AI-based autonomous flight control for the electric passenger aircraft of the nearest future

Informations projet

Corvid

N° de convention de subvention: 782888

DOI

10.3030/782888

Projet clôturé

Date de signature de la CE 1 Août 2017

Date de début 1 Septembre 2017

Date de fin 31 Janvier 2018

Financé au titre de

• SOCIETAL CHALLENGES - Smart, Green And Integrated Transport

Coût total

€ 71 429,00

Contribution de l’UE

€ 50 000,00

Coordonné par

DAEDALEAN AG
Switzerland

Periodic Reporting for period 1 - Corvid (AI-based autonomous flight control for the electric passenger aircraft of the nearest future)

Période du rapport: 2017-09-01 au 2018-01-31

Résumé du contexte et des objectifs généraux du projet

In 2018, experts seem to have reached the consensus that artificial intelligence(AI) is essentially the new electricity—its impact on different industries and markets will be immense. AI has the potential to redefine all areas of life, from healthcare to manufacturing, either expanding the capabilities of human operators or freeing them from mundane and error-prone activities.

At Daedalean we focus on bringing AI to one of the most important global industries—aviation. Aviation, already a highly automated industry, is also highly regulated, and the combination of these factors has contributed to its stellar safety record. However, as consumer demand for commercial flight continues to drive up the number of flights, the industry already experiences a shortage of qualified pilots.

Our ambition is to build and commercialize Corvid—the first AI-based autopilot capable not only of successfully performing all the tasks of human pilots for different aircraft, but outperforming them in nominal and off-nominal situations.

Developing such an autopilot is an extremely complex, challenging and capital-intensive task, which requires solving a lot of technical issues that have never been tackled before. However, the fate of such technology will be ultimately decided by the regulatory bodies in different countries. Realizing this, we have uniquely built our go-to-market strategy in alignment with the regulators. First, we are collaborating with them from day one to determine how an AI-based autopilot can be certified. Second, we are designing our autopilot to pass the pilot tests that already exist for humans to create an important legal precedent, which we hope will also pave the road for wider AI adoption in other heavily regulated industries.

We have identified the nascent market segment with a huge growth potential for rollout of our technology—electric passenger vertical take-off and landing aircraft (VTOLs), marketed to the public largely as flying taxis. Recent advances in battery density and distributed electric propulsions made such small aircraft feasible and experts believe that in 2-3 years they will have sufficient performance parameters for commercial rollouts, replacing the taxi cars. However, to take over the taxi market VTOLs will have to be fully autonomous: 1) there will not be enough pilots for the eventual 1,000,000+ VTOLs deployed globally; 2) the costs of a human pilot will make the business model of the VTOLs-as-taxis unsustainable; 3) a pilot will take up a commercially impractical amount of room in a typical 1-2 seater VTOL; 4) a pilot will be unlikely to cope with the extremely dense VTOL traffic.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The goal of this Horizon 2020 SME Instrument Phase 1 feasibility study was to allow us to understand our target market, validate the problem/solution, and validate the product/market fit. We were able to exceed our expectations greatly—not only we were able to start detailed technical conversations with major VTOL manufacturers, suppliers and other partners, we signed five LOIs and eighteen NDAs, validating their willingness to pay for our solution and confirming our pricing. We also signed one commercial cooperation agreement with an industrial unmanned VTOL avionics manufacturer. This cooperation provides us an excellent engineering validation platform and program.
This gives us confidence to continue development, and provided us with valuable information and business contacts to raise additional funding to support our startup. We also conclude that next to raising funding from private sources, we are fit to apply for the Horizon 2020 SME Instrument Phase 2 subsidy in 2018.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

In the pursuit of one of the critical components of Corvid, we have achieved and in some cases surpassed state-of-the-art in our application of computer vision. In the longer-term, almost all of the systems that we develop will either match or exceed what is currently considered to be state-of-the-art.

The wider societal impacts of Corvid include:
- Reduction of emissions: Electric VTOL aircraft by definition have zero primary (operational) emissions. Uber calculates that one VTOL has the potential to render ~70 personal cars unnecessary.
- Traffic congestion: For many urban dwellers the flying taxi value proposition—a 15-20 min flight instead of a 2-hour slow drive—is sufficient to reduce conventional car usage (either personal or shared). As the price of a kilometre flown with VTOL drops down to the price of kilometre driven, we will see massive reduction of car usage, and thus much less congestion on the streets of cities like Paris or London.
- Reduced noise pollution: At flying altitude, noise from advanced electric vehicles will be barely audible. Even during take-off and landing, the noise will be comparable to existing background noise.
- Productivity increase: Less time spent in traffic ultimately means more time for productive work. An average San Francisco resident spends 10 days/year commuting, an average Mumbaian—7 weeks. The numbers for the major EU cities like London and Paris are not substantially smaller.

The Deadalean Team