When most people think about the concept of swarming drones, they don’t think about them playing aerial bumper cars with each other. The images of the swarming drone fleets that we have seen are those of coordinated flying bots which move in unison, maintaining a safe distance while often mimicking the movements of the others in the fleet.
But a research team from the University of Pennsylvania suggests that we stop coddling our drones. According to them, there’s nothing wrong with a little drone-on-drone contact, so long as we prepare our flying bots for that contact properly. After all, swarming animals bump into each other without risking a complete system failure, so why shouldn’t our drones be able to do the same?
It’s not the first time that crash protection technology has been proposed, but it is one of the first documented applications of crash-proof mechanisms in swarming drones. Swarms of drones have a higher likelihood for contact than those that are flying solo, and the team of Yash Mulgaonkar, Anurag Makineni, and Luis Guerrero-Bonilla, under the guidance of Vijay Kumar, suggested a different approach from the traditional collision-avoidance systems.
Consider it non-critical collision with a purpose.
Given that our robots can sustain collisions, we can fly in dark, unstructured environments and deploy small payloads by colliding, Mulgaonkar said. We demonstrate it here by flying in a dark basement and deploying small magnetic LED flares to illuminate the environment. This payload can be substituted with other small sensors as well, like those for measuring air quality, radiation, chemical contamination, etc.
Deriving its inspiration from the shape of a Gomboc, the collision-ready drone has many advantages over its predecessors, which almost always fall out of commission should they endure any significant level of contact with another drone or obstacle.
For one, the collisions allow the drones to map obstacles, probing – albeit violently – an unknown environment without consequence. The ability to hit objects means that, unlike more fragile drones that have to decelerate to reverse course, the rate of directional change can be much faster in collision-ready drones. You already read about the benefits of payload deployment, as described by Mulgaonkar.
This new perspective may help us reconsider why we have seen collisions as a negative instead of a potential advantage when it comes to drone design. Bees do it, birds do it, and now we are seeing that drones can crash and thrive too.