After burning $100 billion, Robotaxi Robotaxi dares to talk about profitability? Titanium garage

Image source: AI-generated

Robotaxi, which has been in the doldrums for a long time, seems to be ushering in an upward "inflection point" again.

A few days ago, Baidu Apollo officially announced the profit schedule: it will break even in Wuhan by the end of 2024, and enter a profitable period in 2025.

In early April, Tesla CEO Elon Musk revealed on the social media platform X that he would launch Robotaxi on August 8 this year.

Almost at the same time as Tesla's announcement of Robotaxi, GAC Aion and Didi Autonomous Driving announced that their joint venture, Guangzhou Andi Technology Co., Ltd., was approved for an industrial and commercial license, and Andi Technology is 50% owned by GAC Aion and Didi Autonomous Driving.

The reason why it is said to be "re-" is because in 2018 and before, most of the bullets in the industry were aimed at Robotaxi, and almost all of the leading autonomous driving companies at home and abroad announced that they would do Robotaxi, such as Waymo, Pony.ai, Baidu, etc.

However, since 2019, Robotaxi companies have frequently broken out negative news such as financing difficulties, infighting and bankruptcy, layoffs, and listing difficulties, and even the survival situation of leading companies is not optimistic. Although Robotaxi companies such as WeRide and Pony.ai have been rumored to be listed, there has been no news of success.

Therefore, when Baidu, GAC, Didi and Tesla and other leading companies have taken action, does it mean that Robotaxi is good again?

With a "clear schedule"

Compared with the previous wave of Robotaxi, this time the companies no longer depicted just the sea of stars, but gave a specific plan.

Musk announced on his social platform, X platform (formerly Twitter), that he would release Robotaxi on August 8, and said that the company is developing a self-driving taxi Robotaxi that hopes to achieve mass production in 2024, which will be a "huge driving force" for Tesla's future growth.

Although whether the promise can be fulfilled as scheduled in 2024, this still needs to be observed, after all, Tesla's ticket skipping incident is not unprecedented. However, foreign media also pointed out that Tesla is actively communicating with the California Vehicle Authority (DMV), hoping to obtain the same license as Waymo to operate Robotaxi in limited areas in 2024.

On May 15, Baidu also officially released the sixth-generation unmanned car of Radish Express after two years.

As early as the 2022 Baidu World Conference, Baidu released the sixth generation of mass-produced unmanned vehicles - Apollo RT6. At that time, Baidu's external caliber was - it will be the first to be put into use on the radish run in 2023. But it wasn't until May of this year that it was officially unveiled. It is understood that this car is produced by Jiangling New Energy.

Baidu said that the first batch of sixth-generation unmanned vehicles delivered will be put into use in Wuhan from now on, and the deployment of 1,000 unmanned vehicles will be completed in Wuhan within the year.

At the same time, Didi Technology, a joint venture between Didi and GAC Aion, plans to launch its first commercial L4 model in 2025.

On April 26, Pony.ai and Toyota jointly established a new company, Lifeng Intelligent Technology, with a registered capital of 798 million yuan. In the first phase, it is planned to launch 1,000 units of Bozhi 4X L4 driverless Robotaxi in the Chinese market, and provide driverless Robotaxi travel services in major first-tier cities through Pony.ai's operation platform.

Of course, in this wave of Robotaxi progress, the most eye-catching is Tesla. Regarding whether Tesla's Robotaxi can repeat the "catfish effect" of electric vehicles, there are many expectations in the industry.

However, Liu Yudong, an investor in Chentao Capital, told Titanium Media App, "Now many people in the industry have high expectations for Tesla to do Robotaxi, as if Tesla will be able to do Robotaxi this year." I don't think it's that easy. ”

He explained that if our ruler is a real alternative to manned taxis, it will certainly not be possible for Tesla to achieve it in a short period of time. But if we set another yardstick, that is, to compare Tesla and Waymo, to judge whether the product released by Tesla this year is likely to do better than Waymo, everyone may have a different judgment on this matter.

Dare to talk about profits?

As one of the most profitable tracks in previous years, according to McKinsey and Company statistics in 2021, companies and venture capital have invested more than $100 billion in driver-related technologies since 2010.

In the case of Cruise, a subsidiary of General Motors, the first echelon player in the global Roboaxis, according to GM's financial report data, the loss in 2023 alone will be 2.414 billion US dollars (about 17.37 billion yuan).

Cruise was founded in 2013 and was acquired by General Motors in 2016 for $600 million. Cruise has remained independent since its acquisition and was at the heart of GM's massive rollout of Robotaxi. If calculated from 2017, Cruise has burned more than 8.5 billion US dollars (about 61.17 billion yuan).

It has been burning money and cannot make a profit, which is also the controversial place of Robotaxi in the past period. But this wave of Robotaxi also released a second signal worth paying attention to - dare to talk about profits.

Recently, Baidu Apollo announced at Apollo Day 2024 that by the end of 2024, Radish Express will break even in Wuhan and fully enter the profitable period in 2025.

The reason why he dared to put forward this goal, Chen Zhuo, general manager of Baidu's autonomous driving business department, said that on the one hand, it is the reduction of costs, and on the other hand, the increase in demand.

In the part of cost reduction, it is mainly achieved through three parts: vehicle cost, labor cost and service cost.

According to Baidu, the cost of the fifth-generation unmanned car is about 500,000, a 60% decrease from the previous generation car, and the price is only 200,000. Titanium Media App observed that this is about 50,000 fewer than when the sixth generation of unmanned vehicles was just released in 2022.

One of the most important reasons is the price reduction of sensors, taking lidar as an example, this component that originally cost more than $100,000 has been reduced to $500 to $1,000 per piece. On the other hand, the number of lidars used is also decreasing. "In 2022, we used eight lidars when we released this car, but now it's scaled down." Chen Zhuo revealed.

