US 11,718,306 B2
Method and apparatus for acquiring sample deviation data, and electronic device
Houqiang Zhu, Beijing (CN); Pengfei Wei, Beijing (CN); Jia Song, Beijing (CN); Jianxin Jiang, Beijing (CN); and Songtao Chang, Beijing (CN)
Assigned to BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD., Beijing (CN)
Filed by BEIJING BAIDU NETCOM SCIENCE TECHNOLOGY CO., LTD., Beijing (CN)
Filed on Jun. 4, 2021, as Appl. No. 17/339,113.
Claims priority of application No. 202011547791.1 (CN), filed on Dec. 23, 2020.
Prior Publication US 2021/0291843 A1, Sep. 23, 2021
Int. Cl. B60W 40/09 (2012.01); B60W 50/08 (2020.01)
CPC B60W 40/09 (2013.01) [B60W 50/085 (2013.01); B60W 2540/10 (2013.01); B60W 2540/12 (2013.01); B60W 2540/30 (2013.01)] 8 Claims
OG exemplary drawing
 
1. A method for acquiring sample deviation data, comprising:
acquiring, in a manual driving mode, during a continuous time period, a first driving behavior parameter of a vehicle and a second driving behavior parameter of the vehicle in a simulated automatic driving mode;
comparing, the first driving behavior parameter with the second driving behavior parameter during the continuous time period; wherein the continuous time period refers to a time period for comparison;
determining whether there is a deviation between the first driving behavior parameter and the second driving behavior parameter during the continuous time period;
controlling a processor of the vehicle to acquire sample deviation data within a preset time period when there is the deviation; wherein the sample deviation data comprises information of a vehicle body state and information of an environment where the vehicle is located; and
sending the sample deviation data to a cloud server; wherein the sample deviation data is used for instructing the cloud server to optimize an automatic driving model based on the sample deviation data;
wherein the controlling a processor of the vehicle to acquire the sample deviation data within a preset time period comprises:
controlling a processor of the vehicle to collect the sample deviation data in a first time period that takes a time when it is determined that there is the deviation as a starting time, and
extracting, from collected data, the sample deviation data in a second time period that is before the time when it is determined that there is the deviation;
wherein the first time period and the second time period are two adjacent time periods, and there exists the deviation within both the first time period and the second time period;
wherein the determining whether there is a deviation between the first driving behavior parameter and the second driving behavior parameter during the continuous time period comprises:
determining, when the first driving behavior parameter and the second driving behavior parameter meet a deviation behavior rule, that there is the deviation between the first driving behavior parameter and the second driving behavior parameter; wherein the deviation behavior rule is determined according to a driving behavior and the driving behavior parameter corresponding to the driving behavior of the vehicle in both the manual driving mode and the simulated automatic driving mode;
wherein the method further comprises:
determining, the driving behavior parameter corresponding to the driving behavior of the vehicle in the manual driving mode and the simulated automatic driving mode, respectively;
determining, according to the driving behavior parameter corresponding to the driving behavior, a condition under which the driving behavior have a deviation during the continuous time period in the manual driving mode and the simulated automatic driving mode; and
establishing, according to the condition under which the driving behavior has a deviation in the manual driving mode and the simulated automatic driving mode, the deviation behavior rule.