Ornitos.

Leonardo drew ornithopters obsessively. The Codex Atlanticus contains dozens of folios devoted to the problem of flapping-wing flight, and the consensus among historians is that none of them would have worked. The power-to-weight ratio of human muscle is wrong for the job. The wing kinematics are too complex for the materials of the fifteenth century. The whole project was a beautiful failure.

We are returning to it on purpose. Modern brushless motors, modern batteries, modern compliant materials, and five centuries of aerodynamic research have changed the constraint set entirely. The question is no longer whether a person can flap themselves into the air. It is whether a small machine, drawn from those same studies, can do something useful that a rotor-based drone cannot.

"L'uccello è strumento operante per legge matematica." — Leonardo, Codex Atlanticus f. 846v

A single brushless motor drives an eccentric crank. The crank is coupled to two compliant bamboo flexures — the wing roots — through short connecting rods. As the crank rotates, the flexures store and release energy, producing a flapping motion that is not driven directly by the motor but resonantly tuned to it.

The wings themselves are bamboo-spar primaries with hemp-linen membranes, modelled loosely on a kite frame. The trailing edge is deliberately soft; the leading edge is stiff. As the wing moves downward, the membrane scoops air. As it returns, it twists out of the way. This is not new physics — every bird does it — but applying it at the scale of a 1.4 m machine is hard.

What we have learned

• good news The wings flap, the flexures survive 100,000-cycle bench tests, and the system is genuinely quiet.
• bad news Lift is currently 60% of bodyweight. It does not yet fly.
• next Wing kinematics tuning. We think there is another 50% in the wing twist profile.

Ornitos is a research vehicle, and we expect to spend several more years on it. There are two reasons we think the work is worth doing.

First, a machine that flies the way a bird flies has signatures — acoustic, visual, aerodynamic — that no rotorcraft can match. If Ornitos ever crosses the lift threshold, it will change what observation in wildlife and conservation contexts looks like.

Second, the discipline of returning to a 500-year-old set of drawings and asking could you, with what we have now, finally make this work? is exactly the kind of work the workshop exists to do. Every iteration teaches us something about flexure design, bamboo processing, and resonant mechanisms that ends up improving the rotor-based machines too. None of it is wasted.

All Ornitos design files, test data, and lessons learned are published openly. If you are a researcher in flapping-wing flight, compliant mechanism design, or biomimetic aerodynamics, write to the workshop. We are happy to share, collaborate, and credit.

Write to the workshop →