By Adam Zewe | MIT News Agency
Bees are clumsy fliers. It is estimated that foraging bees bump into flowers about once per second, which over time damages their wings. But even though it has many rips or small holes in its wings, bees can still fly.
Air robots, on the other hand, are not so tough. Punch a hole in the robot’s wing motor or cut off part of the propeller, and it will likely be grounded.
Inspired by the resilience of bees, MIT researchers have developed a repair technique that allows an insect-sized aerial robot to sustain severe damage to the actuators, or artificial muscles, that move its wings – but still fly effectively.
They optimized this artificial muscle so that the robot can better isolate defects and deal with minor defects, such as pinholes in actuators. In addition, they demonstrated a new laser repair method that can help robots recover from severe damage, such as a fire that engulfs a device.
Using their technique, the damaged robot is able to maintain flight-level performance after one of its artificial muscles is punctured with 10 needles, and the actuator is still able to operate after a large hole is burned in it. Their repair method allowed the robot to fly even after the researchers had cut off 20 percent of its wingtips.
This could make swarms of small robots more capable of performing tasks in harsh environments, such as carrying out search missions through collapsed buildings or dense forests.
“We spent a great deal of time understanding the dynamics of artificial soft muscle and, through new fabrication methods and new understandings, we were able to demonstrate a level of resistance to damage comparable to that of insects,” said Kevin ChenD. Reid Weedon, Jr. Assistant Professor in the Department of Electrical Engineering and Computer Science (EECS), head of the Soft and Micro Robotics Laboratory at the Electronics Research Laboratory (RLE), and senior author of paper on this latest advance. “We are very excited about this. But insects are still superior to us, in that they can lose up to 40 percent of their wings and still fly. We still have some work to do.”
Chen co-authored the paper with lead co-authors Suhan Kim and Yi-Hsuan Hsiao, who are EECS graduate students; Younghoon Lee, postdoc; Weikun “Spencer” Zhu, a graduate student in the Department of Chemical Engineering; Zhijian Ren, an EECS graduate student; And Farnaz Niroui, EE Landsman Assistant Professor of Career Development from EECS at MIT and a member of RLE. The article appears in Science Robotics.
Robot repair technique
The small rectangular robot being developed in Chen’s lab is about the same size and shape as a micro cassette tape, even though one robot weighs no more than a paperclip. The wings at each corner are powered by dielectric elastomer actuators (DEA), which are soft artificial muscles that use mechanical force to rapidly flap the wings. The artificial muscle is made of layers of elastomeric sandwiched between two razor-thin electrodes and then rolled into a slick tube. When a voltage is applied to the DEA, the electrodes compress the elastomer, which flaps.
But microscopic imperfections can cause sparks that burn the elastomer and cause the device to malfunction. About 15 years ago, researchers discovered that they could prevent DEA failure from one small flaw by using a physical phenomenon known as self-clearing. In this process, applying a high voltage to the DEA disconnects the local electrodes around the small defect, isolating the failure from other electrodes so that the artificial muscle still functions.
Chen and his collaborators use this self-cleaning process in their robot repair technique.
First, they optimized the concentration of the carbon nanotubes that comprise the electrodes in DEA. Carbon nanotubes are very strong but very small coils of carbon. Having fewer carbon nanotubes in the electrodes increases self-cleaning, as they reach higher temperatures and burn more easily. But this also reduces the actuator power density.
“At some point, you won’t be able to get enough energy out of the system, but we need a lot of energy and power to fly the robot. We have to find the optimal point between these two constraints — optimizing the self-clearing property under the constraint that we still want the robot to fly,” said Chen.
However, even optimized DEA will fail if it suffers severe damage, such as a large hole that allows too much air into the device.
Chen and his team use lasers to deal massive damage. They carefully cut the outer contour of the large defect with a laser, which causes minor damage all around. Then, they can use self-clearing to burn off the slightly damaged electrodes, isolating the larger defects.
“On the one hand, we are trying to do surgery on the muscles. But if we don’t use enough power, then we can’t do enough damage to isolate the damage. On the other hand, if we use too much power, the laser will cause severe damage to the actuator that cannot be cleaned,” said Chen.
The team quickly realized that, when “operating” on such a small device, it was extremely difficult to observe the electrodes to see if they had managed to isolate the flaw. Draw previous job, they feed electroluminescent particles into the actuator. Now, if they see a light shining, they know that part of the actuator is operating, but a dark patch means they have successfully isolated the area.
Flight test success
After they perfected their technique, the researchers conducted tests with damaged actuators – some had been pierced by multiple needles while others were hollow. They measured how well the robot performed in flapping, take-off and hovering experiments.
Even with a damaged DEA, repair techniques allow the robot to maintain its flight performance, with altitude, position, and attitude errors that deviate only slightly from those of the undamaged robot. With laser surgery, a DEA that would be damaged beyond repair was able to recover 87 percent of its performance.
“I have to hand it over to my two students, who put a lot of legwork into flying the robot. Flying the robot itself is very difficult, especially now that we are deliberately destroying it,” said Chen.
This refinement technique makes the tiny robots much more powerful, so Chen and his team are now working to teach them new functions, such as landing on flowers or flying in groups. They also developed new control algorithms so the robots could fly better, taught the robots to control their yaw angle so they could maintain a constant direction, and allowed the robots to carry small circuits, with the long-term goal of carrying their own. resource.
“This work is important because of small flying robots — and flying insects! – constantly collide with their environment. A small gust of wind can be a big problem for small insects and robots. Therefore, we need methods to increase their robustness if we hope to use robots like these in natural environments,” said Nick Gravish, a professor in the Department of Mechanical and Aerospace Engineering at the University of California at San Diego, who was not involved in the study. “This paper shows how gentle actuation and body mechanics can adapt to damage and I think is an impressive step forward.”
This work was funded, in part, by the National Science Foundation (NSF) and the MathWorks Fellowship.