In February 2017 South Sudan declared a state of famine, reporting that over 100,000 people were facing starvation. The World Food Programme warned that over the following six months twenty million people would be directly affected across the globe. To address this, an aid delivery programme that is both scalable and efficient is urgently required.
However, current methods are extremely ineffective, expensive, and difficult to scale. In countries with infrastructure as poor as South Sudan, airdrops are often the only viable option – despite the associated costs. This project presents Unmanned Aerial Vehicles (UAVs) as a promising solution: a safer, more scalable, and cheaper way to deliver aid.
DF2 is an innovative autopilot philosophy which introduces resilience through the mass redundancy of sensors, actuators, and flight controllers, eliminating single-points-of-failure. This facilitates the development of UAVs with far longer service lives than conventional UAVs, and with maintenance regimes more akin to civil aviation.
To introduce mass redundancy, the UAV’s control surfaces are subdivided into ‘modules’, each containing a full autopilot and communicating as part of a mesh network. The bespoke architecture and printed circuit boards were developed in an iterative process, with extensive testing in simulations and experiments, including full-scale flight-tests with a bespoke UAV airframe. This demonstrated its viability as a highly resilient autopilot architecture that is suitable for long-range aid delivery missions.
The project has delivered an excellent range of platforms for future development: an extensive simulations environment; a lockable 2-axis gimbal for full-scale wind-tunnel testing; a flightready airframe for full-system testing; and a complete DF2 avionics package. Above DF2 during flight testing Above DF2 during flight testing