CN-122009242-A - Vehicle control method and device based on other vehicle vision blind area and vehicle

CN122009242ACN 122009242 ACN122009242 ACN 122009242ACN-122009242-A

Abstract

The embodiment of the application provides a vehicle control method and device based on other vehicle vision blind areas and a vehicle. After the vehicle starts automatic driving, firstly determining the eye point coordinates of a driver of the vehicle, and then constructing a dynamic vision blind area model of the vehicle based on the eye point coordinates, the structural geometric design parameters of the vehicle and the predicted driving track of the vehicle in a future period. And optimizing the predicted running track of the vehicle in the future period based on the dynamic view blind area model to generate a target running track, and finally controlling the vehicle based on the target running track. Wherein the dynamic view blind area model is used for representing the view blind area of the other vehicle which changes with time in the future period. The method is used for achieving the technical effect of improving the control accuracy of the vehicle.

Inventors

LUO JI
Ding xuan
SHEN CHEN
SHEN KAI
PENG PAI

Assignees

岚图汽车科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260310

Claims (10)

1. A vehicle control method based on a visual field blind area of a vehicle is characterized by comprising the following steps: After the vehicle starts automatic driving, determining the eye point coordinates of a driver of the vehicle; Constructing a dynamic visual field blind area model of the other vehicle based on the eye point coordinates, the structural geometric design parameters of the other vehicle and the predicted driving track of the other vehicle in a future period, wherein the dynamic visual field blind area model is used for representing the visual field blind area of the other vehicle which changes with time in the future period; Optimizing the predicted running track of the vehicle in the future period based on the dynamic view blind area model to generate a target running track; And controlling the vehicle based on the target running track.
2. The method of claim 1, wherein optimizing the predicted travel track of the host vehicle during the future period based on the dynamic view blind zone model to generate a target travel track comprises: Optimizing the predicted running track based on the dynamic view blind area model according to a preset optimization target to generate the target running track; the preset optimization targets comprise at least one of reducing the residence time in the visual field blind area of the vehicle, reducing the position deviation from a preset reference running track and reducing the control quantity of the vehicle.
3. The method of claim 1, wherein said determining the eye point coordinates of the driver of the other vehicle comprises: acquiring fusion perception data of the other vehicle through a multi-sensor fusion perception system; And determining the eye point coordinates of the driver according to the fusion perception data of the vehicle.
4. A method according to claim 3, wherein said determining the eye point coordinates of the driver from the fused awareness data of the other vehicle comprises: identifying the face of the driver from the fused perception data of the other vehicle; and calculating the eye point coordinates of the driver according to the face of the driver.
5. A method according to claim 3, wherein said determining the eye point coordinates of the driver from the fused awareness data of the other vehicle comprises: determining the identification of the other vehicle according to the fusion perception data of the other vehicle; And determining corresponding eye ellipse center coordinates from a preset vehicle parameter database according to the identification of the other vehicle, and determining the eye ellipse center coordinates as eye point coordinates.
6. A method according to claim 3, wherein said determining the eye point coordinates of the driver from the fused awareness data of the other vehicle comprises: determining external geometric parameters of the other vehicle according to the fusion perception data of the other vehicle; inputting external geometric parameters of the other vehicle into a hard point estimation model, and obtaining at least one target hard point coordinate output by the hard point estimation model; calculating the center coordinates of the ellipse of the eye according to the coordinates of the at least one target hard point; And determining the eye ellipse center coordinates as eye point coordinates.
7. The method according to any one of claims 1-6, wherein the constructing the dynamic view blind zone model of the other vehicle based on the eyepoint coordinates, the structural geometric design parameters of the other vehicle, and the predicted travel track of the other vehicle in the future period of time includes: according to the eyepoint coordinates and the structural geometric design parameters of the other vehicle, a static vision blind area is built under the vehicle body coordinate system of the other vehicle; and mapping the static visual field blind area into a dynamic visual field blind area model which changes with time under a global coordinate system through space-time synchronous coordinate transformation according to the predicted running track of the other vehicle.
8. A vehicle control device based on other car vision blind areas, characterized by comprising: The determining module is used for determining the eye point coordinates of a driver of the host vehicle after the host vehicle starts automatic driving; The system comprises a building module, a dynamic visual field blind area model and a control module, wherein the building module is used for building the dynamic visual field blind area model of the other vehicle based on the eye point coordinates, the structural geometric design parameters of the other vehicle and the predicted driving track of the other vehicle in a future period, wherein the dynamic visual field blind area model is used for representing the visual field blind area of the other vehicle which changes with time in the future period; The optimizing module is used for optimizing the predicted running track of the vehicle in the future period based on the dynamic view blind area model to generate a target running track; And the control module is used for controlling the vehicle based on the target running track.
9. A vehicle is characterized by comprising a vehicle main body, a memory and a processor; The memory stores computer-executable instructions; The processor executing computer-executable instructions stored in the memory, causing the processor to perform the method of any one of claims 1-7.
10. A computer readable storage medium having stored therein computer executable instructions which when executed by a processor are adapted to carry out the method of any one of claims 1-7.

