According to the micro-network news, Apollo Day Technology Open Day was held in Apollo Park in Yizhuang, Beijing. Wang Yunpeng, vice president of Baidu and general manager of Autopilot Technology Department, first shared the milestone of Baidu Apollo Autopilot in unmanned driving and commercialization.
Wang Yunpeng said: "In 20021year, Baidu Apollo fully implemented the" unmanned+commercialized "model room in Shougang Park. In 2022, we will make every effort to build Yizhuang's "highland of intelligent transportation demonstration area", relying on the technical accumulation of driverless, 5G cloud driving and vehicle-road coordination to realize the technical route suitable for the complex road conditions in China and quickly seize the commanding heights of autonomous driving technology. "
At present, with the development of autonomous driving technology, autonomous vehicles need to rely on the network to communicate and realize various vehicle control processes such as route allocation, emergency report, emergency control and cloud driving. However, in the actual environment, the network of self-driving vehicles often appears abnormal for various reasons, which leads to certain security risks of self-driving vehicles.
In order to improve the safety of self-driving vehicles, on July 28th, 20021,Baidu applied for an invention patent (application number: 202 110856969.9) entitled "Methods, devices, equipment, media and products for controlling self-driving vehicles". The applicant is Beijing Baidu Netcom Technology Co., Ltd.
According to the relevant information currently published in this patent, let's take a look at this technical scheme. As shown in the above figure, it is a schematic flow chart of the control method of the self-driving vehicle of the present invention. Firstly, the automatic driving system needs to obtain the network information of the vehicle, which is obtained by monitoring the network status of the self-driving vehicle through the terminal equipment carried in the vehicle, including network delay, network bandwidth, network speed and so on.
Secondly, in response to determining that the network information meets the preset network fault conditions, the automatic driving mode of the self-driving vehicle is determined. The preset network fault conditions can judge whether the network has a fault, such as: network delay is greater than the preset threshold, network bandwidth is less than the preset threshold, network speed is less than the preset threshold, and other conditions that can judge network health. When these network failures occur, it is judged that it is necessary to start the automatic driving mode, including: driver manual driving, vehicle automatic driving, cloud driving and other modes.
In this process, the automatic driving decision to adjust the driving state can also be determined according to the dependence of the target driving mode on the network and the degree of network failure. Moreover, the higher the target driving mode depends on the network, the higher the network failure, and the greater the restrictions on the determined autonomous driving decision. For example, when there is a serious network failure, you can only switch to the driver's manual driving mode, but you can't use cloud driving, automatic driving and other methods.
Finally, based on the above judgment and analysis, the self-driving vehicle can be switched by the network operator, and the self-driving vehicle can be controlled according to the automatic driving decision during the switching operation of the network operator, so as to improve the switching effect of the target network operation service provider.
Article picture 2
As shown in the above figure, it is a schematic diagram of the application scenario when the scheme is used to control autonomous vehicles. The executive body of the scheme is the terminal equipment installed on the self-driving car, which can communicate with the server through the network. After that, the terminal equipment will respond to the network failure and the decision-making mode of autonomous driving, thus switching the network operator of the vehicle, and will slow down when switching the network operator of the autonomous driving vehicle, thus improving the safety of autonomous driving.