Self-excited actuation enables adaptive and resilient flapping-wing flight
This paper introduces the first flight-capable flapping-wing robot utilizing asynchronous actuation to emulate insect muscles, demonstrating that this self-excited strategy enables adaptive, resilient, and instantaneous obstacle avoidance without requiring external sensing or control intervention.
Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the natural world, the ability to fly is often a matter of life and death. For the smallest creatures, from the fruit fly to the mosquito, flight is not just about moving from place to place; it is a high-speed negotiation with the air that requires split-second adjustments to stay aloft. To achieve this, nature has evolved two distinct ways for muscles to power a beating wing. The first method is direct and deliberate: a nerve signal tells a muscle to contract, the wing moves, and the brain waits for the next command. This is like a conductor telling an orchestra when to play each note. The second method is more mysterious and self-sustaining. Here, the muscle itself has a special property where stretching it triggers a contraction, which then stretches the opposing muscle, creating a continuous, rhythmic loop that runs on its own without constant orders from the brain. This self-starting rhythm allows insects to flap their wings hundreds of times a second, adapting instantly to changes in their body or the air around them. While engineers have long been able to build robots that fly by mimicking the first, direct method, the second, self-sustaining approach has remained elusive in machines. The question of whether a robot could harness this same self-correcting, automatic power to fly through a messy, obstacle-filled world has been an open challenge.
A team of researchers has now built the first flying robot that uses this self-sustaining muscle strategy. Instead of relying on a central computer to time every wingbeat, they created a machine where the wings generate their own rhythm through a feedback loop built into the motor itself. The robot, which weighs just five grams and has a wingspan of 155 millimeters, uses a clever trick to sense its own motion without adding heavy cameras or sensors. The electric motors that drive the wings also act as sensors; as the wings spin, they generate a small electrical signal that tells the system how fast they are moving. The robot's control system uses this signal to create a delay, mimicking the biological "stretch activation" found in insects. When the wing stretches, the system waits a fraction of a second before pulling it back, creating a continuous, self-sustaining oscillation. This design allows the robot to fly without a central brain constantly calculating the timing of every flap.
The results of this new approach are striking, particularly when the robot faces the chaos of the real world. In tests where the robot flew through a cage filled with vertical poles, the difference between the two methods was clear. Robots using the traditional, computer-timed method crashed quickly. When their wings hit a pole, the computer kept sending the same signal to flap, causing the wing to slam repeatedly against the obstacle with great force, destabilizing the entire machine. In contrast, the robot with the self-sustaining wings reacted instantly to the collision. As soon as a wing hit an obstacle and stopped moving, the feedback loop broke, and the motor naturally stopped pulling. The wing simply ceased flapping, allowing the robot to glide past the obstacle or recover without a violent crash. In these cluttered environments, the self-sustaining robot successfully reached the top of the cage in nine out of twelve attempts, while the traditional robot succeeded only twice.
This adaptability extends beyond just avoiding crashes. The researchers also tested how the robot handled damage. When they trimmed a portion of the wings to simulate injury, the traditional robot struggled to stay airborne, often failing to lift off the ground. The self-sustaining robot, however, automatically adjusted. Because its rhythm is tied to the physical weight and resistance of the wings, the loss of wing area changed the mechanical balance, and the system naturally sped up its flapping frequency to compensate. It increased its beat rate by roughly 12.6 percent, allowing it to generate enough lift to fly despite the damage, all without a computer detecting the injury or reprogramming the flight plan. Even when researchers caught the flying robot in mid-air, the wings immediately stopped beating the moment the motion halted and resumed instantly upon release, demonstrating a level of safety and responsiveness that does not require complex sensing or decision-making.
This work suggests a new path for the future of flying machines. By embedding the intelligence directly into the physical mechanics of the robot, rather than relying solely on software to manage every movement, engineers can create machines that are inherently more resilient. The robot does not need to be told to stop when it hits something; the physics of its own design forces it to stop. This approach mirrors how the most agile insects in nature have survived for millions of years, proving that sometimes the most advanced control system is not a faster computer, but a smarter body.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.