Shulou(Shulou.com)11/24 Report--

Between autopilot ideal and reality bread, Ford chose the latter.

Earlier, Silicon had reported the ebb of the self-driving industry in the United States, and the day before yesterday, the board of directors of another star company, Tucson, suddenly announced the bizarre decision to remove CEO.

This raises the question: will the self-driving industry be all right?

In fact, it is companies that target fully autopilot technology (L4 and above) that may face real survival challenges. Big Brother Mobileye, who has a number of mature businesses, was able to go public successfully despite a sharp drop in valuation.

Another giant in the fully autonomous driving industry announced the closure almost at the same time:

Argo AI, a leader in self-driving behind Ford and Volkswagen, which has so far raised $3.6 billion in financing, was the first to fall during the industry's downturn. It's not just layoffs or downsizing, it's just announcing the closure.

Ford, which had a valuation of more than $7 billion, was within walking distance of going public, and last week Ford reported third-quarter results, although quarterly revenue of $39.392 billion far exceeded Wall Street's expectations. but the net profit attributable to shareholders was a net loss of $827 million. The net loss was mainly due to a failed investment in Argo AI, which was as high as $2.7 billion. At the earnings report, Ford officially announced that Argo AI would stop operating and more than 2000 employees around the world would be disbanded.

Founded in 2016 and headquartered in Pittsburgh, Argo AI is co-founded by Bryan Salesky, head of Google's self-driving car hardware, and Peter Ranger, head of Uber's self-driving project. Its main business is to develop, test and commercialize self-driving technology.

Around 2016 is the early stage of the outbreak of self-driving companies in the United States. in March of that year, General Motors bought self-driving technology startup Cruise for about $1 billion, and then Wyamo announced a spin-off from Google at the end of the year. The valuation has broken through the 100 billion mark all the way, and the self-driving battle among traditional car companies and the technology world as a whole has risen sharply.

So when Argo AI, with the core technologies of Google and Uber self-driving, was born, Ford couldn't wait to do it. In early 2017, Ford announced a $1 billion investment in Argo AI, which was just three months old, when Argo AI had fewer than 12 employees.

Ford announced its investment in Argo AI in February 2017, with Peter Range on the left and rBryan Salesky on the left. the picture shows that Ford was ambitious in the self-driving field at that time, hoping to quickly cut into the self-driving track through Argo AI and take the lead in the industry competition. After investing in Argo AI, Ford said it plans to start selling fully self-driving commercial cars in 2021 and launch robotic taxis and self-driving logistics vehicles in 2022.

In addition, Ford seems to have greater ambitions for the future commercialization of Argo AI. Instead of buying Argo AI directly, Ford chose to become its customer and investor. Although the main task of Argo AI at the time was to develop a "virtual driving system" to be the first to integrate it into Ford's new self-driving cars, Ford also hoped to serve customers other than Ford's own auto ecology in the future.

Ford attached great importance to Argo AI. In addition to the real investment of 1 billion US dollars, Ford also built a test fleet and provided a car factory for Argo AI. In 2019, Volkswagen also made a high-profile bet on Argo AI, injecting $2.6 billion into Argo AI and co-controlling Argo AI with Ford, expanding its cooperation into electric vehicles and self-driving.

Of the $2.6 billion invested by Volkswagen, $1 billion is in cash and $1.6 billion is the merger of Volkswagen's self-driving technology company Autonomous IntelligentDriving (AID) into Argo AI.

This investment was considered at that time an initiative in the traditional car-building industry to jointly dilute the development cost of self-driving technology and win market opportunities. The strong support of Ford and Volkswagen made Argo AI a star in the self-driving startup field, directly pushing Argo AI's market capitalization to $7 billion, and Argo AI quickly became one of the top self-driving companies, with more than 2000 employees expanding rapidly around the world.

In the picture from Twitter in April last year, Argo AI co-founder Bryan Salesky told employees at an internal meeting that as the commercialization of self-driving technology approaches, the company is seeking additional financing and is considering going public as early as 2021. Since then, Lyft has taken a 2.5 per cent stake in Argo AI as an undisclosed round of investment in Argo AI.

At the time, it seemed that the development of Argo AI was going well. It not only tested fully self-driving cars in cities such as Washington, Miami and Austin, but also planned cooperation with Lyft automatic taxi business in various cities, and it also partnered with Wal-Mart to deploy self-driving delivery services.

But since the beginning of this year, as the market environment has returned to calm, Argo AI, which is not profitable and has spent money crazily, has become in jeopardy. In July, Argo AI said it would cut 150 jobs as a result of a change in its business plan, accounting for more than 15 per cent of the company's workforce.

This time, Argo AI's announcement of a direct shutdown not only means that billions of dollars spent by Volkswagen and Ford over the past few years have been wasted, but that the company's more than 2000 employees will be laid off on the spot, except for a small number of employees hired by Ford and Volkswagen.

