The Bell Nexus concept vehicle is shown at the Uber Elevate Summit June 12, 2019 in Washington, DC, ... [+]
Uber’s data scientists are hustling to create simulated air mobility systems for specific cities ahead of the 2023 launch of Uber Air.
The company has said it will roll out the service in Dallas, Los Angeles, and Melbourne that year, but a slide displayed by an Uber data scientist at Northwestern University Tuesday suggests Uber is planning service for many other cities, including London, Delhi, Paris, Buenos Aires, Rio de Janeiro, San Francisco, Mexico City, Mumbai, New York, Santiago, Bogota and Sao Paolo.
For each city, the scientists are plumbing data on point-to-point travel for different modes of transportation, said Zhuyun Gu, a data scientist for Uber Elevate, comparing those data to Uber’s own trip data and modeling the results over different time spans in the future.
“We’re able to construct very detailed Uber Air itineraries,” Gu said. “And by combining different sources together we’re able to simulate the future for air trips (and) compare and assess different cities to see how juicy different markets are. And we’re able to construct the candidate locations for those sky ports in different cities, and ultimately thus is what will empower us to select our markets.”
Uber plans to locate skyports on existing rooftops and parking lots and build multi-level mega-skyports at hubs. They’ll be equipped with electric vehicles that take off and land vertically like helicopters, but with multiple rotors they more closely resemble drones.
“We are currently working with planners to design these electric takeoff and landing vehicles. These vehicles can travel at a flight speed of 150 mph and a range of 60 miles, and unlike traditional helicopters these vehicles are using a distributed electric propulsion technology. Therefore these vehicles can travel much quieter and much safer than traditional helicopters.”
Initial plans call for four passengers and one pilot, but Gu hinted the vehicles will eventually be autonomous.
“During the initial launch we envision these vehicles will be piloted,” she said at the Northwestern University Transportation Center’s annual Technical Workshop. “For most riders your experience will be very much like Uber Pool. You will have two to three co-passengers in your same vehicle.”
But before any of this can happen, Uber has to “know what are the potential barriers to this new mode of transportation (so) we can work our communication plan accordingly in collaboration with the government sector.”
While acknowledging the policy challenge, Gu’s team is focused on the economic. It seeks to answer three questions before the launch:
1 Which Cities Demand Airborne Mobility?
Uber’s first focus will be to determine which cities are good markets for airborne mobility, Gu said, and “how juicy” the market is in each city.
“Is it going to be a niche market for the high-income few or actually can it go massive in the long term?”
For each city, Uber is analyzing how residents balance time savings and price, Gu said. The company will use that data to estimate supply and demand over different time horizons.
2 Where Should Skyports Be Located?
By simulating future trips in each city, Uber hopes to determine the best locations for its skyports. They will likely appear first in the urban core, Gu said, and then radiate to outlying areas.
The team is modeling skyport locations not only at launch, but long after the system develops in each location.
“We would like to know what this evolution looks like in different cities. And we would also like to have an in-depth understanding about what is the maximum throughput expected with each of those skyports, so we can design our skyport layout accordingly.”
3 What Should The Vehicles Look Like?
Uber has floated several prototypes of its vehicles, but its system simulation gives the company an opportunity to design the vehicles to fit anticipated use, Gu said.
“What is the optimal range and speed associated with different markets?” she asked. “And what is the required fleet size for different markets?”
Once the scientists determine those design parameters, Uber will test the system’s profitability within those parameters and run experiments to optimize system design.
“At Uber we have developed this system simulation over the past two years from scratch,” Gu said, “so it is composed of algorithms to understand or answer these three main category questions.”
Uber Elevate data scientist Zhuyun Gu displayed this concept image of an Uber skyport at ... [+]
