The oscillation towards bio-inspired machine intelligence | MIT news

Goncal Arbeliz Mogica A native of Spain, octopus is a popular ingredient on the menu. However, Arbelaiz values ​​octopuses and similar creatures in a different way, through her research into soft robotics theory.

More than half of the octopus’ nerves are distributed through its eight arms, each with a certain degree of autonomy. This distributed sensing and information processing system has intrigued Arbelaiz, who is looking at how to design decentralized intelligence for man-made systems with integrated sensing and computation. At MIT, Arbelaiz is an applied mathematics student working on the fundamentals of optimal distributed control and estimation in the final weeks before completing her Ph.D. in the fall.

It finds inspiration in the biological intelligence of invertebrates such as octopus and jellyfish, with the ultimate goal of designing new control strategies for flexible “soft” robots that can be used in narrow or delicate surroundings, such as a surgical instrument or for research – and rescue missions.

“A soft robotic sponge allows it to adapt dynamically to different environments. Think worms, snakes, or jellyfish, and compare their locomotion and adaptation capabilities to those of vertebrates.” It is an interesting expression of embodied intelligence – the lack of a rigid skeleton gives advantages for some applications. And it helps to deal with real-world uncertainty more efficiently. But this extra smoothness also implies new theoretical challenges to the system.”

In the biological world, the ‘controller’ is usually connected to the brain and central nervous system – it creates motor commands for muscles to achieve movement. Jellyfish and some other soft organisms lack a central nervous center or brain. Inspired by this observation, she is now working towards a theory in which soft robotic systems can be controlled using decentralized sensory information sharing.

“When sensing and actuation is distributed in the robot’s body and computational capabilities on board are limited, it can be difficult to implement central intelligence,” she says. Therefore, we need this kind of decentralized scheme that, although only sharing sensory information locally, ensures the desired global behaviour. Some biological systems, such as jellyfish, are beautiful examples of decentralized control architectures – movement is achieved in the absence of a (central) brain. This is remarkable compared to what we can achieve with man-made machines.”

Smooth transition to MIT

Her graduate studies at the University of Navarre in San Sebastian led to her work with a professor at MIT John Bush in fluid dynamics. In 2015, Arbelaiz invited MIT as a visiting student to investigate droplet interactions. This led to 2018 paper in physical revision fluids, She is pursuing her PhD at the Massachusetts Institute of Technology.

In 2018, her doctoral research moved to the Interdisciplinary Technical Social System Research Center (SSRC), and she is now being advised by Ali Al Jadbaei, professor of engineering at JR East and chair of the Department of Civil and Environmental Engineering; and Associate Dean of the College of Engineering Annette “Biko” HosoiHe is the Neil and Jane Pappalardo Professor of Mechanical Engineering as well as Professor of Applied Mathematics. Arbelaiz also works regularly with Bassam Bamieh, co-director of the Center for Control, Dynamic Systems, and Computation at the University of California, Santa Barbara. She says working with this team of consultants gives her the freedom to explore the interdisciplinary research projects that she has been drawn to over the past five years.

For example, you use System theoretical approaches to design new optimal controllers and capabilities for systems with spatiotemporal dynamics, and to gain a basic understanding of the topology of sensory feedback connections required for optimal control of these systems. For soft robotic applications, this amounts to a rating of important sensory measurements to better stimulate all of the “muscles” of this robot. Is robot performance degraded when each operator has access only to the closest sensory measurements? Her research distinguishes 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

two years Schmidt fellowship for scienceWhich funds young researchers to pursue post-doctoral studies in a field different from their graduate work, will allow Arpels to further explore the intersection between biological and machine intelligence after graduation.

She plans to spend her postdoctoral time at Princeton University with Professor Naomi Leonard, working with researchers in systems biology, computer science, and robotics, to explore the reliability and robustness of biological and synthetic assemblies. Specifically, she is interested in learning how biological systems adapt efficiently to different environments so that she can apply this knowledge to man-made systems, such as autonomous machines, whose exposure to noise and uncertainty creates safety problems.

“I anticipate an unprecedented revolution approaching autonomous and intelligent machines, facilitated by the fruitful coexistence of systems theory, computation, and (neuro)biology,” she says.

push it forward

Arbelaiz grew up in Spain and is well aware of the privilege of receiving a better education from her parents. Her father earned a degree in economics through independent study while working to support his family. His daughter inherited his insistence.

“The difficulties my parents experienced made them cherish self-esteem, lifelong learning, and critical thinking,” she says. “They imparted these values ​​to me, so I grew up to be a curious and diligent person, passionate about science and ready to seize every educational opportunity.”

In her desire to pass this on to others, she mentors STEM students who lack direction or resources. “I firmly believe that we must foster talent everywhere, and mentoring can be the main driver for encouraging underrepresented minorities to pursue careers in STEM fields,” she says.

An advocate for women in science, technology, engineering and mathematics, she was part of the executive committee of Graduates of the Massachusetts Institute of Technology (GWAMIT) and MIT Women in Mathematicsand participates in various seminars and workshops. It also conducts live experiments for children, such as the MIT Museum Girls’ Day events.

“As scientists, we are responsible for sharing our knowledge, informing the public of scientific discoveries and their impact, and raising awareness of the value of research and the need to invest in it.”

Arbelaiz also supports MIT’s Covid-19 outreach efforts, including talks About the mathematical modeling of the virusand translated into Basque by her former teacher, John Bush MIT Covid-19 Indoor Safety App.

This interest in pushing her STEM knowledge forward is something she attributes to her MIT education.

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

Leave a Comment