Drone use is growing in multiple verticals with drone swarms and multiple drones working in tandem becoming a trend in agriculture, industrial and logistics.
As more drones are launched into the air, the greater the chances are of these drones colliding in the air. This becomes even more dangerous with drone swarms that work together for a common goal.
To help avoid costly crashes, researchers at MIT have created a new system called Robust MADER, a multiagent trajectory planner that generates collision-free trajectories even when communication between drones is delayed.
Robust MADER is an algorithm that incorporates a delay-check step during which a drone must wait a specific amount of time before it commits to a new trajectory. If it receives a new trajectory from fellow drones during the delay period, it might abandon its new trajectory and start the optimization process over again.
“MADER worked great in simulations, but it hadn’t been tested in hardware,” said Kota Konda, an MIT aeronautics and astronautics graduate student working on the project. “So, we built a bunch of drones and started flying them. The drones need to talk to each other to share trajectories, but once you start flying, you realize pretty quickly that there are always communication delays that introduce some failures.”
MIT tested MADER in simulations and flight experiments with real drones. It achieved a 100% success rate at generating collision-free trajectories. The travel time was slower than it would be with other approaches, but safety was secured.
“If you want to fly safer, you have to be careful, so it is reasonable that if you don’t want to collide with an obstacle, it will take you more time to get to your destination,” Kondo said. “If you collide with something, no matter how fast you go, it doesn’t really matter because you won’t reach your destination.”
How they tested it
MIT researchers built six drones and two aerial obstacles to test Robust MADER in a multi-agent flight environment.
The original version of MADER had issues and resulted in several collisions. But Robust MADER with a new algorithm did not cause a single crash in any of the hardware testing.
Drones were able to fly 3.4 meters per second with Robust MADER, slightly longer than the average travel time of drones, however it was collision-free in every experiment.
The next steps are to test the drones with the Robust MADER technology outdoors where many obstacles and types of noise can affect communications. Additionally, MIT wants to outfit the drones with visual sensors that can detect other agents or obstacles as well as predict their movements.
The complete research can be found on the arXiv Forum.
