Attention on drone safety and data collection

The Software Technical Committee focuses on software engineering issues for complex and critical systems, including requirements, design, code, test, evaluation, operation and maintenance.

Software continues to enable increasingly capable and autonomous aerospace systems. Unmanned aerial systems autopilot software has become reliable, and in 2017, focus shifted to the development of user interface and payload support software. Open-source drone interface apps now allow users to virtually joystick from a touch screen; capture and livestream drone video; direct a drone to follow a smartphone or tablet based on GPS or follow a visual target; and identify obstacles from affordable sensors, such as video and sonar.

Counter-drone software and systems have also become an important area of study given incidents such as an October collision between a commercial aircraft and a drone in Canada. Counter-drone software must not only detect an intruding drone, but also classify it as sufficiently hostile to “bring down.” New products such as Maryland-based Department 13’s Mesmer counter-drone system — which became available in January — are exploiting weaknesses in digital radio links to take control of a drone’s computer for safe landing.

Comprehensive in-flight data collection is critical to understand system use patterns and reveal software flaws. Real-time manufacturer access also ensures patches and upgrades can be effectively installed. Yet, mobile devices also risk user privacy and security invasion. Drone data collection received increased attention for privacy reasons in 2017. In August, a U.S. Army memo cited “cyber vulnerabilities” as reason for banning DJI products because the Chinese drone manufacturer uses automatic data collection. While DJI code remains closed-source, tests using network analyzer products such as Wireshark have shown that actual data capture and reporting depends on choice of ground station software as well as the specific drone model.

NASA’s UAS Traffic Management continued to make progress in 2017 with a series of network-enabled tests in which UTM checked drone flight plans for conflicts, approved or rejected each plan, and notified users of constraints. UTM aims to collect systemwide data to enable beyond-line-of-sight flight capabilities as well as to provide weather and traffic data as soon as safety-critical network and software systems can be verified and validated.

Commercial passenger aircraft companies have also adopted new fleetwide data collection software products. In January, Boeing announced a new data analytics global services unit in partnership with Microsoft’s Azure cloud platform to provide a variety of services, such as real-time information about atmospheric conditions. In June, Airbus and partner California-based Palantir Technologies announced the release of data platform Skywise to provide an industrywide access point for secure cloud airline operational data. This data can also be analyzed to optimize each aircraft’s performance, track maintenance, and to improve operational efficiency and reliability. Announcing a new Asian Customer Care Center in April, Sikorsky is using advanced analytics to better predict and resolve potential issues before they cause flight groundings or safety issues. Data analytics products uncover patterns in aircraft performance and parts that can help improve flight safety, optimize aircraft operations and significantly reduce costs.

In the space sector, the NASA Space Robotics Challenge finals were in August. The competition required each team to program a virtual R5 robot to autonomously repair habitat damage caused by a Martian dust storm. The winning team, Coordinated Robotics of Newbury Park, California, deployed a virtual R5 (the NASA Johnson Space Center’s Valkyrie) that completed all competition tasks using technology augmented from the earlier DARPA virtual robotics challenge.

Contributors: James Paunicka and Stephen Blanchette