Fly drones hands-free using AeroCommand, a Raspberry Pi-powered AI controller
Flying a drone can be challenging, and this is even more true for individuals with physical impairments in their hands. Gustavs Andersons, a creative maker, has stepped up to address this challenge by developing AeroCommand, a project that enables hands-free drone piloting. Using a Raspberry Pi, Andersons has made it possible for people with limited hand dexterity to control a drone through alternative input methods.
Andersons, a student from Latvia, created AeroCommand as part of the Build2GetherV2 Contest. His goal was to assist those with various hand limitations by using artificial intelligence to interpret different types of input to control a DJI Mavic Pro drone. The innovative system utilizes a camera to capture facial movements and gestures, which are then processed by an AI system to control the drone. For example, nodding the head adjusts the throttle, while winking signals yaw. Additionally, a voice-control option is available for those who may not be able to use facial movements.
The project is powered by a Raspberry Pi 3B+, demonstrating that even older models of single-board computers can handle such tasks. The Pi communicates with the drone’s remote control and an Android smartphone wirelessly and integrates with a Seeed Studio Sense for speech-to-text functionality. Andersons also developed a custom PCB to connect all these components seamlessly, making the system both functional and efficient.
To promote accessibility, Andersons has made AeroCommand entirely open-source, allowing anyone to replicate it at home. The source code is available on GitHub, and a full build guide is posted on Hackster. Andersons is also planning to release a demo video once repairs on a broken motor are complete, so be sure to follow his progress for future updates.
Read more: Fly drones hands-free using AeroCommand, a Raspberry Pi-powered AI controller
