In Yu Gu’s Interactive Robotics Laboratory at WVU, graduate student Trevor Smith observes Loopy, a multicellular robot that learns to respond organically and autonomously to its environment.
(WVU Photo/Brian Persinger)

West Virginia University With Loopy, a “multicellular robot” consisting of a ring of individual, interconnected robot cells, roboticists are working on an alternative path to robot autonomy.

With support from a grant from the National Science Foundation, the WVU team will test Loopy’s ability to “co-design” itself, determining its own shape with limited assistance from human engineers. Because its behavior is not directly programmed, they believe Loopy can learn to use its body to mark the boundary of a contaminated area, such as the site of an oil or toxic spill.

Inspired by natural phenomena such as a swarm of ants gathering around a spilled lemonade or a system of tree roots growing around obstacles, Loopy changes its shape as each of its cells reacts organically to its environment.

Senior Researcher Yu GuThe Mechanical engineering, materials engineering and aerospace engineering Academy of Distinguished Alumni Professors on WVU Benjamin M. Statler College of Engineering and Mineral Resourcessaid that Loopy’s ability to transform itself could represent a breakthrough in robotics, with the potential to respond flexibly to unpredictable situations in the real world that conventional robots do not have.

““Loopy started as a thought experiment in my lab,” Gu said. “It was intended to challenge the prevailing top-down thinking in robotics, where the robot is passive and the human designs, programs and builds it.”

“Loopy, on the other hand, is an example of ‘swarm robotics’. Many small robot cells connect to form Loopy, which allows the cells’ simple, decentralized responses to stimuli to develop lifelike properties and complex, coordinated behaviors such as problem solving.”

Loopy’s body is made up of 36 identical cells physically connected in a circle. Each cell can control its own movement and has sensors that inform it of its joint angle as well as external stimuli such as light and temperature.

To track how Loopy responds to different situations, Gu’s lab is equipped with a tabletop test environment fitted with overhead cameras, a motion capture system and a projector. Underneath the table, heating wires create warm spots that simulate areas of contamination. A thermal imaging camera on top visualizes the heat map, and a temperature sensor is embedded in the feet of each of Loopy’s cells.

With doctoral student and NSF fellow Trevor SmithGu, of Appalachia, Pennsylvania, will test Loopy under a variety of unpredictable conditions, including different surface materials and obstacles. They will evaluate Loopy’s accuracy in circling contaminated areas, Loopy’s reactions to the unexpected, and Loopy’s tolerance to situations about which it has little or inaccurate information.

At the same time, they compare the solutions that Loopy finds organically with a more conventional, centralized approach in which a human designer can access all sensor data and control Loopy’s individual cells.

“The research trajectory on Loopy is likely to be nonlinear and unpredictable,” Gu said. “Most of the time, the outcome of our experiments with Loopy is unexpected, and that has been a source of insight and a stimulus for future research.”

“We want to know whether Loopy’s self-organized problem-solving offers greater adaptability and resilience than programmed behaviors, and how to harness robot swarm behavior for practical applications. Once we have created the conditions that encourage the spontaneous emergence of these complex behaviors in multicellular robots, I believe robots that function like Loopy will have potential for applications as diverse as adaptive leak sealing or interactive art displays.”

While traditional top-down robotic systems are “unnatural and brittle” and have difficulty adapting to new conditions, the collective intelligence of simple cells in swarm robotics enables the natural emergence of new behaviors through a “bottom-up” process.

“Our approach is philosophically similar to permaculture, where human land managers work with nature, not against it, to create self-sufficient, sustainable agricultural ecosystems,” said Gu. “In our robot design process, there are three equal players: humans, robots and the environment.”

Of the various biological models for Loopy, Gu was particularly inspired by studies of plant intelligence. For example, he used chemical signaling in plants as a model for how decentralized information between cells can contribute to collective behavior.

“Plant roots grow by producing new cells,” he explained. “Each of these cells responds to external factors, such as the presence of water or nutrients, and to internal factors, such as hormones. These responses coordinate en masse root growth – where the roots go, what shapes they take. This is just one biological mechanism that underscores the importance of distributed coordination as opposed to central control in complex systems.”

“This work blurs the boundaries between a robot’s physical form, its behavior, and its environment,” Gu added. “Loopy could fundamentally change our understanding of autonomy, adaptability, and design in robotics.”

-WVU-

mm/8/21/24

MEDIA CONTACT: Micaela Morrissette
scientific Assistant
WVU Research Communication
304-709-6667; [email protected]

Call 1-855-WVU-NEWS for the latest West Virginia University news and information from WVUToday.