Apollo turns into a sweeping monk, and autonomous driving has just begun to change the future industrial ecology

On September 28, Baidu Apollo and Beijing Environmental Sanitation Group Environmental Sanitation Equipment Co., Ltd. jointly launched seven driverless sanitation vehicles. These sanitation vehicles can be used in squares, high-speed railways, airports, blocks, exhibition halls and other scenarios.

These sanitation vehicles are primary applications of unmanned driving, suitable for low-speed, closed scenarios, and are still a long way from fully unmanned driving.

However, this sanitation vehicle is small but fully equipped with all the elements of unmanned driving. Technologies such as artificial intelligence, machine vision, image recognition, and precise positioning have all been applied to this sanitation vehicle.

It is very difficult to realize fully autonomous driving. The industry generally believes that it will take a long time to realize true driverless driving. At present, a large number of domestic and foreign car companies, technology giants, and start-ups are planning to enter this field. Waymo, a subsidiary of Google, plans to launch a driverless taxi service without a safety officer this year. Although driverless driving is far from mature, it has entered the fast lane of development.

Autonomous driving has been applied in certain scenarios and fields. Many cars have already been equipped with L3 level and below autonomous driving functions. Although it cannot completely free people from driving work, it has greatly simplified the driving work and can undertake driving work in scenarios such as highways and traffic jams.

BAT’s competition for autonomous driving

Currently, many companies in the field of autonomous driving have set their goal to achieve Level 4 autonomous driving (all driving operations are completed by the vehicle, and human drivers do not need to pay attention under limited roads and environments). However, autonomous driving technology cannot be achieved overnight and needs to be developed step by step.

For Baidu, the unmanned sanitation vehicle is a verification of existing autonomous driving technology. Applying existing technology to less demanding scenarios as soon as possible will also be conducive to the further development of autonomous driving technology.

The core of autonomous driving technology lies in artificial intelligence, which relies on the machine to make judgments and decisions. The improvement of artificial intelligence requires continuous training, learning and data accumulation. Road test data is a very critical indicator for measuring the level of a company's autonomous driving technology.

For Baidu, the greatest significance of launching driverless sanitation vehicles lies not in the implementation of technology, but in providing training scenarios and data accumulation for the further development of technology.

Baidu entered the field of autonomous driving early and is in a leading position in China. However, Alibaba and Tencent, which entered the field later, are also catching up. Alibaba's major advantage over Baidu in developing autonomous driving is its large-scale logistics business. The logistics business includes both urban transportation and highway transportation, which can provide Alibaba with rich data support.

Baidu itself lacks similar businesses. By accelerating the implementation of technology and expanding application scenarios, Baidu Apollo can obtain more test data, thereby iterating and upgrading the technology and maintaining its competitive advantage.

The competition between BAT is not only about technology, but also about ecology. Baidu started early in autonomous driving, advocated open cooperation and win-win, and established a huge Apollo ecosystem. However, this ecosystem is more like a loose alliance. Baidu is more open to the autonomous driving industry chain than to control it.

In contrast, Alibaba has always been very aggressive in its layout of autonomous driving, emphasizing its control over the industrial chain. For example, Alibaba has deployed high-precision maps through its wholly-owned acquisition of AutoNavi; it has established Qianxun Location in cooperation with China North Industries Group Corporation to master high-precision positioning technology; it insists on independently developing in-vehicle OS, and has established Banma in cooperation with SAIC to be responsible for the implementation of Internet cars and smart cockpits.

Baidu's openness is actually limited by its own scale. Baidu's market value is already several steps behind Alibaba and Tencent. The lack of capital has prompted Baidu to build a more open ecosystem, focus on building its own technological advantages, and use this to leverage the entire autonomous driving industry chain.

It is still difficult to say which of the two models is more advantageous. Alibaba's full industry chain layout will undoubtedly be more efficient in forming landing products. But whether car companies will eventually buy into it is another matter.

Cars are the foundation for autonomous driving, and neither Baidu nor Alibaba has built cars on its own, but has chosen to cooperate with car companies. Powerful car companies will not be content to be just OEMs, and they also hope to have more say in the autonomous driving industry. Baidu's low profile and openness may make it easier to reach a consensus with car companies.

Compared with Baidu and Alibaba, Tencent's layout in autonomous driving is shallower and slower. Tencent participates in the autonomous driving and new energy vehicle industries more in the form of investment. Tencent does not have a controlling stake in most of the invested companies.

Tencent's positioning in autonomous driving is more 2C than 2B, with more emphasis on the implementation of services and products and the projection of the mobile phone ecosystem on cars. The huge advantage in the social field can easily allow Tencent to get a share in the smart car and other fields, but the lack of technical genes and 2B products is likely to make Tencent a second-rate player in the tide of autonomous driving reform.

China has its own ideas on autonomous driving

Google was the first to develop autonomous driving, and the leading companies are American companies represented by Google and GM. In the initial stage of autonomous driving development, China was still a follower of the United States.

