An autonomous quadcopter I'm designing as a personal engineering project, with the goal of operating for extended periods without manual recharging — and a study in letting the energy budget dictate the architecture. Onboard solar math ruled out fully self-powered flight, so the harvest is split: cells on the airframe for top-up and loiter margin, and a solar-powered charging station on the ground for the real recharge. The 190 mm airframe shown below is shaped by a second constraint most solar-drone concepts ignore: on a quadrotor, the propellers occupy the exact real estate you'd want to put panels on.
The airframe is modelled parametrically in Python (CadQuery / OpenCASCADE) rather than drawn by hand, so wheelbase, propeller size, wing width and panel dimensions are driven variables. Six parts export to STEP for SolidWorks and STL for printing; all meshes are verified watertight and manifold before release.
Arms print in PETG-CF with layer lines running across the arm, not along it — layer adhesion is the governing failure mode. Solar side uses flexible ETFE monocrystalline cells into 1.2 mm recesses, feeding a CN3791 MPPT charge controller; a plain diode would discard roughly 30% of an already small harvest.
On the ground side, a solar-powered charging station recharges the aircraft between flights: when the battery reaches a low charge level, the drone is designed to return, dock, recharge, and resume operation on its own. I'm also exploring computer vision and autonomous navigation so it can interpret its surroundings in real time, avoid obstacles, and perform simple tasks independently — with applications like package delivery, environmental monitoring, and surveillance in mind.
Status: In development since May 2026. I designed the components and full assembly in SolidWorks, including the parametric airframe shown above, and 3D-printed the custom parts for prototyping. I'm currently assembling the system and beginning initial testing and iteration in my free time.