Tesla fatality could boost autonomous driving

The first reported fatal car crash in a semi-autonomous car could spur the development of more sensitive and accurate self-driving technology.

Read time 6min 10sec
The first reported fatal crash in a Tesla vehicle in semi-autonomous autopilot mode is likely to spur the development of better self-driving technologies.
The first reported fatal crash in a Tesla vehicle in semi-autonomous autopilot mode is likely to spur the development of better self-driving technologies.

Concerns raised by the first reported fatality in a semi-automated car were expected to speed adoption of more sensitive technology to help vehicles see and drive themselves safely, increasing demand on the emerging autonomous vehicle technology industry, investors and analysts said.

Goldman Sachs forecasts the market for advanced driver assistance systems and autonomous vehicles will grow from about $3 billion last year to $96 billion in 2025 and $290 billion in 2035. More than half of that revenue in 20 years, Goldman estimates, will come from radar, cameras and lidar, a sensor that uses laser - all tools considered essential to building vehicles that can pilot themselves.

The 7 May death of Ohio technology company owner Joshua Brown in a Tesla Model S while the car's semi-automated autopilot system was engaged highlighted the limitations of current automated driving systems.

Tesla's Autopilot system uses cameras and radar, but not lidar. The company said its system would have had trouble distinguishing the white semi-trailer blocking Brown's path, against a bright sky.

Industry executives and analysts told Reuters they expect the Tesla crash will spur investment in self-driving vehicle systems that combine multiple of sensors, including lidar.

"As we move to a higher level of autonomy in vehicles, you're going to want to have more redundancy," which radar and lidar can provide, said Dan Galves, senior VP at vision safety system maker Mobileye, in an interview. "The more sensors, the better."

Carmakers have been using multiple sensors in prototypes that are in testing but not yet ready for market. A variety of technologies with overlapping capabilities is seen as a way to increase safety under a wider range of circumstances.

The valuations of some self-driving start-ups "may even increase if there are companies that can solve some of the issues" the Tesla accident raised, said Quin Garcia, MD of AutoTech Ventures, a Silicon Valley investment firm.

Semi-automated systems such as General Motors' SuperCruise and Traffic Jam Pilot from Volkswagen's Audi are due on the market in 2017-2018. Ford expects to deploy a semi-automated system, using Velodyne lidar, in 2018.

Toyota, which is investing more than $1 billion in such self-driving technologies as robotics and artificial intelligence, said it aims to put fully driverless cars on the road in time for the 2020 Tokyo Olympics.

Delphi Automotive plans to build lidar vision systems with technology from Quanergy Systems, which makes solid-state lidar systems. Delphi plans to combine information from the lidar system with radar and other driver assistance technology to create a 360-degree view around a car, a company official said.

Delphi has an investment in Quanergy, one of more than 50 self-driving start-ups that together have raised more than $800 million in investment capital in the past decade, according to a Reuters analysis of publicly available data.

At least two of the start-ups - namely Quanergy, which makes solid-state lidar sensors, and Zoox, which is developing fully automated vehicle systems - have jumped in value to more than $1 billion each since GM's $1.2 billion acquisition earlier this year of another self-driving start-up, Cruise Automation.

Quanergy and Zoox hope to follow the lead of Mobileye, an Israeli supplier of vision-based safety systems to 25 global automakers, including Tesla. Co-founded in 1999 by a computer science professor at Hebrew University, Mobileye went public in 2014 and today is valued at nearly $10 billion.

Mobileye plans by 2020 to offer a hardware/software system that can gather, fuse and analyse data from 20 different sensors, including cameras, lidar and radar. The company's new EyeQ5 "system on chip" will be a key component in a fully autonomous driving system that is being jointly developed with BMW and Intel, and is aimed at production in 2021.

Like Mobileye, Velodyne, a supplier of laser-based lidar systems, works with many of the world's top automakers, including Ford, GM, BMW, Toyota Honda and Daimler's Mercedes-Benz.

"Our clients want to [combine] lidar and cameras," said Velodyne's Marta Hall, president of business development, in an interview with Reuters. Automakers are stepping up orders as lidar systems come down in size and price, she said.

Among the potential beneficiaries of this growing interest is LeddarTech, a relatively young start-up based in Canada's Quebec City. The company is providing LED-based lidar systems to French supplier Valeo, which also buys vision-based systems from Mobileye.

Germany's Robert Bosch, which has been developing self-driving components and systems for more than 15 years, buys lidar from an unnamed tier two supplier and intends to package it in a highly automated "highway pilot" system intended for series production in 2020, said spokesman Tim Wieland.

"Bosch sees the necessity for a sensor setup that includes radar, video and lidar," Wieland said. The three sensors "complement each other very efficiently".

Regulation and litigation

Regulation and litigation are two big wild cards for the autonomous driving sector.

Safety regulators and industry executives have said self-driving cars ultimately could slash traffic fatalities - about 35 000 last year in the United States and more than 1.2 million globally - by up to 90%. But regulators are also concerned that drivers could be lulled into unsafe behaviour by systems that take control for a time, but expect human operators to re-take command in an emergency.

The National Highway Traffic Safety Administration is investigating the role of autopilot in the Florida accident and another crash in Pennsylvania involving a Tesla vehicle. The agency is also expected to roll out later this year broad guidelines for deploying autonomous vehicle technology.

"I hope NHTSA does not overreact" to the crash, said Stefan Heck, co-founder of Nauto, another Silicon Valley self-driving start-up with corporate backing. "The trade-off is quite clear: some safety improvement is better than none."

Product liability for automated vehicles is uncharted territory. The US Transportation Department has said an automated driving system could be considered the "driver" for regulatory purposes.

Industry executives are betting that consumer interest in the technology will rise.

A survey conducted by AlixPartners in June - before the Tesla accident was reported publicly - found that 90% of respondents would be interested in a self-driving car that would let the driver take the wheel from time to time. The same survey noted that nearly 80% of respondents would pay for the technology - including 10% who would spend up to $5 000.

The favourable response rates are much higher than in previous surveys on self-driving technology.

News of the Tesla crash "is not going to put too much of a dent in public perception" of self-driving cars, said AlixPartners' Mark Wakefield.

Login with