Black Swift Technologies’ STL (Safe-to-Land) AI Solution

Coupling “failure prediction” with a vision system can be used to avoid ground obstacles during a forced landing. Now Black Swift Technologies (BST) has used AI neural networks to develop SwiftSTL™ so a UAS can autonomously locate a safe landing site in the event of system or flight failure.

The Boulder, Colorado, firm says it essentially employs AI and other technologies as a “Sully-factor.” The nickname references Chesley Sullenberger, the airline pilot who in 2009 landed US Airways Flight 1549 in the Hudson River after its engines failed, without any fatalities.

Leveraging AI and online machine vision classifiers to identify people and obstacles for safe landing.

SwiftSTL combines AI, algorithms, machine learning and onboard pixel-level processing to examine position, energy state, altitude, wind speed and other flight factors. If danger is detected, the technology invokes an AI (artificial intelligence) algorithm to steer to a safe landing site. The AI uses real-time video or still images to classify potential hazards along a hierarchy from people to terrain while it determines said optimal landing spot. These classifications are made visible via color-coded semantic segmentation maps. The technology monitors any objects on the landing zone as the UAS descends from higher altitudes.

The AI also is said to be able to predict servo, propulsion, battery, comms or GPS failures before they happen, to allow for preventive maintenance.

Black Swift CEO Jack Elston summarized SwiftSTL’s capabilities. “This technology is of value to all commercial drone operators, whether they are delivering packages or medical supplies,” he said in a release. “Regardless, the drone needs to identify where it is at all times so that when it makes its delivery that it avoids people, vehicles or structures. The same is true if the UAS needs to make a safe landing.”

Black Swift also produces the Black Swift S2™, a research sUAS capable of conducting scientific missions in harsh environments; it’s been chosen to support the East Greenland Ice Core Project. Other products include the “smart” SwiftCore™ Flight Management System and various field-swappable payloads.

Images courtesy of Black Swift Technologies.