CN-122024210-A - AR-HUD vision focus intervention control method for preventing cognitive tunnel effect

CN122024210ACN 122024210 ACN122024210 ACN 122024210ACN-122024210-A

Abstract

The invention relates to the technical field of vehicle control, in particular to an AR-HUD vision focus intervention control method for preventing cognitive tunnel effect. And obtaining perception data of the target vehicle. And determining each risk target object according to the perception data. And projecting each virtual image layer and each risk target object to an equivalent imaging plane. And determining the competition index between each virtual layer and each risk target object and the environmental risk level of each risk target object according to the perception data and the parameters of each virtual layer. And determining the cognitive tunnel risk index of each risk target object according to the competition indexes and the environmental risk grade of each risk target object. And determining the virtual layer intervention level of each risk target object according to the cognitive tunnel risk index. And determining the virtual layer forming a strong competition relationship with the risk target object according to the competition index of the risk target object and each virtual layer. And performing intervention control according to the intervention level of the virtual layer so as to prevent the driver from generating the cognitive tunnel problem.

Inventors

ZENG MENGJIN
HU WEIMING
TANG QIUYANG
JIANG WEI
LI RUIJIE
WU ZE
CHEN HAOLIN
JING YAN
TANG BANGBEI

Assignees

中汽院(江苏)汽车工程研究院有限公司
中国汽车工程研究院股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (10)

1. An AR-HUD vision focus intervention control method for preventing cognitive tunnel effect is characterized by comprising the following steps of: S1, acquiring perception data acquired by a vehicle perception system of a target vehicle, and acquiring virtual layers of AR-HUD projection and parameters of the virtual layers; s2, determining all risk targets in a central driving corridor of a target vehicle according to the perception data, wherein the central driving corridor is an area which can be observed by a driver through a front windshield and can be used for the target vehicle to run; S3, projecting each virtual image layer and each risk target object to an equivalent imaging plane in the eye box of the driver; S4, according to the perception data and parameters of the virtual layers, determining competition indexes between each virtual layer and each risk target object respectively, and determining the environmental risk level of each risk target object; s5, determining a cognitive tunnel risk index of each risk target object according to each competition index and the environmental risk grade of each risk target object; S6, aiming at each risk target, determining a virtual layer intervention level corresponding to the risk target according to a cognitive tunnel risk index of the risk target, and determining a virtual layer forming a strong competition relationship with the risk target according to a competition index of the risk target and each virtual layer respectively; s7, performing intervention control on the virtual layer forming a strong competition relationship with the risk target object according to the intervention grade of the virtual layer corresponding to the risk target object and preset intervention control operation so as to reduce the visual competition between the virtual layer and the risk target object and prevent the driver from generating a cognitive tunnel problem.
2. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein in S4, according to the perceived data and parameters of the virtual layers, determining a competition index between each virtual layer and each risk target object, respectively, specifically includes: determining a projection area of each virtual image layer and each risk target object in the equivalent imaging plane; Calculating the space overlapping degree, the proximity degree, the dynamic conflict degree, the depth conflict degree and the continuous occupation degree between the projection area of each virtual layer and the projection area of each risk target object according to the perception data and the parameters of each virtual layer; And determining the competition index between each virtual layer and each risk target object according to the space overlapping degree, the proximity degree, the dynamic conflict degree, the depth conflict degree and the continuous occupancy degree between the projection region of each virtual layer and the projection region of each risk target object.
3. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 2, wherein the parameters of each virtual layer include refresh frequency, displacement direction vector and depth plane distance of each virtual layer; According to the perceived data and the parameters of each virtual layer, calculating the spatial overlapping degree, the proximity degree, the dynamic conflict degree, the depth conflict degree and the continuous occupancy degree between the projection areas of each virtual layer and the projection area of each risk target object, wherein the method specifically comprises the following steps: Calculating the space overlapping degree between the projection area of each virtual layer and the projection area of each risk target object; Calculating the distance between the center of the projection area of each virtual layer and the center of the projection area of each risk target object, and normalizing the distance to obtain the proximity; Determining dynamic conflict degree according to the refreshing frequency and displacement direction vector of each virtual layer and the optical flow direction vector of each risk target object; determining depth conflict degree according to the depth plane distance of each virtual layer and the depth data of each risk target object; And determining the accumulated time of the contact of the projection area of each virtual layer and the projection area of each risk object in the preset time length, and determining the continuous occupancy according to the accumulated time.
4. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein said perception data comprises a longitudinal relative distance between each risk target and said target vehicle, a longitudinal relative velocity, a lateral offset, a center point lateral distance, a lateral velocity component of each risk target; S4, determining the environmental risk level of each risk target object according to the perception data, wherein the method specifically comprises the following steps: For each risk target object, determining the collision risk of the risk target object and the target vehicle according to the longitudinal relative distance and the longitudinal relative speed of the risk target object relative to the target vehicle; determining a closing speed risk of the risk target according to the longitudinal relative speed of the risk target relative to the target vehicle; determining the proximity degree of the risk object and the target vehicle according to the lateral offset of the risk object relative to the target vehicle; determining the lane conflict risk of the risk target object and the target vehicle according to the transverse distance of the center point of the risk target object relative to the target vehicle and the transverse speed component of the risk target object; determining a risk coefficient preset by the risk target object according to the type of the risk target object; and determining the environmental risk level of the risk target object according to the collision risk, the closing speed risk, the proximity degree, the lane conflict risk and the risk coefficient.
5. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein S5 specifically comprises: aiming at each risk target object, determining the competition index with the largest value from the competition indexes between the risk target object and each virtual layer; And determining the cognitive tunnel risk index of the risk target object according to the environmental risk grade of the risk target object and the competition index with the maximum numerical value.
6. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 5, wherein determining the cognitive tunnel risk index of the risk target according to the environmental risk level of the risk target and the competition index with the largest value comprises: Determining a preset semantic priority weight of a virtual layer corresponding to the competition index with the largest numerical value; and determining the cognitive tunnel risk index of the risk target object according to the environmental risk level of the risk target object, the competition index with the largest numerical value and the semantic priority weight.
7. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein in S6, according to the competition indexes of the risk target and each virtual layer, determining the virtual layer forming a strong competition relationship with the risk target, comprises: Determining a competition index exceeding a preset competition threshold value from competition indexes between the risk target object and each virtual layer aiming at each risk target object; and determining the virtual layer corresponding to the competition index exceeding the preset competition threshold value as the virtual layer forming a strong competition relationship with the risk target object.
8. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein in S7, intervention control is performed on a virtual layer forming a strong competition relationship with the risk target object according to a preset intervention control operation, and the method specifically comprises: And under the condition that the intervention level of the virtual layer corresponding to the risk target object is the first-level intervention level, reducing the brightness and the contrast of the virtual layer forming a strong competitive relationship with the risk target object.
9. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein in S7, intervention control is performed on a virtual layer forming a strong competition relationship with the risk target object according to a preset intervention control operation, and the method specifically comprises: And under the condition that the intervention level of the virtual layer corresponding to the risk target object is a secondary intervention level, transferring the virtual layer forming a strong competition relationship with the risk target object to the edge of the AR-HUD projection area for two-dimensional display.
10. The AR-HUD vision focus intervention control method for preventing cognitive tunneling according to claim 1, wherein in S7, intervention control is performed on a virtual layer forming a strong competition relationship with the risk target object according to a preset intervention control operation, and the method specifically comprises: and hiding the virtual layer forming a strong competition relationship with the risk target object under the condition that the virtual layer intervention level corresponding to the risk target object is three-level intervention level.

