While Ford's autonomous vehicles are learning to drive on Miami's bustling streets, the company is simultaneously mapping out a business strategy for the driverless future by tapping local merchants for input on its first purpose-built AV.

Why it matters: The U.S. market for AVs will be $332 billion by 2026, according to Ford. About 40% could be for goods delivery including small businesses that are often constrained because they can't afford to hire drivers or buy delivery vehicles for only sporadic use.

The background: Ford is testing a small fleet of AVs — with safety drivers but simulating fully autonomous operation — in hopes of deploying tens of thousands of self-driving vehicles to move people and goods in multiple cities, starting in 2021.

Lessons learned from those tests — about preferences for interacting with driverless vehicles and how the car's interior might be configured to keep food orders separate or dry-cleaning pressed — will influence Ford's AV design and refine its business strategy.

• Today, ride-hailing costs about $2.50 per mile. Cut out the driver and Ford estimates that cost could fall to $1 per mile, boosting demand.
• Add in revenue from other things like fleet management, deliveries and digital content and they think the profit potential dwarfs the 6%–10% margins of traditional auto-making.

“The opportunity of that market is massive. It’s a margin that is like no automotive markets today.”

To ensure a profitable business, the company says it will need to keep its AVs running round-the-clock:

• During peak commuting times, they'll be used mostly for ride-hailing by people looking to complete the "last mile" of their trip.
• During business hours they could be used for delivering packages or food.
• In the evening, they might be turned into roving party buses, ushering fans to a concert, for example.

Details: Ford is working with 7 local Miami merchants to assess the level of demand for commercial deliveries using AVs. The 2 month trial was eye-opening, Lydia and Roland Losas of Giralda Dry Cleaners told me.

• By assigning the AV test vehicle to bring dirty laundry from their four stores to their central cleaning plant, they freed up 2 employee drivers to serve customers at luxury hotels.
• Using an AV for laundry delivery helped them avoid costs like car payments, liability insurance, gasoline, tolls and part-time seasonal help.
• They think it opens the door for new growth opportunities like wash-and-fold services for college students, Lydia said.

The bottom line: Automakers are targeting small businesses as a potential market for AVs that might fill their intermittent needs, allowing them to save on labor and fixed costs.

AV developers are using their test vehicles to collect data on the interactions between vehicles, pedestrians and cyclists, but it is currently limited to specific cities like San Francisco, Sid Misra, CEO of Perceptive Automata, writes for Axios Expert Voices.

The big picture: The cultural norms of driving vary widely from one region to the next. To operate safely and be deployed widely, AVs will need to draw on global data sets that are locally customized and continuously updated to account for both changing behaviors and new modes of transportation, like electric scooters.

Jaywalking is one example of these regional differences.

• In Japan, only 7% of people cross the street against a red light, as opposed to 67% in France.
• In big U.S. cities like Boston and New York, jaywalking is so commonplace that police officers rarely enforce laws against it.
• In Germany, most people obey the Ampelmännchen (“little traffic light man”), whereas pedestrians in the U.K. tend to be more defiant.

What's needed: Massive amounts of vehicle data have to be processed in real-time for vehicle path planning — without cloud connectivity — to ensure AVs perform safely and smoothly. At the same time, AVs also act as probes, capturing anomalous road behaviors that are then identified and processed through cloud-enabled feedback loops.

• Once an anomalous situation is detected, it's flagged and transferred wirelessly to the cloud for offline processing, including tuning of the machine learning models used on the vehicles.
• These tuned models are transferred back to AVs via ongoing system updates, ensuring they continue to improve over time.

The bottom line: Though it might be a while until self-driving cars have spread across the world, technology developers that master scalable solutions for these regional differences will pull ahead of the competition and make for a smoother transition of AVs onto human-dominated roads.

Go deeper: Read the full post.

Brookings Institution's Adie Tomer unpacks for Axios a new study he co-authored that found 9.5 million people in transportation-related jobs will be impacted as digital technology, including AVs, diffuses into transportation products and services.

Why it matters: Every week brings announcements around technology breakthroughs, capital infusions and new consumer-facing services. But that news often distracts from a pressing need to develop the workforce that will create, manage and maintain AVs, in addition to overseeing the digital services and built environment around them.

Where it stands: The public conversation so far has been either too narrow and focused on drivers alone, or too wide and swept up in all workers who drive as part of their professional routine.

Reality check: The pool of workers who will be affected by digital mobility — from professional drivers to employees in the motor vehicle and transportation support industries — is bigger than sizable sectors like finance, real estate and administrative services.

• And that's without counting the many people working in the computing and telecommunications industries on the R&D that will power new transportation technologies.
• Critically, the transportation industry represents a core component of work in every state — not just those that manufacture vehicles or have a higher proportion of truck drivers. No state employs less than 5% of its workforce in mobility-related occupations.

The bottom line: AVs and digital mobility platforms are coming to market fast. Preparing the workforce for them will require partnerships among cities and states, educational institutions, civic organizations and private employers — and a clearer understanding of the jobs and skills that will be most in demand.

I want to hear — and share — what you're reading about AVs. Send me a link to an article and your expert analysis of why it matters to [email protected].

Location, Location, Location? How autonomous vehicles will shake up the commercial real estate market. (Alan Ohnsman — Forbes)

• The big picture: The most desirable office space used to be located close to talent, urban amenities and transportation. But if AVs are done right, people might not mind commuting longer.

What's your name again? Waymo to launch its commercial robotaxi service next month under a new name. (Tom Randall — Bloomberg)

• My thought bubble: Why does Google, er, Alphabet, er, Waymo, keep coming up with new names? I barely got used to Waymo for the self-driving car unit.

AV dealmaker: Michigan's governor has more deals with foreign countries than President Trump. (Joann Muller — Axios)

• Why it matters: By creating research partnerships with foreign countries, Gov. Rick Snyder aims to harmonize global standards while reasserting his state as the center of AV development.

This week I traveled to Miami for some seat time in Ford's autonomous test vehicles. I rode in 3 separate Ford Fusions, each with a different pair of safety drivers up front.

Background: Miami's streets can be hectic and confusing, between random lane jogs, construction detours and occasional flash floods not to mention jaywalking tourists and wrong-way bicyclists.

Details: The AV's safety drivers kept their hands and feet ready to react, but only once did they opt to take control.

• At about 20 mph, the car was starting to change lanes to the right but aborted when it detected another car traveling up from behind at a higher speed.
• It moved back into the current lane and let the other car pass on the right.
• The driver then took control to execute a quick double-lane change to stay on the intended route.

Ford and its AV tech partner, Argo AI, are trying to master "naturalistic driving" — which means not being overly cautious so as to annoy other drivers. The cars mostly succeeded.

• In one instance, the AV had to make an unprotected left turn across two lanes of heavy traffic.
• The car waited for a natural gap to turn left, leaving just enough space for two more aggressive drivers to cut in front from the right, nosing into oncoming traffic.
• It then waited until it was safe to turn, but then had to stop in the middle of the intersection to let pedestrians cross.
• Behind us, a Miami driver laid on the horn.
• The car? It was unfazed.

The bottom line: All were relaxing, uneventful experiences, which says a lot about how close we're getting to the driverless car era.