The road of L4 is too long for Ford and Volkswagen to burn. The main reason behind the closure of Argo AI is the continuous large-scale burning of money and the difficulty of commercial landing. Ford said in its third quarterly report that Argo AI has been unable to attract new investors recently. In other words, if the Argo AI continues to run, Ford and Volkswagen will have to pay for it themselves.

However, in the face of the current market environment and the technological progress of L4 self-driving, the two auto giants finally chose to stop loss and leave the market.

At the earnings report, Ford CEO Jim Farley also basically expressed Ford's current views and commercial attitudes towards the commercialization of self-driving: first of all, Ford believes that the commercialization of L4 self-driving still has a long way to go. Jim Farley said that while Ford is optimistic about the future of the L4 ADAS, the large-scale commercial landing of driverless cars will cost billions of dollars and will take at least five years. In the future, even if Ford needs L4 technology, it can be acquired through other ways such as purchase, and it is not necessary to develop this technology on its own.

Even Ford's internal executives in charge of product development and technology bluntly say that developing self-driving cars and commercializing them on a large scale is now "harder than sending a man to the moon." As a result, Ford also made a strategic decision to shift resources to the development of L2 + and L3 technical driver assistance systems, temporarily abandoning self-driving car technology that can be directly applied to self-driving taxis.

Farley said that Ford's most important task now is to develop excellent, differentiated and safer L2 + and L3 auxiliary driving systems, which can be used to expand the current possibility of increasing revenue and profits. Electric transformation will be a significant priority over autopilot.

In addition to Ford, another "financier" Volkswagen also announced the abandonment of Argo AI. Volkswagen said on Wednesday that it would no longer invest in Argo AI and instead explore a new self-driving partnership.

But unlike Ford, Volkswagen has not completely given up on its own investment in self-driving systems. At present, Volkswagen's main partners in the field of autopilot include Bosch and Qualcomm, the former of which is auxiliary autopilot and L3 autopilot, while the latter is L4 autopilot. In addition, in the Chinese market, Volkswagen announced this month that it would set up a joint venture with Horizon Robot to develop autopilot assistance systems and autopilot solutions.

At present, the ID.Buzz of Volkswagen's special electric van is in the experimental stage, and the first results of the cooperation between Volkswagen and Bosch software are scheduled to be launched in 2023.

| L4 Mountain, which is difficult to cross, who will be the next domino?

The suspension of Argo AI is just another microcosm of the current dilemma in the self-driving industry, especially those companies that once had ambitions to achieve autopilot at level L4 and above are now facing the same commercial dilemma.

The American Society of Automotive Engineers (SAE) divides the autopilot technology into five levels of L0-L5 according to the degree of control of the vehicle control task. The driving system of the general system L1-3 mainly plays an auxiliary function, while the system function of L4 has already handed over all the driving of the vehicle to the system, that is to say, there is no need to control it artificially. L5 is an idealized state of pure intelligent operation.

In theory, autopilot can be commercialized on a large scale when L4 is realized. one of the existing problems is that there is not a subdivision standard below L4. Factors such as lighting conditions and road complexity will lead to great differences in test results. Therefore, although many car factories and technology providers have indicated that they have achieved L4 autopilot, the L4 level is actually uneven, and many of them have not reached the level of commercialization.

Picture from Intel at present, domestic and foreign self-driving manufacturers mainly have two kinds of market-oriented routes. The first is the "gradual evolution" route, which gradually adds the auxiliary function of self-driving to traditional cars, and collects a large amount of data from the real world through the cars that have been sold, eventually helping to transition to the stage of fully self-driving. Such as Tesla, Weilai and so on.

The other is to take the "one-step" route, that is, after conquering the self-driving technology that can be applied on a large scale, we will produce "brand-new" self-driving cars that really do not need drivers and pedals, such as Waymo, Cruise, Argo AI and so on.

At present, these companies that pursue one step are facing a lot of challenges. In addition to the Argo AI shutdown, the valuation of Waymo, the industry leader, has fallen from a peak of $175 billion to $30 billion. Although Waymo has started testing in more than 20 cities and recently launched unmanned taxis for commercial use, it is a drop in the bucket compared to Google's huge investment of more than a decade.

In addition, GM's bet on Cruise is also making slow progress and is still in the testing stage. In March this year, Softbank Corp., who suffered from blood loss, also had to withdraw from his investment in Cruise to stop his loss.

The reality is that there is still a long way to go to achieve L4. In addition to computing platforms, chips, lidar and other hardware and software to reach a considerable technical level, self-driving supervision and landing also requires a long game run-in. If the project does not generate any revenue during this period, the huge cost may only be affordable by a rich and wayward financier like Google.

The fall of Argo AI is undoubtedly a wake-up call to Waymo, Cruise and other car companies with L4 or even L5 dreams. After all, at present, enterprises must be kept alive before they can hope to see the day of large-scale self-driving.

This article comes from the official account of Wechat: Silicon Man (ID:guixingren123), article: Juny, Editor: VickyXiao

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.