When Kyle Vogt, the founder of General Motor’s self-driving car subsidiary Cruise, got into the passenger side of one of San Francisco’s first driverless taxis recently, he was so nervous that he tried to put on the wrong seat belt.
However, his nerves were misplaced, and the vehicle responded as it should when Vogt hit “drive” on his app and took him on his way. Vogt joked that he was more nervous than the car.
So far, so good for San Francisco’s foray into the world of driverless transport.
It has been a while coming, with some speed bumps along the way. Still, California’s transport regulator recently approved GM’s Cruise and Google’s sister company Waymo to commercialize “robo taxi services” in the city with limits.
Cruise, for example, can operate only from 10 pm to 6 am and in certain areas. At the same time, Waymo vehicles must have a safety driver at the wheel who can override the driverless controls if needed.
These are promising developments, but it seems that after years of hype and spending of an estimated USD80 billion by automakers and technology companies, the new era of driverless cars is off to a low-key beginning.
Part of the reason for that is that the technology is still unproven and the unexpected still occurs.
Waymo, for example, had clocked up millions of driverless kilometers so far but had problems with its service in Arizona recently when its vehicles failed to respond to temporary traffic cones.
In San Francisco, its vehicles continued to drive down the same cul-de-sac after failing to properly read a road sign prohibiting traffic.
Driverless trucking company TuSimple employs a safety driver who had to slam on the breaks because he feared the truck’s array of cameras and sensors had not picked up an approaching Jeep in danger of collision.
TuSimple’s fusion of cameras, lidars, and radars can see 1000 meters ahead and use AI and machine learning, but it seems their solutions might have some more to learn yet before it can replicate driver best practices.
Despite these setbacks or doubts, the driverless vehicle push is continuing, but it appears there may be limits – in the medium term at least – to what is achievable. Anyone who thinks that autonomous vehicles will drive anywhere, at any time, on any surface, or any weather condition will find that reality, at least initially, will fall short of their expectations.
Nevertheless, the investments in the driverless industry are continuing to gain momentum, with the U.S. and China seemingly the important centers for development.
They might be two of the oldest names in motoring, but GM and longtime rival Ford are considered leaders in the US. GM, for example, has hired around 1200 people in autonomous vehicle jobs in the last year.
Ford has announced it would partner with retail giant Walmart on self-driving cars, and testing is already underway for autonomous delivery services in three US cities. Ford is leveraging its $1 billion investment in AI company Argo to develop its offering.
The traditional automakers compete with new technology companies such as Google and Apple, which have announced plans for a self-driving car by 2025.
Apple’s efforts were hit when the key executive in charge of its project defected to Ford, but the project is reportedly back on track. The Apple vehicle reportedly has no steering wheels or pedals, and the interior will be designed for hands-off driving with passengers potentially configured in a U-shaped seating configuration.
Meanwhile, in China, the leading startup is WeRide, and the company recently announced a strategic co-operation agreement with automaker GAC Group and its mobility service platform OnTime.
The collaboration is around the commercialization of a “Robotaxi,” with the most recent model integrating OnTime’s ride-hailing platform. This was on display at the Guangzhou Auto Show in November.
Baidu is another player in the market and launched its Apollo Go robotaxi services in five cities, with plans to expand to 30 cities in three years.
In August, the company launched its first Level 5 self-driving robocar with no steering wheel and an upgraded autonomous platform called Luobo Kuaipao. Baidu’s robotaxi had completed more than 400,000 rides and 16 million kilometers at the end of September 2021.
The potential rewards from early leadership are significant. Global consultancy HIS Markit has forecast that the size of China’s self-driving taxi market will surpass US$200 billion by 2030.
Elsewhere, a glimpse of the future was offered at the Tomorrow Mobility World Congress in Barcelona in November 2021, where more than one hundred companies showed off their wares.
These included the Deliverbot, a small autonomous vehicle developed by U.K. company Delivers.ai designed for deliveries in a maximum radius of one mile. More of a device than a vehicle, it moves along the pavement and uses cameras, radars, and sensors to avoid pedestrians.
The congress also showcased the Auve Tech micro-shuffle, a small vehicle designed for customized routes.
Developed in Estonia, it also boasts a hydrogen fuel cell engine, combining two next-generation technologies and pointing the way forward for the vehicle of the future.
Lachlan Colquhoun is the Australia and New Zealand correspondent for CDOTrends and DigitalWorkforceTrends, and the editor of NextGen Connectivity. His fascination is with how businesses are reinventing themselves through digital technology and collaborating with others to become completely new organizations. You can reach him at [email protected].
Image credit: iStockphoto/metamorworks