Among China's major Internet giants, Baidu and its founder Robin Li have been betting on intelligent transportation and autonomous driving tracks "without borders". Recently, it has also launched the "7 Lectures on Intelligent Transportation" on the APP.
At the end of February, Yangquan City, Shanxi Province, officially issued a commercialization time permit notice to Robocop, a Baidu self-driving travel service platform. This is another commercial landing city after Beijing and Chongqing.
The internal reason for the radish run to choose to land in Yangquan is not difficult to guess: Yangquan is the hometown of Baidu CEO Robin Li, just as Liu Qiangdong opened a Jingdong customer service center in his hometown of Suqian, Jiangsu Province, Baidu also opened an important cloud computing center here. In addition, the test center of Baidu's driverless car Apollo project is also located here.
This is also highly related to the geographical characteristics of Yangquan. Geographically, Yangquan is located in the eastern foothills of the Loess Plateau, the terrain of the whole city is changeable, and the road conditions are more complex than those of the plain city, which naturally provides a changeable test environment for Baidu's unmanned driving; at the same time, the lower labor costs on the one hand reduce Baidu's research and development expenses, on the other hand, it also provides opportunities for digital transformation for Yangquan, as a traditional resource-based city.
Driverless Apollo is equivalent to Baidu's "moon landing" project
In the early morning of February 26, a reporter from the Science and Technology Innovation Board Daily took a high-speed train from Beijing to Yangquan, Shanxi. After 2 hours and 2 minutes, the speeding high-speed train arrived at Yangquan North Railway Station. After another 40 minutes by car, the reporter came to the Baidu Cloud Computing (Yangquan) Center located at No. 61 Dalian East Road, Yangquan Economic Development Zone. Inside the central park, Baidu insiders introduced the development process of Baidu's unmanned driving project to the "Science and Technology Innovation Board Daily Reporter".
Entering the first floor of the exhibition hall, the reporter of the Science and Technology Innovation Board Daily noticed a driverless car placed inside the exhibition hall. It is understood that this is the Apollo Robocop automatic inspection vehicle jointly developed by Baidu and FAW Hongqi, which is used to check for violations such as infrastructure failures or speeding of motor vehicles on the road.
It is worth noting that Robocop is also the first batch of platform-level products in China to combine L4-level automatic driving with automatic road inspection and maintenance. In addition to detecting daily scenes such as speeding, reverse driving, and occupied road construction, customized expansion can also be carried out, and the efficiency can be increased by two to three times compared with ordinary inspection vehicles.
In order to make products such as Robocop play a greater role, since 2018, Apollo has open sourced vehicle-road collaboration solutions to combine smart vehicles with smart roads. Its final product form is the central platform of the city's car city network. So far, Yangquan has realized the intelligent upgrading of vehicles and roads in both horizontal and vertical kilometers and more than 50 intersections in the core area.
Specifically, at each core intersection, Baidu's Apollo vehicle-road coordination suite will be installed, and after completion, it will be presented in the form of a pillar of light on the central screen of Baidu's car city network. On the left side of the screen are the types and number of smart devices that are already running, each of which can be clicked individually to enlarge.
It is understood that at present, all the autonomous vehicles in the Baidu Cloud Computing Park in Yangquan City have been connected to the above-mentioned central system of the car city network and can be displayed on the central control screen. Taking a traffic violation incident as an example, after the incident, each element will be recorded by the Apollo intelligent transportation system, which greatly reduces the cost of urban traffic management and the difficulty of refined control.
At the same time, at each intersection where the intelligent transformation is completed, the real-time traffic flow, signal light color and countdown status of the traffic light will be collected and fed back to the big screen of the central system of the car city network. In addition to the data center based on big data collection and analysis, Baidu launched a 5G cloud driving control cabin in 2020 for each driverless car launched.
For Baidu, its vision is to achieve a truly L5-level driverless product, which is an important opportunity for the birth of the cloud's driving control warehouse.
Taking the cloud driving control cabin currently displayed as an example, when encountering other unexpected situations such as the car accident in front, the unmanned vehicle will issue a request to the control center of the cloud; when the cloud navigator receives the request, he will connect to the control of the unmanned vehicle to get out of trouble. So far, Baidu's cloud cockpit has been able to achieve one-to-many monitoring and one-on-one to get out of trouble.
The current cloud driving control cabin is far more than used in unmanned vehicles such as Robotax, but can also be applied to driverless operations in various extreme environments. In 2020, Baidu has cooperated with related companies to control special facilities such as mining towers in Inner Mongolia to achieve some monitoring operations in mining towers. Of course, this is just the tip of the iceberg for the application of 5G cloud driving technology.
