterns that, despite only sharing sensory information locally, ensure the desired global behavior. Some biological systems, like the jellyfish, are fine examples of decentralized control architectures – locomotion is achieved in the absence of a (centralized) brain. It’s fascinating compared to what we can achieve with man-made machines.

A smooth transition to MIT

Her graduate studies at the University of Navarre in San Sebastián led her to work with Professor John Bush of MIT in fluid dynamics. In 2015 he invited Arbelaiz to MIT as a visiting student to study droplet interactions. This led to their 2018 article in physical examination fluids, and his pursuit of a doctorate at MIT.

In 2018, his doctoral research was transferred to the Center for Interdisciplinary Sociotechnical Systems Research (SSRC) and is now advised by Ali Jadbabaie, JR East Professor of Engineering and Head of the Department of Civil and Environmental Engineering; and Associate Dean of the School of Engineering Anette “Peko” Hosoi, who is the Neil and Jane Pappalardo Professor of Mechanical Engineering as well as Professor of Applied Mathematics. Arbelaiz also works regularly with Bassam Bamieh, associate director of the Center for Control, Dynamical Systems, and Computation at the University of California, Santa Barbara. She says working with this team of advisors gives her the freedom to explore the multidisciplinary research projects that have attracted her over the past five years.

For example, she uses systems-theoretical approaches to design new optimal controllers and estimators for spatio-temporal dynamic systems, and to gain a fundamental understanding of the sensory feedback communication topologies needed to optimally control these systems. For soft-robotic applications, this amounts to prioritizing which sensory measures are important to best trigger each of the “muscles” of this robot. Has robot performance degraded when each actuator only has access to the closest sensory measurements? His research characterizes such a trade-off between closed-loop performance, uncertainty, and complexity in spatially distributed systems.

“I am determined to bridge the gap between machine autonomy, systems theory and biological intelligence,” she says.

Next chapter

A two-year Schmidt Science Fellowship, which funds young researchers to pursue postdoctoral studies in a different area of ​​their graduate studies, will allow Arbelaiz to further explore the intersection of biological intelligence and human intelligence. artificial after graduation.

She plans to spend her postdoctoral time at Princeton University with Professor Naomi Leonard, and work with researchers in systems biology, computer science and robotics, to explore the reliability and robustness of biological and artificial assemblies. Specifically, she wants to learn how biological systems adapt effectively to different environments so that she can apply this knowledge to human-made systems, such as autonomous machines, whose vulnerability to noise and uncertainty creates security issues.

“I foresee an unprecedented revolution in autonomous, intelligent machines, facilitated by a fruitful symbiosis between systems theory, computation and (neuro)biology,” she says.

pay ahead

Arbelaiz grew up in Spain, aware of the privilege of having access to a better education than her parents. His father earned a degree in economics through independent study while working to support his family. His daughter inherited his perseverance.

“The hardships my parents went through made them cherish self-directedness, lifelong learning and critical thinking,” she says. “They transmitted these values ​​to me, so I grew up to be a curious and persevering person, passionate about science and ready to seize all educational opportunities.”

In a desire to pass this on to others, she mentors STEM students who lack guidance or resources. “I strongly believe that we should promote talent everywhere, and mentorship could be the main driver to encourage underrepresented minorities to pursue careers in STEM,” she says.

An advocate for women in STEM, she served on the executive committee of Graduate Women at MIT (GWAMIT) and MIT Women in Mathematics, and participates in various panels and workshops. She also leads live experiences for children, such as at the MIT Museum’s Girls Day events.

“As scientists, we have a responsibility to share our knowledge, to inform the public about scientific discoveries and their impact, and to raise awareness of the value of research and the need to invest in it.

Arbelaiz also supports MIT’s Covid-19 outreach efforts, including discussions of mathematical modeling of the virus and translation into Basque of his former mentor John Bush’s MIT Covid-19 Indoor Safety app.

This interest in paying for her STEM knowledge is something she attributes to her education at MIT.

“MIT has been one of the best experiences of my life so far: it has brought tremendous academic, professional, and personal growth,” she says. “I share MIT’s taste for collaborative and multidisciplinary research, the attraction for intellectual challenges and the enthusiasm for advancing science and technology for the benefit of humanity.”

Leave a Reply

%d bloggers like this: