At the massive Consumer Electronic Show (CES) show in Las Vegas in early January, Bell demonstrated a model-size cityscape with scale flying versions of its Nexus passenger air taxi and Autonomous Pod Transport (APT) operating with Bell’s AerOS urban air mobility operating system.
Calling it a “smart city ecosystem,” Bell president and CEO Mitch Snyder explained, “This year, we’re demonstrating what governing, operating, working, and living in a smart city will look like.”
The Bell Nexus City demo at the CES Mobility Hall was designed to highlight how “mobility as a service” software like AerOS can manage a metropolitan area’s urban air mobility (UAM) activities. Bell intends to offer AerOS, which runs on Microsoft’s Azure platform, to cities to speed up their adoption of UAM capabilities.
Bell has also settled on a smaller version of its Nexus passenger vehicle, with four rotors instead of the six previously shown at CES. The Nexus is designed for all-electric or hybrid-electric power, but is “propulsion-agnostic,” according to Bell, “depending on customer needs.” The four-rotor Nexus, a mockup of which was shown at CES, will initially have a 60-mile electric range, but that could be greater with hybrid-electric power.
At CES, the smart city demo included tablet stations where visitors could interact with AerOS, choosing departure and destination, and then watching in real-time how the flying models interacted. The flying models were not controlled by individuals flying them, but by the AerOS software, which constantly assesses demand across the scale-size city and deploys the vehicles to meet that demand. The AerOS software also takes into account problems that inevitably come up during passenger and cargo flying operations, for example, weather events that might require all vehicles to land immediately. AerOS creates an optimal flight schedule based on goal-seeking optimization algorithms and artificial intelligence to anticipate passenger behavior and desires as determined from the booking engine and the vehicle’s needs for battery recharging to meet the schedule.
There is much more to the battery needs than simply recharging them periodically, Snyder explained. Some flights will be so short, for example, that it isn’t necessary to fully recharge the vehicle’s batteries. Those batteries that get a deep discharge will need more time to recharge, which adds cost to the entire system, so if the software can intelligently schedule vehicles to avoid deep discharges, this lowers the average system cost.
The software also looks at each day’s operations and predicts upcoming demand, then afterward analyzes how that deviated from the original plan for the day. That way, each day, the prediction model improves to keep the system running more efficiently. “We are using goal-seeking optimization algorithms to build the best-fit schedule,” he said.
“We are working on modeling simulation tools now. We need to do better than have a good model, but we have tools in-process to refine and update that without an army of PhD data scientists. This solves the digital backbone need of aerial mobility.”