Description

Vehicle control method and device based on other vehicle vision blind area and vehicle Technical Field The application relates to the technical field of automatic driving, in particular to a vehicle control method and device based on a visual field blind area of a vehicle and the vehicle. Background In the field of automatic driving, it is important to ensure safety and comfort when the host vehicle interacts with other vehicles. Among them, how to make the host vehicle recognize and avoid long-time stay in the blind area of his car is a key challenge. At present, the blind area of the vehicle is estimated through a preset model mainly based on general characteristics of the type, the size, the speed and the like of the vehicle, and the longitudinal speed (such as acceleration or deceleration) of the vehicle is mainly controlled and adjusted through the longitudinal speed so as to avoid entering the blind area of the vehicle. However, the prior art has a technical problem of low accuracy in controlling the vehicle. Disclosure of Invention The embodiment of the application provides a vehicle control method and device based on a visual field blind area of a vehicle and the vehicle, which are used for achieving the technical effect of improving the accuracy of vehicle control. In a first aspect, an embodiment of the present application provides a vehicle control method based on a visual field blind area of a vehicle, including: After the vehicle starts automatic driving, determining the eye point coordinates of a driver of the vehicle; Constructing a dynamic visual field blind area model of the other vehicle based on the eye point coordinates, the structural geometric design parameters of the other vehicle and the predicted driving track of the other vehicle in a future period, wherein the dynamic visual field blind area model is used for representing the visual field blind area of the other vehicle which changes with time in the future period; Optimizing the predicted running track of the vehicle in the future period based on the dynamic view blind area model to generate a target running track; And controlling the vehicle based on the target running track. In a possible implementation manner, the optimizing the predicted running track of the host vehicle in the future period based on the dynamic view blind area model to generate a target running track includes: Optimizing the predicted running track based on the dynamic view blind area model according to a preset optimization target to generate the target running track; the preset optimization targets comprise at least one of reducing the residence time in the visual field blind area of the vehicle, reducing the position deviation from a preset reference running track and reducing the control quantity of the vehicle. In one possible embodiment, the determining the eye point coordinates of the driver of the vehicle includes: acquiring fusion perception data of the other vehicle through a multi-sensor fusion perception system; And determining the eye point coordinates of the driver according to the fusion perception data of the vehicle. In one possible implementation manner, the determining the eye point coordinates of the driver according to the fusion perception data of the other vehicle includes: identifying the face of the driver from the fused perception data of the other vehicle; and calculating the eye point coordinates of the driver according to the face of the driver. In one possible implementation manner, the determining the eye point coordinates of the driver according to the fusion perception data of the other vehicle includes: determining the identification of the other vehicle according to the fusion perception data of the other vehicle; And determining corresponding eye ellipse center coordinates from a preset vehicle parameter database according to the identification of the other vehicle, and determining the eye ellipse center coordinates as eye point coordinates. In one possible implementation manner, the determining the eye point coordinates of the driver according to the fusion perception data of the other vehicle includes: determining external geometric parameters of the other vehicle according to the fusion perception data of the other vehicle; inputting external geometric parameters of the other vehicle into a hard point estimation model, and obtaining at least one target hard point coordinate output by the hard point estimation model; calculating the center coordinates of the ellipse of the eye according to the coordinates of the at least one target hard point; And determining the eye ellipse center coordinates as eye point coordinates. In one possible implementation manner, based on the eye point coordinates, the structural geometric design parameters of the other vehicle and the predicted running track of the other vehicle in a future period, constructing a dynamic vision blind area model of the other vehicle includes: according to the