Aircraft maker Airbus is turning to smart industrial drones, data analytics and machine learning to make aircraft inspections easier and faster.
One day while working on a shiny new Airbus A350 aircraft, Ronie Gnecco figured it was time to build a better relationship between drones and passenger airplanes.
His bold idea to use flying robots for aircraft safety inspections worked so well it has — among other projects — it inspired aircraft manufacturer Airbus to move deeper into the industrial drone revolution.
Within a couple of years, the company’s intelligent unmanned aerial vehicles (UAV) systems could be used for safety inspections at airports around the world, making planes safer with more on-time flight departures. To make that happen, Gnecco said it will require pioneering efforts from technology experts, regulators and airport authorities from around the world.
“Drones, sensors, database, cloud computing and machine learning are really key for what we are putting together,” said Gnecco, innovation manager for UAV Development & Applications at Airbus.
Gnecco is showing why drones are a valuable tool for the airline industry. Drone technology innovation is increasingly benefiting many industries. Sales of commercial drones – used for farming, construction, product delivery and other industries — are expected to grow from a $2 billion global market today to as much as $127 billion by 2020, according to PricewaterhouseCoopers LLP.
It usually takes two hours for two men hoisted in a cherry picker to quality inspect a new Airbus aircraft. Gnecco knew a camera-equipped UAV could do it faster and produce better data with great value for traceability.
“When I suggested it in 2013, people looked as if I was playing around, like I wanted to bring my hobby to work,” said Gnecco.
He was dead serious.
He brought his inexpensive drone, flight controller and a GoPro camera to the Airbus hangar at Toulouse-Blagnac airport in France. The high-definition images captured 25 meters in the air, looking down at an A350, where transferred online to the Airbus design office in Getafe, Spain. In a matter of minutes, the images verified the perfect condition of the new airliner.
To make results repeatable across Airbus, Gnecco sought a more robust, reliable drone. He turned to Ascending Technologies, an Intel-owned manufacturer of commercial UAVs. Together they designed a proof of conceptusing an AscTec Falcon 8. –The V-shaped octocopter that features the AscTec Trinity Autopilot, which allows the operator to use programmed waypoints that direct the drone’s flightpath safely and manually without GPS. They equipped it with Intel RealSense cameras, giving the UAV “robotic eyes” for obstacle navigation. The payload also included a 42-megapixel full-frame camera that captures data with millimeter accuracy.
“That RealSense camera helps us meet one of our main needs, which is anti-collision,” Gnecco said.
He said a new aircraft can cost hundreds of millions of Euros, so it’s critical to have an ultralight drone smart enough to avoid collisions on its own.
“The drone is intelligent enough to recognize all the features of the aircraft,” he said. “Any object or surface of the aircraft that is too close in proximity to the drone is identified and avoided.”
Programmed flight paths and trajectories allow the drone to fly safely around industrial and commercial areas, according to Gnecco. The drone flies autonomously but safety regulations require a human pilot to supervise the flight.
The drone takes aerial high quality photos used to create a 3D model of the aircraft. The digital images are analyzed by Airbus design-specific software that allows inspectors to identify damage, locate it precisely and document traces for a verifiable Inspection.
Digitizing Safety Inspections
Airbus made its first public flight at Farnborough 2016 International Airshow.
“With the Falcon and using software that we’ve developed, it takes about one flight 10 minutes to fly all around the aircraft, taking 150 high-definition pictures,” he said.
The need for speed, new technologies and repeatable processes was essential after Airbus compressed the A330 development timeline. Gnecco’s idea to use drones helped meet a need. It also sparked new opportunities for Airbus and the airline industry as a whole.
By the end of the year, the drone inspection system will be ready for use across manufacturing and development processes. Gnecco expects to bring hundreds of drones to work for Airbus next year. By digitizing inspections with drones, Airbus could save thousands of hours, speed communications across teams and ramp up aircraft production rate.
A quality inspector can zoom in and out, pan and twist the model searching for anything out of the ordinary, said Gnecco. If damage is identified, it get matched against a model of a flawless fuselage.
The precision of the pictures helps the computer system find the diagnosis. With each inspection, the database grows smarter with help from machine learning algorithms.
With each new aircraft inspection, Airbus’ data analytics system gets smarter by using machine learning leverages what Gnecco calls a living database.
“We are performing more than 3,000 inspections per year at Airbus, and we are creating a great database of damage for automatic damage recognition,” he said.
To build the complete system, Gnecco and his team assessed hundreds of companies, analyzing their drone technologies. He spent much of 2015 traveling around the world looking for the best pieces to build a reliable digital inspection system.
“Airlines are contacting us in order to have this technology already, but first we need to ensure a good and reliable experience before deploying it outside. There’s also work to be done with regulations.”
Before this digital system was built, Gnecco said inspections were rife with paper work, reference numbers, frames, figures and alphabetic letters. Now it’s faster and easier to transfer inspection data across different locations. That means more analysis can be done quickly if needed, and there is always a digital record of the inspection.
Future of Smart Drones
Beyond safety inspections, Gnecco is exploring how drones can be used for better weather measurement and airport surveillance.
“During instrumentation, we use the drone to do the weather measurement so that we can better analyze the acoustics of the engine during certification,” he said. “Before we were using a balloon with sensors that measured conditions every 30 minutes. With a drone we can get better telemetry of the data without losing a sensor at every measurement.”
Looking ahead, Gnecco can imagine drones with more onboard computer processing and memory so they could be loaded with database info. This would allow them to process images, identify damages and cross-check with the database in real time as it flies around the Aircraft. This could speed up the inspection process even more.
“The future of drones is really about the data they capture and what you do with it,” he said.
This all came about a few years ago after Gnecco shared his idea with Airbus BizLab, the internal innovation and utilization network. Airbus gave Gnecco the green light to create a proof of concept, identify regulations needed. It also led to the creation of a computer analysis and data visualization tool.
“Innovation is not just having an idea, it’s actually making that idea happen.”
Editor’s note: Learn more about the next generation Falcon, the Intel Falcon 8+ unmanned aerial system with full electronic system redundancy. It is designed with safety, ease, performance and precision for the North American markets. The Falcon 8 and Falcon 8+ devices have not been authorized as required by the rules of the Federal Communications Commission. These devices are not, and may not be, offered for sale or lease, or sold or leased, until authorization is obtained. Safety Certification, CE assessment and other country approvals not yet completed.