Wayfinder’s Origin Story

All great teams need an origin story.

Wayfinder’s story began in 2016 as part of the team building Vahana, one of Airbus’ eVTOL demonstrators originally housed within A3 by Airbus. At first, we planned to integrate key components to build a working detect and avoid system for the Vahana Alpha aircraft. The system would show that it was possible to detect obstacles in the air and on the ground without a pilot to see them.

Figure 1: Team preparing for a data collection flight.

Figure 1: Team preparing for a data collection flight.

Before long, it became clear that no pure off-the-shelf system could be a solution for our targeted demonstration. We therefore decided to build a team to do much of the work internally. It was also clear that by doing so, Airbus as a company could benefit from the development experience and establish in-house expertise.

Over the course of 2017 and 2018, we designed and built solutions to two key problems for  self-piloted vehicles:

  1. How to detect and avoid uncooperative aircraft.

  2. How to make sure the desired landing site is free of dangerous obstacles.

The resulting detect and avoid system is now integrated into Vahana Alpha - a full-scale eVTOL vehicle that our colleagues at Vahana have been testing at the Pendleton UAS Range in Oregon since early 2018.

Figure 2: Vahana AlphaOne in-flight December 2018.

Figure 2: Vahana AlphaOne in-flight December 2018.

In mid-2018, we further demonstrated the full capability of our system in testing on surrogate vehicles (small multi-rotor UAVs) and in hardware-in-the-loop simulation of Vahana.

Figure 3: Surrogate testing of our landing zone evaluation module.

Figure 3: Surrogate testing of our landing zone evaluation module.

Figure 4: Surrogate testing of our drone detection algorithm. This example shows our success at solving a challenging problem, where the environment contains difficult clutter and noise both on the ground and in the background clouds.

Today, the Vahana flight test team has completed more than 60 test flights of the full-scale vehicle. During each flight, our system logs data that we will use to refine and improve the products that we are building for future vehicles.

The technologies and processes we are building have huge implications across Airbus. With strong support from A3 by Airbus and our colleagues in Europe, we launched our own project, Wayfinder, to build the team to deliver that impact. Here at Wayfinder, we embrace the innovative spirit of Silicon Valley, and we plan to solve the most challenging problems in developing scalable, certifiable autonomy systems to power self-piloted aircraft applications throughout Airbus, from small urban air taxis like Vahana to large commercial airplanes.

We have already expanded the range of challenges we can tackle to include a larger Airbus initiative developing solutions for Autonomous Taxi, Takeoff, and Landing (ATTOL). Stay tuned for more details about ATTOL in upcoming blog posts!

There is nothing more exciting than watching a brand new vehicle that you had a hand in building fly for the first time. After almost two years of hard work with many ups and downs (so to speak) I was lucky enough to experience the first flight of Vahana in January 2018, and it was a highlight of my career to date. Now that Wayfinder has launched to bring our systems to vehicles across Airbus, I look forward to many more of these highlights in the near future. If you’d like to help us get to our next first flight, please apply to any of our open positions or email us at hello@airbus-sv.com!

  • Alex Naiman, Director of Engineering, Wayfinder

Welcome to Wayfinder!

In a world where annual air travel volume is expected to double by 2036 (to 7.8B passengers from 4B in 2017) and the number of pilots cannot scale quickly enough to meet that demand (450K - 870K pilots are expected to be needed), adaptable, autonomous systems will be required to provide an alternative solution.

At Wayfinder, the latest project to launch out of Airbus’ Silicon Valley Innovation Center, A3, we aim to develop scalable, certifiable autonomy systems to enable self-piloted aircraft applications. These systems — vital to the busy future of air transportation — will be usable across Airbus, from small urban air vehicles such as air taxis to large commercial airplanes for which teams are already investigating single pilot operations.

Figure 1: An Airbus A320neo, from the world’s most advanced and fuel-efficient single-aisle aircraft family.

Figure 1: An Airbus A320neo, from the world’s most advanced and fuel-efficient single-aisle aircraft family.

