Last year was a “banner year for autonomous vehicle technologies”, according to Michelle Avary, Project Head of Autonomous and Urban Mobility at the World Economic Forum (WEF).
As the WEF’s annual meeting in Davos draws to a close, the organisation believes that the 2019 autonomous vehicle (AV) sector has the potential to surpass 2018 – as you would hope in these early days for the industry. If it didn’t, something would have gone wrong.
Evidence for the WEF’s upbeat assessment comes from Alphabet’s AV division Waymo beginning public services in Phoenix, Arizona, and GM’s Cruise division planning a public launch in San Francisco.
Autonomous taxis will certainly grab the headlines this year, not least because ride-share rivals Uber and Lyft are filing for IPO. Both are exploring driverless systems.
Last year, Japanese car giant Toyota invested $500 million in Uber, so it (and other giants of the traditional automotive sector) will want to see the promise of payback. The two companies announced that they were extending their technology collaboration to bring autonomous mobility as a service (‘AutonoMaaS’) to market at scale.
Heightened expectations will certainly characterise 2019, and that spells both opportunity and risk.
Onto the superhighway
Not everything will be as smooth on the road to autonomy as Avary appears to suggest. With two fatal crashes last year involving Tesla and Uber vehicles running under software control, public trust is at a tipping point. At least in the US.
Last year, a report from the American Automobile Association (AAA) found nearly three-quarters (73 percent) of US drivers saying they would be too afraid to ride in a fully self-driving vehicle, up from 63 percent in late 2017. Meanwhile, nearly two-thirds (63 percent) of US adults said they would now feel less safe sharing the road with an AV.
Of particular concern for automakers and other tech innovators was that the biggest fall in consumer confidence was among millennials. The percentage of young people who would be too afraid to ride in a fully self-driving vehicle increased from 49 to 64 percent, said the AAA.
Megan Foster, AAA Director of Federal Affairs, said that while autonomous vehicles are in their testing phase, there is always a chance that they will encounter a situation that challenges even the most advanced sensors and software.
While safety-driver error is known to have been a factor in the fatal crash in Tempe, Arizona, last March – in which a woman was struck and killed by an Uber Volvo – it was also established that the test car’s self-driving system failed to identify the pedestrian from onboard sensor data until it was too late.
More, the Volvo’s own safety systems had been disengaged by Uber’s technology.
These stories and the cultural challenges of US car ownership aside, what issues have grabbed the attention of the WEF as self-driving vehicles move off the development hard shoulder and onto the highway?
Three themes will dominate the year, according to Avary. The first is new types of design – or form factors, she says.
Cruise Automation is partnering with Honda Motors to develop a purpose-built autonomous vehicle. “There is no reason to believe these purpose-built AVs will resemble the vehicles we currently see on the road,” Avary writes in a WEF report published to coincide with the Davos event.
“While current AV fleets are retrofitted mini-vans, SUVs, golf carts, all-terrain vehicles, and shuttles jammed with AV technologies, future vehicles will be specifically designed to meet the needs of a particular market segment.“The first AVs on the road in meaningful numbers will be owned by service providers operating a fleet in urban and suburban areas. Focusing on the transportation of people, not goods, we can assume that the vehicle will be shared, capable of seating four people, like a traditional taxi or up to six people in a shuttle form.
“Each vehicle will be highly utilised. These vehicles will be electric, likely optimised for lower speeds, and will be fully, wirelessly connected. Connectivity will be cellular-based and used for over-the-air software updates, remote operations, and real time in-cabin monitoring for safety and notifications.
“Headlights won’t be necessary for the vehicle to see, which could allow for softer, more pleasant lights to be deployed.
“We don’t know if the vehicles will have exchangeable batteries or will be conventionally charged, but we can assume they will utilise company-owned and/or maintained charging facilities.
“We would expect AVs to report on traffic conditions, speed of travel, number of stops, and number of passengers. Over time, we will expect AVs to get data directly from the city, including road closures, signal changes, and lane changes.”
Avary predicts two types of on-demand transport developing as the market becomes established: luxury and utilitarian. A luxury service could offer more entertainment and networking options, as well as comfort.
“It’s possible a more luxurious AV ride service would have greater restrictions on sharing, or not be shared at all,” she says. “On the other hand, a more affordable, highly utilised AV would be built for sharing, would be more durable, and less likely to have entertainment features. It is possible that AV service providers will focus on short trips in a city and move at lower speeds.”
Meanwhile, the US is sweeping aside regulations and redefining the concept of ‘the driver’. The aim is to spur the testing and uptake of driverless technologies in America at a time when China is in US policymakers’ headlights.
But for AV purpose-built vehicles to be street legal, regulations in the US will have to evolve beyond simply allowing vehicles to be built without a steering wheel, explains Avary.
“Regulatory change will take time, and lobbying efforts could be an interesting indicator of what commercial interests are planning. Companies will be issuing petitions and making the case for their particular engineered solution.”
At least 47 cities worldwide are piloting AVs. As a result, regulations in each of these places will have to change to accommodate the building, manufacturing, and operation of purpose-built vehicles.
As AVs mature beyond autonomy Level 3 (advanced driver-assisted systems) and move fully into self-driving features, international guidelines are being updated to address remote operation, as well as licensing and permit changes.
China, the EU, New Zealand, Canada, and other countries are joining the US in passing legislation or working on frameworks to move beyond allowing AV pilots in constrained areas. Such guidelines include rules on occupancy and reporting requirements, electrification, and considerations about land use and transit planning.
Designing, testing, and manufacturing a new vehicle is expensive – as Uber has found, with some investors calling for it to spin out its driverless division to underpin the future profitability of the main ride-share business.
Some companies will seek to minimise risks by partnering with others, which is one reason why Avary expects to see more mergers, collaborations, partnerships, and consolidation in the AV industry.
While numerous automaker and AV partnerships have already hit the headlines in recent months – drawing in giants such as Ford, Jaguar Land Rover, Apple, Honda, Toyota, Mercedes, and more – less well known are the multitude of companies making the supporting technologies, such as sensors.
“At the last check, there were something like 53 LiDar companies in California,” says Avary. “I’m not sure the market can sustain 53 LiDar companies. At some point, the technology stack options will narrow and clear choices will be made.
“Similarly, the choice between creating a highly integrated, in-house solution will be weighed against the more traditional strategy of integrating multiple tier-one suppliers.”
So expect to see more collaborations and consolidation in 2019. “We’ve been hearing rumours about VW and Ford joining forces to develop AV solutions. Maybe that announcement will be forthcoming in the next few months,” she says.