Researchers at MIT have developed algorithms that enable trajectory planning and control of electronic vertical takeoff and landing (eVTOL) fixed-wing aircraft, specifically drones used for tasks such as search-and-rescue or parcel delivery.
The MIT algorithm can execute maneuvers like sideways or upside-down flight and are computationally efficient enough to plan complex trajectories in real-time.
Previously, other methods to simplify trajectory planning had to use two different models: one for helicopter mode; and another for airplane mode. However, neither could plan and execute trajectories that the MIT algorithm can do, researchers said.
“We wanted to really exploit all the power the system has,” said Ezra Tal, researcher scientist for the Laboratory for Information and Decisions Systems (LIDS) at MIT. “These aircraft, even if they are very small, are quite powerful and capable of exciting acrobatic maneuvers. With our approach, using one model, we can cover the entire flight envelope — all the conditions in which the vehicle can fly.”
Researchers used trajectory generation and control algorithms to show how aircraft could do complex maneuvers like loops, rolls, climbing turns and even a drone race that sped through aerial gates and acrobatic maneuvers. The algorithms can also allow these drones to fly into dynamic environments like a collapsed building or avoid obstacles while searching for survivors.
Challenging conditions
While trajectory generation and control algorithms do exist for fixed-wing aircraft, most focus on calm trajectories with slow transitions. Challenging conditions need more sophisticated algorithms to enable the fixed-wing aircraft to reach peak performance.
To create the algorithms, researchers used a global dynamics model to apply to all flight conditions ranging from vertical take-off to forward, or even sideways. Then they leveraged a technical property known as differential flatness to ensure that the model would perform efficiently.
By using differential flatness, MIT can use a mathematical function to check whether a trajectory is feasible. This approach avoids system dynamics and plans a trajectory for the aircraft. The algorithm then uses differential flatness to check the feasibility of that trajectory.
“That check is computationally very cheap, so that is why with our algorithm, you can actually plan trajectories in real-time,” Tal said.
The full research can be found in the journal IEEE Xplore.