In the last round of the Internet economy, Chinese companies followed the pace of American companies and achieved good development. Although domestic Internet companies are not aggressive enough in foreign markets, they at least hold on to the domestic market. In terms of specific business models, most Chinese companies are still conducting localized innovation based on the American model.

At present, the development of autonomous driving is still dominated by single-vehicle intelligence. The American car culture is very developed, and most cities are built around cars. The population density of the United States is also much smaller than that of China. This leads to a much friendlier traffic environment in the United States than in China. Correspondingly, the difficulty of realizing autonomous driving in China is higher than that in the United States.

To achieve autonomous driving, artificial intelligence alone is far from enough. The upgrade of various sensors, high-precision maps, chips and car operating systems will bring new costs to cars. At present, the transformation of the automotive industry is not only about autonomous driving, but also smart cockpits and new energy vehicles will increase the cost of cars. Only by seeking a more inclusive technical route can autonomous driving really fly into the homes of ordinary people.

On September 6, Alibaba announced its autonomous driving technology route, focusing on achieving autonomous driving through the cooperation of autonomous vehicles and road infrastructure. On the same day, Alibaba DAMO Academy and the Ministry of Transport's Highway Research Institute announced the establishment of a joint vehicle-road collaboration laboratory. On September 14, Baidu announced that it would officially open source its Apollo vehicle-road collaboration solution by the end of this year.

The two Internet giants and the heavyweight players in autonomous driving have announced that autonomous driving will rely on vehicle-road collaboration. China's autonomous driving is likely to take a different path from now on.

The biggest difference between vehicle-road collaboration and single-vehicle intelligence is the addition of intelligent road infrastructure. This intelligent road infrastructure consists of a variety of sensors and computing units, which can provide autonomous vehicles with surrounding road conditions information.

Vehicle-road collaboration can reduce the vehicle's own sensor requirements, thereby reducing the cost of implementing autonomous driving. Vehicle-road collaboration technology can directly provide the car with more comprehensive road condition information, while also reducing the requirements for artificial intelligence for autonomous driving.

The vehicle-road collaboration system actually transfers some of the car sensors to public infrastructure, which can not only provide assistance to self-driving cars, but also transmit information to cars driven by humans. The utilization efficiency of sensors is greatly improved, and the cost of self-driving is easier to spread.

As a result, the focus of autonomous driving will shift from improving the performance of automotive sensors and the level of artificial intelligence to road infrastructure construction. The government will play a more important role in this, and China's advantages in infrastructure will promote the promotion of autonomous driving technology.

Vehicle-road collaboration means stable and instant network connection between vehicles and between vehicles and road infrastructure. This puts forward requirements for the realization of the Internet of Vehicles. The rapid development and improvement of 5G technology has laid the foundation for the realization of the Internet of Vehicles. China also has advantages in 5G communications, which will accelerate the realization of the Internet of Vehicles and vehicle-road collaboration.

On September 14, AliOS announced a strategic cooperation with Intel and Datang Telecom on intelligent transportation-vehicle-road collaboration to lay out the construction of intelligent road transportation network. On September 27, China Unicom, Huawei and Baidu jointly demonstrated autonomous driving and remote driving services based on 5G networks. It can be seen that Alibaba and Baidu have significantly accelerated their layout in vehicle-road collaboration and Internet of Vehicles in recent years.

Although fully autonomous driving is still a long way off, it will gradually be implemented in scenarios such as airports, industrial parks, mines, and ports.

BAT is not the only company in the field of autonomous driving. There are also many startups that specialize in a certain application scenario of autonomous driving. With the scale of China's market, the innovation and accumulation of these companies in the sub-sectors may breed new unicorns.

The most eye-catching application at present is the application of autonomous driving in long-distance freight scenarios on highways.

Long-distance freight is labor-intensive, requires high qualifications for drivers, and has a high rate of traffic accidents caused by human factors. In China's long-distance freight industry, driver labor costs can reach 30-40% of the total cost. Long-distance freight requires autonomous driving to replace the pain points of drivers.

In fact, the significance of autonomous driving is not only to save labor costs, but more importantly, to speed up trucks through platooning. Speeding up means improving freight efficiency and road utilization, which will generate huge economic benefits. Moreover, the road conditions in the long-distance freight industry are mostly relatively simple and closed highways, so the difficulty of implementing autonomous driving is relatively low.

The reason why autonomous driving has a bright future is not because of its cool concept, but because it can actually improve efficiency and reduce costs. As autonomous driving gradually takes root in some fields, it will gradually change people's perception of cars and transportation, and change the entire automotive industry ecosystem.

As a winner of Toutiao's Qingyun Plan and Baijiahao's Bai+ Plan, the 2019 Baidu Digital Author of the Year, the Baijiahao's Most Popular Author in the Technology Field, the 2019 Sogou Technology and Culture Author, and the 2021 Baijiahao Quarterly Influential Creator, he has won many awards, including the 2013 Sohu Best Industry Media Person, the 2015 China New Media Entrepreneurship Competition Beijing Third Place, the 2015 Guangmang Experience Award, the 2015 China New Media Entrepreneurship Competition Finals Third Place, and the 2018 Baidu Dynamic Annual Powerful Celebrity.