Daedalean’s first machine learning avionics system soon in the market

Founded in 2016, Daedalean has been developing autonomous dynamic flight control systems and software for safety-critical onboard applications, based on computer vision and neural networks. The company’s ambition is to build an AI-based autopilot that will enable future aircraft to fly autonomously.

Since 2018, Daedalean has been improving its suite of products including landing guidance, detect-and-avoid, GPS-independent navigation systems which could be implemented as a cockpit aid for existing aviation. In 2019, the startup signed an agreement on technological partnership with Honeywell and their subsidiary Bendix King to jointly bring to the market new products for aviation as well as for Urban Air Mobility (UAM). Subsequently, the team has been working on specifying and flight testing the functions for the needs of General Aviation.

Commercialisation partnership with avionics pioneer
In a further strategic move, the startup has signed a commercialisation partnership with Avidyne, a Melbourn (Florida) based aviation electronics company specialising in innovation and product design make flying safer, more accessible and more enjoyable for pilots and their passengers. The two companies will jointly develop, certify and bring to the market a detect-and-avoid system, designed as several cameras and a powerful computation unit that will act as a pilot assistance tool in existing cockpits.

Avidyne already offers traffic advisory products for aeroplanes and helicopters. However, these active surveillance systems only detect aircraft that are equipped with transponders or automatic dependent surveillance-broadcast (ADS-B) systems. By contrast, Daedalean is using convolutional neural networks, to analyse visual images from high-resolution cameras and can identify potential hazards. The new product will hence detect both airborne and ground threats, including non-cooperative traffic.

The startup will also benefit from Avidyne’s expertise in hardware design and certification to take the product through the next steps to commercialisation. The first application will be aimed to first provide a cockpit aid to manned civil GA aircraft (both fixed-wing and rotorcraft). The system is also well suited to address the demands of the emerging UAM industry, enabling safe flights of eVTOLs. Daedalean plans to launch the system in the market in 2021.

“We’re delighted with this agreement,” said Daedalean's CEO Luuk van Dijk, explaining that it will allow Daedalean to focus on its strengths as a software developer. “The commercial agreement is really about bringing this detect-and-avoid function to the market together — they do the box, we do the software.”

To sustain its growth, Daedalean is currently raising $25m financing round. in the past, it has raised CHF15m and was awarded a €2.3 million Horizon 2020 grant.