We Want to Invite You to Build Flying Cars with Us, No Kidding

MasterPa
Generalized Intelligence
8 min readFeb 26, 2019

In the past five years, autonomous and electric cars are trending around the globe. Everyone seems to think that the future will be filled with autonomous electric automobiles. However, looking at the future described in sci-fi movies, we see a future with cars that fly. If we ask a tech enthusiast 30 years ago about the future in his mind, he probably will not think of autonomous electric cars, but he will most definitely think of flying cars. After 30 years, we could not agree more when Adam K. Raymond said, “sure, we’ve got tiny supercomputers in our pockets and countertop speakers that can tell jokes, but our cars still drive on four wheels.”

We don’t believe in a future without flying cars. We believe we are empowered to reach for the future promised by The Fifth Element and Blade Runner. Instead of waiting, we would rather expedite the coming of flying cars. This is what motivates us to develop GAAS, an open source Generalized Autonomy Aviation System. We would like to allow more people to take parts in this flying car revolution.

Why do we need flying cars?

Imagine flying from San Francisco to Downtown St. Jose in just 15 minutes, while driving usually takes 2 hours. Everything will be changed, especially how we feel about distance. We will be able to spend more time on lifestyles rather than on commutes. Emergency medical services will be able to save more lives without traffics. People will also be able to have efficient transportation in under-developed areas. What we described here is a future that will only be achieved by the flexibility, resilience and low resource intensity of flying cars.

Do we really need a reason to build a future with flying cars? Flying cars ARE the single, most valid reason of all.

Flying is the most romantic dream of mankind, ever since we stood on this planet. One hundred years ago, the pioneers of the industry have defeated gravity. For the first time in human history, mankind was able to soar in the sky. However, during the past 100 years, the way we fly has not changed much. Flights are still mostly for long-distance transportation but are rarely used for day to day commute. Flying cars can change all this. Flying cars can allow us to commute in the sky.

Testing GAAS at low light

Three Major Challenges

Before getting into GAAS, I would like to talk about the three challenges flying cars are facing:

1. Level of Awareness and Acceptance

2. Safety of Aviation Vehicles

3. Technology Barrier for Building Flying Cars

The First Challenge: Level of Awareness and Acceptance

We asked ourselves, why people today don’t have the urge to buy a flying car, as they do to autonomous electric automobiles?

Currently, the majority of people are still in the process of learning and understanding what drones can do. When I am flying my DJI Mavic, I often get the “what is that” look from the street. People are curious because drones are not commonly seen yet. Whether it be industrial grade drone or commercial drone, it is still not a product of main-stream (though more main-streamed than 5 years ago). This is definitely not a market that is ready for flying cars.

Because of the high safety level required by passenger-carrying vehicles (which we will talk about in the next section), thousands and millions of flights must first be proven to be safe on drones, before we would consider having a person taking the ride. For this reason, Unmanned Aerial Vehicles (UAVs) are not only the technical foundation of flying cars but also a symbol as well as a preacher for flying cars.

Therefore, the first challenge is to prepare the market for flying cars through democratizing UAVs.

The Second Challenge: Safety of Aviation Vehicles

Safety is always the first thing that comes to mind when we think of any passenger-carrying vehicles. How do we know for sure that an aviation vehicle is safe enough to fly on top of Downtown Manhattan with passengers?

Therefore, the second challenge is to ensure the safety of the flight.

The Third Challenge: Technology Barrier for Building Flying Cars

Building flying cars is a collaborative project that requires specialists from different disciplines. The technology barrier is what prevents specialists from different field to work collaboratively. The lower the technology barrier is, the more talents the industry can attract from different disciplines.

The three major components of flying cars are the vehicle (hardware), the engine (engineering) and the operating system (software development). A lot of the hardware and engineering design barriers can be lowered by taking advantage of the well-established hardware manufacturing facilities and supply chains. It is said that it only takes 150k USD to build a flying car (hardware only) in China. However, the advantage is valid with one pre-requisition: the software.

The software, or aviation operating system, is the brain of flying cars. It is also the component that has the highest technology barrier. A flight control specialist cannot develop an aviation system on their own, while a full-stack developer does not understand drone physics, flight control protocols etc.

If we look at the history of UAVs. Once upon a time, drones were still being referred to as model aircraft. Model aircraft was only appealing for a very small group of model hobbyists at that time. Ever since APM democratized the flight control technology, within 10 years, we have seen a number of uprising UAV companies such as DJI, 3DR, and Ehang. The drone community has been growing rapidly ever since. We expect the same trend for flying cars.

