Method and device for operating autonomous vehicle

Abstract

The invention discloses a method and device for operating an autonomous vehicle. According to a specific implementation mode, the method for operating the autonomous vehicle comprises the steps of collecting a driving scheme and driving scene information adopted by a driver during performing abnormal intervention on a semi-autonomous vehicle, determining the level of risk according to the drivingscheme; decomposing object information, vehicle running state information and driving environment information of a driving environment where the vehicle is in from the driving scene information; learning the object information and the corresponding level of risk to recognize a risk object associated with the level of the risk; for the risk object, learning the object information, the vehicle running state information, the driving environment information and the driving scheme to determine an association relationship between the combination of the object information of the risk object, the vehicle running state information and the driving environment information and the driving scheme as a driving strategy; and using the driving strategy to determine a candidate driving scheme. According tothe specific implementation mode of the method for operating the autonomous vehicle, the candidate driving scheme is optimized according to behaviors of the driver.

Description

[0001] This application is a divisional application of a Chinese patent application with the application number CN201610825323.3, the application date is September 14, 2016, and the invention title is "Method and device for operating an autonomous vehicle". technical field [0002] The present application relates to the field of vehicle technology, in particular to unmanned vehicle technology, and in particular to a method and device for operating an autonomous vehicle. Background technique [0003] Autonomous driving mainly involves three main technologies: recognition, decision-making, and control. For autonomous driving, safety is the first condition. In the prior art, manual methods are usually used to mark risky objects identified by on-board sensors, and further assess the risk of each object or establish a driving plan through repeated tests. [0004] However, relying entirely on manual labeling of risk factors is extremely costly. Moreover, in an open driving envir...

AI Technical Summary

Problems solved by technology

[0004] However, relying entirely on manual labeling of risk factors is extremely costly

Moreover, in an open driving environment, the existing risk factors are difficult to be exhausted and manually marked by designers and producers, so that the system can automatically identify them

In addition, even if all the current risk objects are pre-marked when the vehicle is designed or manufactured, after the vehicle is put into use, new things that may cause driving risks will continue to appear during the driving process of the vehicle, and the existing methods cannot timely detect these Labeling of new risk objects that continue to appear, and it is impossible to establish risk assessment strategies or driving strategies for these new risk objects in a timely manner

Images