|
UAF makes historic unmanned flight
First FAA-approved true beyond-visual-line-of-sight domestic flight of an unmanned aircraft system flown over the TAPS corridor Steve Sutherlin Petroleum News
The Alaska Center for Unmanned Aircraft Systems Integration at the University of Alaska Fairbanks has flown the country’s first FAA-approved true beyond-visual-line-of-sight, BVLOS, domestic flight of an unmanned aircraft system under the small UAS rule.
The flight is a step toward more routine approval of commercial BVLOS unmanned aircraft flights.
Operators from UAF flew a Skyfront Perimeter long range hybrid-electric unmanned aircraft 3.87 miles along the trans-Alaska oil pipeline corridor July 31 near the Chatanika River on the Elliott Highway.
Onboard and ground-based detection systems, rather than human observers, were used to detect and avoid other aircraft in the airspace.
“The ability to use UAVs for surveillance in remote areas of the pipeline increases the tools at our disposal to operate TAPS more reliably and safely and better protect Alaska’s environment,” said Tom Barrett, president of Alyeska Pipeline Service Co.
The flights were a part of the Unmanned Aircraft Systems Integration Pilot Program, a national initiative from the U.S. Department of Transportation and the White House.
“The Integration Pilot Program is helping us advance the safe, secure and reliable integration of drones into the national airspace,” said FAA Acting Administrator Daniel K. Elwell.
“We want to ensure the safety of manned aviation while opening new opportunities for unmanned aircraft cargo deliveries to villages, monitoring of infrastructure, mammal surveys and a host of other missions of use to Alaskans,” said Dr. Catherine F. Cahill, ACUASI director.
Detect and avoid The BVLOS flight was made possible by an onboard computer vision detect and avoid system known as Casia, which was developed by Iris Automation Inc. of San Francisco.
“Beyond line-of-sight of the pilot, for the first time ever, really makes drones effective and efficient enough to do things like gas pipeline inspection - especially these long distance lineal infrastructure inspections - honestly, that just hasn’t been possible until now,” Iris CEO and co-founder Alexander Harmsen said.
“We’re replacing human vision,” he said. “These drones see the world the way pilots do, to be able to avoid collisions. All of that happens in real time; our system itself makes those decisions.”
The system detects, tracks and classifies aircraft and other objects in the sky.
“Our system is able to distinguish between a cloud, a bird, or a helicopter, and as you can imagine they are all very different risk profiles for the drone,” Harmsen said. “There’s a lot that’s happening all in real-time on board; we need to be able to have processing similar to the human brain. We’re able to make decisions based on the information we get in real time, we don’t want to have to rely on a communication.”
Casia relies on redundancy on board to be able to make sure that the system is running optimally with no dropout.
“We don’t want to fly with just a single algorithm, so many times we have parallel algorithms that continuously check in with each other to make sure they have the same values,” Harmsen said. “We’re holding the system to pretty high level of reliability partly because it’s aviation and there are certain standards within aviation, but also because we’re working in these industrial sectors where safety is the highest priority.”
Harmsen said it took two-and-a-half years of testing with the FAA to get approval for the July 31 flight.
“This is why it’s such a big deal for us and for the industry as a whole,” he said.
| 