Drones - See-and-avoid the NASA way

NASA algorithms keep unmanned aircraft away from commercial aviation

NASA's DAIDALUS algorithms dole out maneuver guidance for the unmanned system pilot on the ground to remain "well clear" of other aircraft

It is one of the major issues of letting large unmanned aircraft share the sky with commercial airliners: preventing a disaster by keeping the two aircraft apart - or "well clear" in flight.

Commercial airliners and many larger private planes have onboard technology (and air traffic controllers as well as live pilots) to detect and avoid other aircraft in the sky but unmanned systems do not.

"The most difficult problem we are trying to solve is how do we replace the eyes of the pilot in the cockpit? We have developed, and are currently testing, detect-and-avoid algorithms. We're also running multiple research experiments to support the validation of this technology," said Maria Consiglio, who leads the NASA Langley Sense and Avoid/Separation Assurance Interoperability, in a statement.

One such system, known as DAIDALUS, or Detect and Avoid Alerting for Unmanned Systems works by using algorithms to process the incoming traffic surveillance sensor data that some larger unmanned aircraft have onboard.

The system spits out maneuver guidance for the unmanned system pilot on the ground to remain "well clear" of other traffic, NASA said. DAIDALUS doesn't just relay passive alerts, though. It also "sees" safe paths out of potentially dicey situations.

Researchers at NASA's Langley Research Center wrote of DAIDALIS: "In the case of a predicted well-clear violation, DAIDALUS also provides an algorithm that computes the time interval of well-clear violation. Furthermore, DAIDALUS implements algorithms for computing prevention bands, assuming a simple kinematic trajectory model. Prevention bands are ranges of track, ground speed, and vertical speed maneuvers that are predicted to be in well-clear violation within a given look-ahead time. These bands provide awareness information to remote pilots and assist them in avoiding certain areas in the airspace. When aircraft are not well clear, or when a well-clear violation is unavoidable, the DAIDALUS prevention bands algorithms compute well-clear recovery bands. Recovery bands are ranges of horizontal and vertical maneuvers that assist pilots in regaining well-clear status within the minimum possible time. Recovery bands are designed so that they do not conflict with resolution advisory maneuvers generated by systems such as TCAS II [An air traffic/alert collision avoidance system]."

"NASA has taken a nebulous concept - well clear - and come up with an elegant mathematical solution for it that can be implemented in software. So those who are interested now can take that and run with it. They can make avionics software that will keep people safe," said Keith Arthur, a co-project engineer on the NASA Langley team.

And the testing continues, NASA says. In simulations at NASA Langley's Air Traffic Operations Lab, or ATOL, researchers are looking at how well air traffic controllers can manage traffic in simulated airspace that contains unmanned aircraft equipped with detect-and-avoid systems and traditional manned aircraft operations.