It's a Bird, It's a Plane... It's a Robot!

Quadcopter drones are among the most agile unmanned aerial vehicles around, with the ability to hover, make sharp turns, and land on a dime. But these drones are also notorious for the way that they quickly drain their batteries. Continuously spinning rotors are excellent for keeping an aerial vehicle under control, but that comes at the expense of very poor energy efficiency.

If you want agility and energy efficiency, then you need look no further than the birds flying around just outside your window. Some species of birds are known to go several years at a time without touching the ground, powering their flight with little more than the insects that they snack on in flight. This incredible efficiency inspired a group of engineers at Northwestern Polytechnical University to design a flying robot that mimics the flight mechanics of birds.

A few years ago, this resulted in the development of RoboFalcon, a flapping-wing robot with a special mechanism that drives its wings to morph in flight. But while this robotic bird was great at a cruising speed, it was incapable of either taking off or flying at low speeds, which greatly limited its practicality in real-world applications.

To address these issues, the team has just reported on the creation of RoboFalcon 2.0. This new and improved robot can take off on its own and it has no problem with low-speed flight.

Instead of relying on a single, repetitive flapping motion like insect-inspired robots or hummingbird-like drones, the researchers developed a mechanism that couples flapping, sweeping, and folding into a single wingbeat cycle. Known as flap-sweep-fold (FSF) motion, this system more closely resembles the way birds and bats actually fly at low speeds. By folding the wings on the upstroke and sweeping them forward on the downstroke, RoboFalcon 2.0 can generate both lift and thrust in a way that mirrors natural bird takeoff.

Rather than use multiple heavy actuators, which would add too much weight for a mid-sized flying robot, the researchers created reconfigurable mechanisms that redistribute the motion of a single powerful actuator into FSF movements. These mechanisms, along with carefully tuned decouplers, allow RoboFalcon 2.0’s wings to morph in mid-flight while keeping the overall system light enough to get off the ground.

The team validated their design through a combination of wind tunnel testing, computer simulations, and real-world flights. The wind tunnel results demonstrated that increasing the wing sweep not only boosted lift but also improved pitch control — both essential for self-takeoff. Simulations confirmed that the aerodynamic benefits correlated with strong leading-edge vortices, while dynamics models showed how sweeping amplitude could be adjusted for stability. Finally, flight tests confirmed that RoboFalcon 2.0 could indeed launch from the ground and remain in stable flight at low speeds.

Despite these advances, RoboFalcon 2.0 is still a work in progress. The researchers note that its energy efficiency during takeoff lags behind that of real birds or even insect-scale robots. Its hovering ability is also constrained by limited yaw control, and at higher speeds, the lack of a tail elevator makes stability more difficult to maintain. Even so, this is a meaningful step forward for flapping-wing robots. With a bit more refinement, they will be ready to fly alongside real birds.