The wriggly gait of centipedes is well known. They can move through any terrain without stopping thanks to their tens to hundreds of legs.
“At the point when you see a dashing centipede, you’re fundamentally seeing a creature that occupies a world that is totally different than our universe of development,” said Daniel Goldman, the Dunn Family Teacher in the School of Material Science. “Our development is to a great extent overwhelmed by dormancy. I land on my foot and move forward when I swing my leg. However, in the world of centipedes, if they stop moving their limbs and body parts, they basically stop moving right away.
A group of physicists, engineers, and mathematicians at the Georgia Institute of Technology is making use of this style of movement to their advantage in order to investigate the possibility that the numerous limbs might be useful for locomotion in this world. They discovered that a robot with redundant legs could move across uneven surfaces without the need for additional sensing or control technology, as predicted by a new theory of multilegged locomotion and the creation of many-legged robotic models.
“Our movement is primarily governed by inertia.” I fall on my foot and go forward when I swing my leg. However, in the world of centipedes, if they cease wiggling their body parts and limbs, they effectively stop moving.”
Daniel Goldman, the Dunn Family Professor in the School of Physics.
These robots could be used for agriculture, space exploration, and even search and rescue because they can move over difficult, bumpy terrain.
In the papers titled “Multilegged Matter Transport,” the researchers described their findings. A System for Velocity on Boisterous Scenes,” in Science in May, and “Self-Impetus by Means of Slipping: In March, researchers published “Frictional Swimming in Multilegged Locomotors” in the Proceedings of the National Academy of Sciences.
A leg up
For the science paper, the researchers were inspired by the communication theory of mathematician Claude Shannon, which explains how to reliably transmit signals over long distances. They wanted to figure out why a multilegged robot could move so well. The hypothesis of correspondence recommends that one method for guaranteeing a message gets from point A to point B on an uproarious line isn’t to convey it as a simple message but to break it into discrete computerized units and rehash these units with a fitting code.
Credit: Science (2023). DOI: 10.1126/science.ade4985
Baxi Chong, a postdoctoral researcher in physics, stated, “We were inspired by this theory, and we tried to see if redundancy could be helpful in matter transportation.” Therefore, we began this project to see what would happen if the robot had more legs: four, six, eight, or even sixteen legs.”
A theory developed by a group led by Chong and consisting of Professor Greg Blekherman and Daniel Irvine, a postdoctoral fellow in the School of Mathematics, suggested that the addition of leg pairs to the robot improves its capacity to move steadily over challenging surfaces. This idea is referred to as spatial redundancy.
The robot’s legs can function independently thanks to this redundancy, without the need for sensors to interpret the environment. Assuming one leg vacillates, the overflow of legs keeps it moving in any case. On difficult or “noisy” landscapes, the robot transforms into a reliable system for moving itself and even a load. The idea is similar to how, without having to engineer the environment, punctuality on wheeled transportation can be guaranteed if the track or rail is smooth enough.
“With a high-level bipedal robot, numerous sensors are normally expected to control it continuously,” Chong said. ” However, in applications like search and rescue, exploring Mars, and even microrobots, a robot with limited sensing must be driven. This initiative to eliminate sensors has numerous justifications. The environments can change so quickly that there isn’t enough time for the sensors and controllers to respond, or the sensors can be expensive and fragile.
Juntao He, a robotics Ph.D. student, and Daniel Soto, a master’s student at the George W. Woodruff School of Mechanical Engineering, built terrains to resemble an inconsistent natural environment to test this.
After that, he put the robot through its paces by gradually increasing the number of its legs by two, beginning with six and eventually increasing it to 16. As the leg count expanded, the robot could all the more easily get across the territory, even without sensors, as the hypothesis anticipated. In the end, they put the robot through its paces outdoors on real terrain, where it was able to move through a variety of conditions.
Next steps
“It’s really noteworthy to witness the multilegged robot’s capability in exploring both lab-based territories and open-air conditions,” Juntao said. “Our multilegged robot uses leg redundancy and can accomplish similar tasks with open-loop control, whereas bipedal and quadrupedal robots heavily rely on sensors to traverse complex terrain.”
Next steps The findings are already being applied to farming by the researchers. Goldman is a co-founder of a business that wants to use these robots to get rid of weeds on farmland where weed killers don’t work.
“They’re similar to a Roomba yet outside for complex ground,” Goldman said. “A Roomba works since it has wheels that work well on level ground. Prior to the development of our framework, we were unable to reliably predict locomotor reliability on terrain that was rocky, bumpy, and littered with debris. We presently have the starting points of such a plan, which could be utilized to guarantee that our robots cross a yield field in a specific measure of time.”
The specialists also need to refine the robot. They are now determining the optimal number of legs to achieve motion without sensing in a manner that is both cost-effective and still retains the benefits. They are aware of the reasons why the framework of the centipede robot works.
“We asked, “How do you predict the minimum number of legs to achieve such tasks?”” in this paper. Chong stated, Currently, all we know is that the minimum number of legs exists, but we do not know the exact number. In addition, the tradeoff between robustness, speed, energy, and complexity must be better understood.”
More information: Baxi Chong et al, Multilegged matter transport: A framework for locomotion on noisy landscapes, Science (2023). DOI: 10.1126/science.ade4985
Baxi Chong et al, Self-propulsion via slipping: Frictional swimming in multilegged locomotors, Proceedings of the National Academy of Sciences (2023). DOI: 10.1073/pnas.2213698120