Beijing, Chongqing and Yangquan have opened the "Radish Run" charging operation
From the perspective of development history, Baidu's autonomous driving technology has undergone several changes. Since 2013, Baidu's unmanned vehicle project has officially started research and development; in 2015, the first unmanned concept car was unveiled at the World Internet Conference.
In 2017, the Apollo plan was officially proposed. Three months later, at the Baidu AI Developer Conference, Robin Li drove to the fifth ring road in Beijing in an unmanned car. Although Li Yanhong ate a ticket because of legal problems, in a realistic sense, this is an important milestone in the official landing of unmanned driving.
In 2019, Baidu's main driverless models began to enter the mass production stage, and it was also the same year that Baidu's Apollo project began research and development and testing for the first time in the Yangquan section of the Wulu Expressway in Shanxi.
This is also the starting point of Baidu Apollo Park and Yangquan.
In the follow-up vehicle test ride of the radish fast run, the reporter of the Science and Technology Innovation Board Daily experienced the driverless travel of the radish fast run. In the five-kilometer test ride section, the safety officer in charge of the emergency situation introduced to reporters that from February 27, 2022, users can log on to the Radish Run APP to use the travel services of driverless cars throughout Yangquan City. Up to now, Radish Run has covered 8 cities in Beijing, Shanghai, Guangzhou, Shenzhen, Chongqing, Changsha, Cangzhou and Yangquan; among them, Beijing, Chongqing and Yangquan have all started official fee operations.
In the process of vehicle driving, the reporter of the Science and Technology Innovation Board Daily found that when there is no obstacle in front, the speed of the driverless car can reach a maximum of 50 km / h; but when there are small vehicles such as takeaway vehicles in front, the operating speed will drop to 20 km / h.
At the same time, in front of each seat, a monitoring screen displaying the real-time traffic situation around the perimeter is installed, and the user can obtain the surrounding traffic situation in real time. In the face of intersections and road conditions with more vehicles, driverless cars will also make corresponding feedback based on real-time changes in road conditions.
At the technical level, this real-time traffic feedback relies on the transmission of 5G signals, which also has certain requirements for network bandwidth. It is understood that when the network condition is not good, the vehicle will prioritize the allocation of network resources to the perception of the surrounding environment, and the network transmission back to the data center will be weakened accordingly.
When autonomous driving "meets" Yangquan
During the investigation and investigation of the "Science and Technology Innovation Board Daily", Li Zhihua, head of the management committee of Yangquan High-tech Zone, introduced the "Yangquan Model" of the digital economy to the reporter of the "Science and Technology Innovation Board Daily".
For a long time, Yangquan as a typical resource-based city, the "coal-based" development model has been the bottleneck restricting Yangquan's economic development in the new historical period. Because of this, how to turn the resource-based economy into the development of emerging industries has always been a new proposition to be solved by the leading group of Yangquan City. At the newly held 12th Party Congress, the Yangquan Municipal Party Committee put the digital development strategy in the first place, which shows the importance it attaches to the digital economy.
The history of Yangquan is not long, but in terms of location conditions, Yangquan is at the junction of Jinxi and Hebei, and the distance from Taiyuan and Shijiazhuang is 100 kilometers. At the Shanxi Provincial Party Congress, Yangquan was also included in the development strategy of the central urban agglomeration, which put Yangquan at the crossroads of development. The construction of a bridgehead facing Beijing-Tianjin-Hebei has become a key task for Yangquan in the new historical period.
Based on this, in accordance with the idea of "accelerated rise of leading industries, rapid cultivation of emerging industries, and advanced layout of future industries", Yangquan High-tech Zone has implemented a cultivation plan of five major industries and tens of billions of industrial clusters, including digital economy, equipment manufacturing, new energy and energy conservation and environmental protection and modern services. The core of this is the digital economy represented by the Baidu Cloud Computing Center.
In terms of specific measures, Yangquan High-tech Zone uses the opening of scenes and the pilot of policies to drive the development of follow-up industries, and takes Baidu apollo park as a model and demonstration to link the development of Yangquan's overall smart city.
Baidu's latest financial report shows that in the fourth quarter ended December 31, 2021, the number of manned orders for radish fast running was about 213,000, nearly doubling from the previous quarter. Baidu does not hide its claim that Baidu's position as the world's largest self-driving travel service provider is being further strengthened.