This project began within Vahana, an early A3 project and current Airbus Urban Air Mobility (UAM) eVTOL demonstrator that has recently completed its 50th successful test flight. The Wayfinder team developed the sense and avoid system that Vahana has been testing since the project began at A3 in 2016. Over those years, it became clear that the sense and avoid system (specifically for the developed technologies and processes) could have greater impact across the Airbus organization especially with respect to commercial aircraft.

We launched Wayfinder to meet these broader needs and did so within A3 to take advantage of the unique ecosystem of Silicon Valley — a place where where speed of execution, agility, talent, and automotive synergies combine enabling Wayfinder to excel faster. Our ambition is to provide both near-term and long-term value to Airbus and, in doing so, contribute to the organization’s ongoing position as the leading aerospace company, as well as one practicing innovation in unique and exciting ways.

Where We Stand

Our primary focus today is on developing a common set of software and hardware that we can use in a scalable way for a range of aircraft. To actually create such a system, we are concentrating on three main categories that need to be explored for a future with safe, self-piloted flight: Technology, Data-Driven Development Processes, and Safety.

Figure 2: Photorealistic simulation allows Wayfinder to rapidly train and test neural networks on large, scalable, and highly customizable datasets. Here we simulate a forward facing camera on final approach to runway 9 at San Diego International Airport. Our neural network architecture is able to detect runways and runway markings, as well as the remaining distance to the runway and vertical and lateral deviations from the nominal approach flight path.

Figure 2: Photorealistic simulation allows Wayfinder to rapidly train and test neural networks on large, scalable, and highly customizable datasets. Here we simulate a forward facing camera on final approach to runway 9 at San Diego International Airport. Our neural network architecture is able to detect runways and runway markings, as well as the remaining distance to the runway and vertical and lateral deviations from the nominal approach flight path.

Technology

For autonomous systems the technology is divided between software and hardware.

  • Software enables the aircraft to perceive the environment around it and decide how best to proceed. At the core of our software are perception methods based on both machine learning algorithms and traditional computer vision. We use the best techniques that have been developed for other applications such as image recognition and self-driving cars, and expand them to the requirements of autonomous flight. We also develop decision-making software to safely navigate this world that we sense around the aircraft.

  • Hardware includes the suite of sensors and powerful computer required to feed and run our software. We are partnering with the leaders in autonomous vehicle sensors and computing to mold their products to fit our needs. In particular, we are bringing the expertise of Airbus in certifying aircraft and their systems to help our partners meet the rigorous requirements on hardware to ensure safe and reliable operation in the air.

Figure 3: Experimental sensor configuration on Vahana Alpha demonstrator vehicle, including cameras, radar, and lidar.

Figure 3: Experimental sensor configuration on Vahana Alpha demonstrator vehicle, including cameras, radar, and lidar.

Data-Driven Development Processes

Much like flight testing of new, demonstrator aircraft, we anticipate that the development of autonomous aircraft will be an iterative process. Starting with an application of the autonomous system in a limited environment, we will train the system to fully understand that environment, and then expand the environmental envelope over time. Our iterations will include a cycle of massive data collection followed by system testing, subsequent development, and then verification of system safety. Once each new iteration has been verified, it will be deployed to start the next cycle of data collection and continuous improvement of the system performance.

Safety

In keeping with Airbus’ long history of safe and certified aircraft, safety is at the core of all our activities. Everything we work on must pass rigorous testing to ensure what we create is not only revolutionary but certifiable and secure. Our dedication to safety coupled with our accelerated pace of innovation is Wayfinder’s biggest differentiator in the increasingly crowded urban air mobility market.

Personally, my background is in autonomous cars and ground-based robots, but thanks to the sheer wonder inspired by the thought of extending autonomy into the third dimension I am now fortunate enough to lead this project. When I think about how we will impact how humans travel and, as a result, interrelate, I cannot wait to get to work everyday. We’re actively looking to add new members to the Wayfinder team who embody this same spirit.

We’re working to develop products that have never been built before, and are looking for team members who are inspired by this challenge and seek to build the future, not just contribute to the present. If that sounds like you, please apply to any of our open positions or email us at hello@airbus-sv.com.

  • Arne Stoschek, Project Executive, Wayfinder