This Professor Turns Dead Birds Into Drones to Study Animals and Development of the Aviation Industry

On the other hand, a research team from the Massachusetts Institute of Technology (MIT) has developed a multi-agent trajectory planner for drones that produces a collision-free trajectory even when communication between agents is delayed.

This system, called Robust MADER, allows drones to create optimal and collision-free trajectories with other drones that may occupy the same airspace.

This system was developed in response to a challenge encountered when testing an initial system, the Multi-Agent Decomposed Evaluation of Risk (MADER), on a real drone.

The researchers found that if a drone does not have up-to-date information about the trajectories of its companions, it may inadvertently choose the path leading to a collision.

MADER incorporates a delay-check step that prevents such collisions. During the delay check period, the drone spends a certain amount of time checking communications from other drones to see if its new trajectory is clear.

If it detects a potential collision, the drone will leave the new trajectory and start the optimization process from scratch.

MADER is an asynchronous, decentralized and multi-agent path planner that optimizes collision-free trajectories using an algorithm that combines the paths received from other agents.

By continuously optimizing and broadcasting their new trajectories, drones can avoid collisions. Armed with Robust MADER, each drone formulates its trajectory, and while all agents must agree on each new trajectory, they don’t need to agree at the same time.