In addition, Baidu Apollo also adopts a battery swap mode to reduce labor costs. Chen Zhuo said, "Now the fastest fast charging time is at least half an hour to charge a round-trip, and 3 minutes is enough to change the battery." This saves nearly an hour of time, which allows us to serve our users more and improve operational efficiency. ”

While reducing costs, we must also increase the scale and increase revenue. As of April this year, Baidu's autonomous driving test mileage has exceeded 100 million kilometers, and the cumulative service of Radish Express has exceeded 6 million.

In its recently updated prospectus, Ruqi also described Robotaxi's plan at length, predicting that it will commercialize Robotaxi around 2026 and enter a mature commercialization stage by 2030. Ruqi Mobility has launched a hybrid operation model of manned ride-hailing and Robotaxi, and its Robotaxi services have been launched in Nansha, Guangzhou and Shenzhen in 2022 and early 2024 respectively.

Of course, there are not many Robotaxi platforms that talk about profitability timelines at the moment, and as for the companies that have been proposed, can they achieve as promised? Will more platforms follow up to announce profit plans? These will be the next key observations for the Robotaxi industry.

Advance and retreat are intertwined

Frost & Sullivan estimates that in 2019, the cost of a passenger taxi was 1.7 yuan/km, when the cost of Robotaxi was 23.3 yuan/km, and in 2023, the cost of a manned taxi will rise slightly to 1.8 yuan/km, and the cost of Robotaxi will drop to 4.5 yuan/km. It is expected that the cost of Robotaxi will further reduce to 2.1 yuan/km in 2026 and 1.0 yuan/km by 2030, when the cost of Robotaxi may be lower than the cost of passenger taxis.

However, the main reason for restricting the commercialization of Robotaxi is not only cost, but also the lack of mature technology and lagging legislation are obstacles to the commercialization of autonomous driving. Roland Berger, a top management consulting firm in Europe, pointed out that the commercialization of Robotaxi includes five core elements, namely policy regulation, technology, cost, operation services, and market acceptance.

An industry person once analyzed the titanium media App that although the industry is urgent, from a policy point of view, is the urgency of unmanned vehicles high? The answer is not necessarily, especially compared with new energy, unmanned can only be regarded as a long-term plan.

However, even if the pace of legislation is not as fast as the industry expects, it is still moving forward.

On November 17 last year, the four departments officially issued the "Notice on Carrying out the Pilot Work of Intelligent Networked Vehicle Access and Road Access" (hereinafter referred to as the "Notice"), which will select intelligent networked vehicle products equipped with automatic driving functions that have mass production conditions and carry out access pilots.

Previously, in June 2022, Shenzhen took the lead in testing the waters of L3 autonomous driving legislation and issued the "Regulations on the Administration of Intelligent Connected Vehicles in the Shenzhen Special Economic Zone". However, the Regulations do not clearly divide the core responsibility of L3 autonomous driving, while the Notice issued in November 2023 has a relatively clear division: the car company is the main body responsible for the accident when the L3 function is turned on, and the driver is responsible when the function is not turned on, but the car company can also recover from the supplier.

In addition to laws and regulations, some industry insiders mentioned to the Titanium Media App that there are still many long-tail problems in autonomous driving that have not been solved, such as whether there is an insurance company willing to provide insurance services if there is a problem with the vehicle. Domestic Robotaxi are all temporary licenses, and they need to be changed frequently.

Another point is that although the concept of Robotaxi has been proposed in China for five or six years, the number of products is expected to be about a few thousand units, and it is only in a few first- and second-tier cities for demonstration operations. If you want to achieve city-level deployment, it will bring more requirements for different scenarios. This is different from the original small-scale operation, after the city-level large-scale connection, the traffic scenarios of complex roads, including morning and evening rush hours, dense traffic and a series of risks need to be solved technically.

Previously, Cruise's Robotaxi lost its qualification to operate and test in California due to a series of accidents, such as blocking traffic through collective stalls, falling into concrete, hitting fire trucks, and dragging pedestrians resulting in serious injuries, and rekindled a crisis of public confidence in autonomous driving.

Since then, Cruise's valuation has plummeted and nine executives have been fired, with the resignation of the company's chief executive, Kyle Vogt, and the company's co-founder, Dan Kan, while Cruise has also laid off a quarter of its workforce. GM also said on the earnings call that it will cut spending on its self-driving unit, Cruise, by about $1 billion in 2024.

While Cruise recently announced that the next testing phase will begin in Phoenix and gradually expand to the city's suburbs. However, Cruise has not yet disclosed when it plans to resume driverless commercial operations.

In addition, following Apple, Samsung Electronics has stopped research on self-driving cars. Samsung Electronics' Samsung Advanced Institute of Technology (SAIT), which is responsible for medium- and long-term development, has excluded autonomous driving from its research projects and shifted its developers to robotics as part of a selection and concentration strategy, as reported by the media on May 9.

Karl Iagnemma, CEO of Hyundai Motor Group's autonomous driving subsidiary, Motional, also recently announced, "We have delayed the commercialization of our autonomous driving products and laid off some employees." ”

Although the Robotaxi industry is gradually showing more hope, policies, laws and regulations, and resources ...... It's not just one of them that gets stuck in Robotaxi, and it takes time to fix them.

It is true that the actions of the head companies are more representative, but for these companies, how much gap between the symbolic meaning and the actual value of Robotaxi still needs to be observed in the future. In the foreseeable time, "interweaving advance and retreat" will be the most common phenomenon in the entire industry.

(This article was first published in Titanium Media App Author|Han Jingxian Editor|Zhang Min)