Judgment method and device for lane changing trend of vehicle and computer storage medium

Abstract

An embodiment of the invention provides a judgment method and device for lane changing trend of a vehicle and a computer storage medium. The judgment method for the lane changing trend of the vehicle comprises the following steps: selecting a frame of road image as a current frame of road image; recognizing and selecting an identification point of a targeted vehicle; recognizing a lane line; calculating a perpendicular distance from the identification point to the lane line and storing the perpendicular distance; and judging whether the targeted vehicle has the lane changing trend or not according to a perpendicular distance between at least one frame of road image prior to the current frame of road image and the current frame of road image, if the targeted vehicle has the lane changing trend, outputting a lane changing trend result, and if the targeted vehicle does not have the lane changing trend or the lane changing trend cannot be judged, repeating the above steps. According to the embodiment, transverse lane changing trend of the targeted vehicle can be effectively predicted in time, and therefore, danger of lane changing of the targeted vehicle to a current vehicle is avoided effectively.

Description

technical field [0001] Embodiments of the present invention relate to a method for judging a lane-changing tendency of a vehicle, a judging device, and a computer storage medium. Background technique [0002] In the current field of autonomous driving research, target recognition technology is one of the key technologies of autonomous driving technology, and it is also the main source of information input for planning decision-making units and vehicle control units. Therefore, timely and accurate target recognition results can bring great convenience and safety to the subsequent vehicle motion control. In the current commercial autonomous driving technology, the highway application scenario is the entry point for major companies, but the application scenario where the target vehicle suddenly cuts into the lane on the highway has always been one of the challenges in the performance evaluation process of the current automatic driving system . For the target vehicle that sudd...

AI Technical Summary

Problems solved by technology

In the current commercial autonomous driving technology, the highway application scenario is the entry point for major companies, but the application scenario where the target vehicle suddenly cuts into the lane on the highway has always been one of the challenges in the performance evaluation process of the current automatic driving system

For the target vehicle that suddenly changes lanes and cuts into the lane, it cannot be identified in time, which often leads to a high risk of safety accidents in autonomous vehicles.

[0003] However, because the current commercial automatic driving system will not judge the lateral movement of the target in time or even make a wrong judgment or will recognize the target late, it is impossible to detect the vehicle that is changing lanes in time, and due to the detection accuracy and resolution of the corresponding sensors. The problem leads to inaccurate calculation of the target's lateral movement speed, which makes the current commercial autopilot system unable to judge the target's lateral movement trend in time

For example, when the vehicle behind changes lanes, the lane-changing vehicle will break into or occupy the lane. If the self-driving vehicle in the occupied lane cannot respond in time, it will easily collide with the lane-changing vehicle, causing traffic accidents

All of the above problems will lead to the driving safety problem of the self-driving vehicle braking too late in the aforementioned scenarios.

Images