Therefore, the third challenge is to lower the technology barrier in software development, in order to take advantage of established supply chains to build the hardware.

What is GAAS?

Generalized Autonomy Aviation System (GAAS) is an open source project dedicated to autonomy flight and operating control system.

Our short-term goals are to unleash the potential of UAVs by making it fly autonomously and allow more developers to work on UAVs. The long term goal of GAAS is to expedite the coming of passenger-carrying aviation vehicle, possibly in the form of flying cars.

GAAS Beta includes features such as SLAM, collision avoidance (fly around obstacles autonomously), route planning and flight control. GAAS supports UAVs as well as manned rotors and helicopters and will be supporting VTOL and eVTOL in the future.

Outside testing of GAAS

How does GAAS tackle these challenges?

1. Making Drones Ubiquitous with Autonomy Aviation System

Drone technology has been around for several decades, why hasn’t it been something ubiquitous, like smartphones or personal computers? The reason is the Unmanned Aerial Vehicles are not in fact “unmanned”. Piloting UAVs requires skills and expertise. Flying a drone from one room in building A to another room in building B is considered fairly difficult for most pilots. In fact, in the industrial settings, on average it takes 5 skilled pilots to fly one drone. The lack of skilled pilots is the main obstacle that stands in the way of democratization of aviation vehicles.

GAAS, the autonomous aviation system, allows drones to fly autonomously, without the supervision of pilots. Piloting skills and amount of pilots are no longer a constraint to the drone market. The more mature the drone market is, the easier it is for flying cars to enter.

2. Making Aviation Vehicles Safer with Autonomous Flight and Accumulated Drone Flight Data

Improving the safety of aviation vehicles can be done in two steps: implementing more safety assurance mechanism, and collecting enough data to support these mechanisms work.

In Fast Forwarding to a Future of On Demand Uban Air Transportation, Uber Under the Hood has pointed out that more than half of flight accidents in Alaska are caused by human errors in adverse weather conditions. A good supplement to human pilots is to utilize automated collision avoidance and SLAM (Simultaneous Localization and Mapping) to improve safety.

In order to validate the performance and stability of the system, we will be concentrating on implementing GAAS in drones, not flying cars, because we need time and data to ensure the system is safe for passenger-carrying aviation vehicles.

More important than everything else, GAAS team will be continuously focusing on improving the safety of aviation vehicles.

3. Lowering the Software Technology Barrier with Easy-to-Use Autonomy Aviation System

GAAS is a user-friendly software and hardware (2 in 1) system. Users will be able to implement the system with minimal installation effort, without the trouble of learning flight control, SLAM, computer vision etc. When the software is in place, experts with the manufacturing background can look into utilizing existing supply chains and facilities to build the hardware of flying cars. People in the automobile or smartphone industry will also be able to take part in the flying car revolution.

GAAS hardware’s prototype, you could use your own hardware instead of it

The Future Is Around the Corner

Of course, the future may not be as ideal as we imagined, but I have found some interesting historical data that may help us keep the spirit high:

In 1885, the first automobile was invented. On August 19th, 1897, the first taxi was driving on the road. In 1900, there was already a well-established taxi community consisting of 100 taxies in New York.

In 1903, Wright Brothers invented the first airplane. In 1909, mankind has already flown across the English Channel.

The future is closer than we think.

Join Us at GAAS

This is GAAS, an open source project dedicated to autonomous flight and operating control system for futuristic aerial vehicles. We would like to invite people who share the vision to join us and expedite a future with flying cars.

If you are interested in learning more about GAAS, here is how you could contribute:

Developers: You may find us at https://github.com/generalized-intelligence/GAAS on GitHub. GAAS Beta is licensed under BSD-3.

We are more than happy if you could simply follow through the introductory tutorial and let us know about your testing environment and result. Here is the tutorial: https://gaas.gitbook.io/guide/

We are grateful if you are kind enough to make any contribution to the repo.

Investors: We are happy to talk about the project at adamwang@accelerator4.ai.

Media: We are happy to talk about the project at adamwang@accelerator4.ai.

Everyone Else: Follow us on Medium and Twitter(@GAAS_ooo) for our latest updates!

Let’s bring flying cars back to the future!

--

--