Autonomous flight control for the electric personal aircraft of the near future
Daedalean is a Zürich-based startup, and our mission is to build a certified airworthy autopilot that can also pass the exams for human pilots.
After more than a century of powered aviation the navigation, guidance and control of aircraft is still a very manual affair, with many functions relying directly on the human pilot. This presents a technical hurdle to expanding the economic applications of flight. It is bound to become a more pressing problem with more than 100 companies trying to bring small (mostly electric) VTOL aircraft ("flying cars") through certification and to market. Replacing the human in the loop with modern robotic systems will become a crucial enabler for whole new markets in aviation and urban transport.
Safety is tightly regulated in aviation and rightly so because it has made flying one of the safer means of transportation. Yet these regulations and the built-in desire for proven solutions also present a barrier to the adoption of new technology. There is no incremental path from uncertified UAV control to certification for personal transport (DAL-A), nor one from current DAL-A level avionics to full autonomy. To justify the absence of a human pilot we need to build a system that is sufficiently deterministic to pass certification, yet can deal with unexpected situations.
Relying on our expertise in robotics, computer vision and machine learning, as well as a thorough foundation in classical systems engineering, avionics and piloting, we have set out to build guidance navigation and control systems that replace and outperform the human pilot on every measurable scale. To autonomously fly in VFR, without relying on extra rules or infrastructure, requires that navigation, guidance and collision avoidance are able to use visual information first and last, alongside other instruments currently available in the cockpit. Our visual systems provide situational awareness and semantic understanding to safely guide any "flying car" from take-off to landing.
Luuk van Dijk, Founder and CEO
PhD in Physics (UvA, RuG). Google (Gmail, Maps, Go), SpaceX (flight software).
Anna Chernova, Co-founder and CPO
BSc in Physics (СПбГУ), mathematical biology (Univ. of Oxford) and bioinformatics (NIMR MRC). Google (Maps). PPL(A) student (30 hrs), PPL(H) license holder (270 hrs).
Andrea Schlapbach, CCO
MSc in Physics/Chemistry (ETHZ). Swiss Re, SIX Group, Hexagon/Leica Geosystems, FLARM (co-founder), SAFEmine COO/CFO (co-founder). EASA LAPL/SPL/FI(S) (over 2000 hrs), UAV expert.
Thomas Kindler, Legal
Dr. jur. (Bern, Switzerland), LL.M. (Georgetown, Washington DC). Legal counsel for siroop AG.
Matthias Kraft, Computer Vision
PhD in Physics, plasmonic systems (Imperial College London). Imperial College London Mathematics department (MCMC samplers and application to machine learning).
MEng Mechanical Engineering (Imperial College London). Rolls-Royce plc (graduate scheme, Development Engineer - civil jet engines), Omnilid (co-founder).
Tatiana Yusupova, Autonomous Flight
PhD in Mathematics (Number Theory, University of York), Basics of Information Security (Moscow State University). Worldpay UK (Software Engineer and Team Security Champion). EASA PPL(A) license holder with SEP type and night rating (140 hrs in Cessna 172).
Armin Burgmeier, Simulation and Integration
PhD in Physics (KIT). DESY (Higgs Search at CMS), Bloomberg. FAA PPL(H) student (100 hrs in Robinson R44 II).
Corentin Perret-Gentil, Deep Learning
PhD in Mathematics (ETH Zürich), Mathematical Sciences Research Institute (Berkeley, CA), Centre de recherches Mathématiques (Montréal).
Zoltan Pillio, Computer Vision
MSc in chemistry (SZTE). Aero Glass (software engineer).
Harleen Hanspal, Systems (Intern)
Pursuing M.Sc. in Robotics, Systems and Controls (ETH), B.Tech in Electrical (IIT Bhubaneswar).
Łukasz Janyst, Autonomous Flight
MSc in Computer Science (Jagiellonian University). Institute of Nuclear Physics PAN (ATLAS experiment at LHC), CERN (Compilers, Massive Data Handling Systems).
Emmanuel Mogenet, Research
DEA/MS Computer Science. Google (Sr.Eng. Mgr, Head of Research), Apple (Sr. Scientist), Sony Pictures.
Ruben Polak, Deep Learning
MSc in Artificial Intelligence (University of Amsterdam), BASc in Aviation Engineering (University of Applied Sciences, Amsterdam). Aiir Innovations (Computer Vision Engineering, Sales), AerData (Technical Consultant).
Sasha Tsvyashchenko, Autonomous Flight
MSc Computer Science (Kyiv National University), Materialise (CAD / 3D printing), Automated Industrial Machinery Inc (CAD / CAM), Google (applied ML for YouTube / Ads /..., ad fraud detection).
Janet Darling, Operations
B.Sc.(Hon). Business Links (Consulting, Ops), IBM (Fin. consulting, Bus. transformation, Project management), CitiGroup (Process optimisation, TQM, SixSigma), iDarling GmbH (Process consulting), ClearSight (Fund of Fund ops), PALA (Mining investment, Ops, HR and IT), Agys (Fintech, Bus. Automation, Market Dev).
