SEL, A little secret about the future of transportation

Photo by Denys Nevozhai on Unsplash

Traffic. We have all been through it. But while this daily occurrence may seem like an annoying obstacle to your next destination, this contributes to a global problem. Here’s our solution. As of 2018, 8 billion tons of Carbon dioxide were emitted from transportation, including planes, buses, and cars. 75% of all global emissions as of 2018 were emitted within cities, 90% of which are coastal, and two-thirds of all energy is used within cities globally. In 2016, San Francisco reported that 42% of its carbon emissions came from cars and congestion emissions. Not that bad, right? Except that a report from CNBC demonstrated that cities underestimate their carbon emissions by a whopping 20%!

Traffic congestion plays a major role in transportation emissions. A study by The University of California at Riverside shows that just moving from 35mph to 55mph can drastically reduce carbon emissions from all vehicles from 166 tons to 146 tons within a Southern California highway.

Some companies have already begun to introduce autonomous VTOL, or electric vertical takeoff and landing. The only issue is, they have low approval and are limited by their current fuel economy, which means there is a large cost in maintaining infrastructure for a non-existent demand until 2040.

While Hydrogen fuel cars are also an alternative, a study by the University of Toronto shows that 90% of all car emissions are produced from the bottom 25% of fuel economy cars. In other words, Hydrogen fuel cars can lightly reduce emissions, but can not reduce congestion.


Here’s the idea: a hydrogen-powered Vertical takeoff and landing (VTOL) flying car. The model is estimated to hold up to 7 people during a single ride and can travel over 70 miles within the VTOL form. Meaning there can be vertishops, (or landing sites) spread across further distances, reducing cost. Likewise, the hydrogen fuel car can go 300 miles in a single charge according to automotive technologies, as compared to 200 with an electric battery.


Oh yeah, did I mention you don’t have to be a professional pilot to operate the machinery? It is a completely autonomous car-plane hybrid. SEL will be using a series of AI models which will be used to conduct autonomous processes.

NLP (Natural Language Processing)- NLP will be used to understand ATC to make correct interchanges and altitude changes

CV (convolutional layer)- SEL will have 15 surrounding ultrasound sensors and cameras which will be used to detect other vehicles and surroundings

RNNs (Recurrent Neural Networks)- RNNs will be used to process data from ATC like other aircraft and their altitudes throughout airspace, restricted airspace. Also mathematical calculations to determine where to land and to merge back onto the road if entering the airspace.

CNN's (Convolutional Neural Networks), to recognize landing pads and obstacles.

It will use Natural Language Processing (NLP) to understand the ATC and correct interchanges and then make altitude changes. An amazing overview of NLP can be found here:

In order to give the SEL car the ability to move and maneuver properly and effectively, it will use a CV AI to modulate the different cameras and sensors around the vehicle. The cameras and ultrasonic sensors are used to detect other vehicles and surroundings.

Convolutional Neural Networks (CNN) will also be used to detect landing pads and obstacles. CNN's are incredibly effective in image, speech, and audio input recognition. They work by utilizing layers of nodes. Each of these layers contains an input layer, one or more hidden layers, and an output layer. Each node connects to another and has an associated weight and threshold. If the output of any individual node is above the specified threshold value, that node is activated, sending data to the next layer of the network. Otherwise, no data is passed along to the next layer of the network.

The SEL car will also feature Recurrent Neural Networks (RNN) to process data from ATC like other aircraft and their altitudes throughout airspace, restricted airspace. It will also use mathematical calculations to determine where to land and to merge back onto the road if entering the airspace. RNNs are types of artificial neural networks which use sequential data or time-series data. These deep learning algorithms are commonly used for ordinal or temporal problems, such as language translation, natural language processing (NLP), speech recognition, and image captioning; they are incorporated into popular applications such as Siri, voice search, and Google Translate.


To ensure safety from cyberattacks, i.e hacking ODB-II ports (Bluetooth), there are special precautions on the AI made to follow safety protocols if something goes wrong. Our AI will be acquainted with an internal machine learning model called ElasticSearch that logs and analyzes data while it is stored and can easily detect malicious activity. As SEL data logs are being stored within the system, ElasticSearch can make an algorithm to detect malicious activity easily.


In order to effectively construct and manufacture the product, there are three general considerations:

Propulsion systems and fuel:

SEL uses reversible hydrogen fuel cells, which means that they have the same fuel economy as a standard fuel cell vehicle, which already outperforms electric cars by 100 miles according to AutomotiveTechnologies, except that it can be produced using solar and wind power. For review, hydrogen fuel cells release heat and water as byproducts, making them environmentally friendly. A survey in 2017 shows that a majority of auto executives believe hydrogen power cars will outcompete electric cars. VTOLs using hydrogen power already exist, and generally outperform average electric batteries in distances past 50 miles. The propulsion system will follow a relatively similar model to typical hydrogen power VTOL (electric vertical takeoff and landing).

Diagram of hydrogen fuel cell propulsion systems


The system is set up as an autonomous taxi, where users download an app, choose their point B, and await their incoming VTOL. While SEL travels, there will be vertishops every 40 miles, each of which will cost 30,000 a year. Because SEL is also a car, it can easily land in parking spaces with its rotational wheels and vertically land. The vertishops will include landing spaces, repair stations, hydrogen fuel stations, an operator, and traffic managers. Local gas stations can also be available if they have hydrogen fuel stations available. SEL can vertically land as a car and recharge. (It’s the equivalent of Tesla Supercharger stations)

While flying in urban areas for VTIK may be around 2.5USD/km, it is estimated that a lot of these costs will fall dramatically for optimization, i.e replacing platinum anodes with nanoparticles and improving cybersecurity.

Consumer standards:

Let’s be honest, if you told someone to get on a vehicle to fly at low altitudes, piloted by a computer, they would be scared. My computer can’t even run zoom properly.

According to a Nasa literature review, these are the optimal standards for passenger journeys:

Each of these standards is integrated into the actual autonomous systems, and issuing a command means that SEL can not exceed any of these limits when flying.

Breaking points for VTOL customers

Also, cloth seats, as well as an insulated blanket, will optimize the customer experience.

In Summary

SEL is an environmentally friendly VTOL system that is convenient and efficient for traveling from city to city, or just to avoid traffic. Download a mobile app, wait for your vehicle, and hop on.

SEL is a way to take the crow’s path, literally.

NOTE: 2in1 Tires developed by Goodyear

Made by Andy Jacoby, Alejandro Leyva, Venkat Yarlagadda, and Mahnoor Sargana

Visit to learn more about the product

Innovator at (TKS) and Founder of NeeroSolutions