Karol Rychwalski, Recruitment
MA in Sociology (Warsaw University of Life Sciences). Roche, Toolbox for HR (cooperating with various tech start-ups).
Grigory Yakushev, Simulation & Modelling
MSc in CS (МГУ), Eagle Dynamics, Google (Calendar, Search), Mediahead, NVidia (Gameworks).
Theo, Emotional Support Animal
Italian sheep dog, anti cat and squirrel defense.
Peter de Lange, Simulation & Modelling
Msc in Mechanical Engineering (TU Delft), Pilatus Aircraft (simulation), Moog Inc. (haptics).
PhD Computer Science, Postdoctoral Fellow (MIT and UC Berkeley), Ass Prof Computer Science (Brown Uni.), Co-founder / Chief Scientist for OpenText, Prof Computer Science (Uni Darmstadt), Dir Fraunhofer IPSI, Dir Eng and co-site Lead (Google ZH), Prof Data Analytics (ETH), Co-founder / CTO 1plusX, Co-director Max Planck-ETH Center for Learning Systems.
Reza Madjidi - Avionics Certification (Advisor)
BSc Computer Science & MBA in Technology Management. ConsuNova CEO (strategy, avionics certification programs). FAA Systems & Equipment DER certification auditor, trainer, consultant. Atego, VP Global Services. HighRely Inc., cofounder. Developer of critical embedded avionics and safety-critical systems.
Your role: to come up with algorithms and code to completely and reliably replace the human pilot, only bound by the laws of Nature and rules of the air. Preferred qualifications and experience: be smart, get stuff done, at least 5 years of hardcore C++ software engineering experience, proven research abilities, and applied engineering skills.
Daedalean became one of the latest recipients of the grant
awarded by European Commission under its European Innovation
Council (EIC) Accelerator Pilot programme. We are one of 75
startups selected from 1800+ applicants, which ranks Daedalean
among the top 4% of European tech startups. The funding is
provided for a two year project and will be spent for the
continuous delivery and launch to the market of our main
technology product related to the autonomous flight control.
Daedalean and Honeywell, Inc. have signed an agreement on joint
testing and technological partnership in developing solutions for autonomous takeoff, landing and GPS-independent
navigation and collision avoidance for GA aircraft and electric vertical takeoff and landing vehicles (eVTOL).
Additionally, Honeywell Ventures has joined the Swiss startup’s pool of investors, the amount and conditions
UAVenture and Daedalean have published a teaser video announcing a new guidance system for professional UAVs. The system is called Magpie and provides unrivaled AI-powered functions – vision based identification of emergency landing sites during flight, as well as visual navigation that helps sustain GPS outages – in a less than 500g package, and completely integrated with the AirRails flight controller: no additional complex setup & coding required. No pre-marking of landing spots is required, Magpie is able to operate without pre-surveying of the area and recognise dynamic obstacles on the ground. The product will be available soon for the users of AirRails, UAVenture’s flight control system for professional drones.
Watch to learn the features of the AI-autopilot for the autonomous flight – computer vision-based flight control – already demonstrated: situation awareness and perception, detecting flying obstacles, GPS-independent simultaneous localisation and mapping, recognizing obstacles on the ground and safe landing advisory. On the roadmap: wire detection and more.
This additional round brings the total amount of the company funding to CHF 11.8 million (nearly $12 mln). The funding round is led by Carthona Capital. Other investing VCs are Redalpine and Amino Capital. Along with them, three former engineering managers of Daedalean’s founder and CEO, Dr Luuk van Dijk, back in his times at Google, are investing as business angels. Dr Harald Nieder, Partner at Redalpine, is joining Daedalean’s board as an observer.
On 8 May 2019 Daedalean and EASA concluded a fruitful 3 day workshop to kick off their Innovation Partnership Contract (IPC).
The project titled “Concepts of Design Assurance for Neural Networks” aims to examine the challenges posed by the application of Neural Networks in aviation,
in the broader context of allowing Machine Learning and more general Artificial Intelligence on board aircraft.
Experts from EASA and Daedalean are bundling their expertise to create concepts and safety standards for the application of
this branch of Artificial Intelligence in safety critical avionics. The project is expected to run through January 2020 when a final report will be presented.
Daedalean and Volocopter conducted joint flight tests on Wednesday 27 th March 2019 at the airfield in Bruchsal (D) in order to test Daedalean’s Visual situational awareness system
(“Raven”) on board an experimental Volocopter.
The tests demonstrated crucial environmental perception capabilities for visual navigation and collision detection in landing and en-route scenarios, using 3 high definition cameras
mounted outside the fuselage of the aircraft, the data from which are fed to the on-board computer vision and artificial intelligence (AI) algorithms.