Description

AR-HUD vision focus intervention control method for preventing cognitive tunnel effect Technical Field The specification relates to the technical field of vehicle control, in particular to an AR-HUD vision focus intervention control method for preventing cognitive tunnel effect. Background With the wide application of Augmented REALITY HEAD-Up Display (AR-HUD) in intelligent automobiles, virtual graphics such as navigation arrows, lane guidance, front target labeling, speed limit prompt and the like can be superimposed in the view of the road in front of the driver. The technology effectively reduces the frequency of looking over the instrument or the central control screen by the driver at low head, but brings a complex visual environment in which the virtual image layer and the real road scene are simultaneously presented in the same front view field. Existing AR-HUD systems typically continuously render highlighted, dynamic or spatially bound three-dimensional graphics in the center of the driver's field of view or near the planned trajectory, as required by navigation, assisted driving and cockpit traffic. When the brightness, contrast, edge complexity, flicker frequency, or spatial position setting of the virtual graphic is not reasonable, the virtual graphic may overlap, be adjacent to, or visually compete with high-risk targets in the real road in the field of view, resulting in a delay in the perception of the real risk targets by the driver. The problem is not a simple problem of whether to look ahead or not, but a problem of visual priority competition between a real risk target and a virtual image layer in a front view field, and can be expressed as continuous dependence of a driver on the virtual image, insufficient acquisition of real environment change and delayed response to braking and steering opportunities, so as to form a cognitive tunnel effect. In the prior art, mainly by means of a driver monitoring system (Driver Monitoring System, DMS), driver states such as fatigue, distraction, eye closure, head posture deflection or vision area of a driver are detected and identified, and a judgment object thereof is the driver itself. However, even if the driver's sight line is detected to fall in the front area, it is difficult to further distinguish whether the driver is focused on an actual road target or an AR-HUD virtual layer, and it is difficult to solve the problem of cognitive tunneling caused by competition between the AR-HUD virtual layer and a real risk target. Therefore, the specification provides an AR-HUD vision focus intervention control method for preventing cognitive tunnel effect. Disclosure of Invention The specification provides an AR-HUD visual focus intervention control method for preventing cognitive tunnel effect, so as to solve the problems in the prior art. The technical scheme adopted in the specification is as follows: the specification provides an AR-HUD vision focus intervention control method for preventing cognitive tunnel effect, which comprises the following steps: S1, acquiring perception data acquired by a vehicle perception system of a target vehicle, and acquiring virtual layers of AR-HUD projection and parameters of the virtual layers; s2, determining all risk targets in a central driving corridor of a target vehicle according to the perception data, wherein the central driving corridor is an area which can be observed by a driver through a front windshield and can be used for the target vehicle to run; S3, projecting each virtual image layer and each risk target object to an equivalent imaging plane in the eye box of the driver; S4, according to the perception data and parameters of the virtual layers, determining competition indexes between each virtual layer and each risk target object respectively, and determining the environmental risk level of each risk target object; s5, determining a cognitive tunnel risk index of each risk target object according to each competition index and the environmental risk grade of each risk target object; S6, aiming at each risk target, determining a virtual layer intervention level corresponding to the risk target according to a cognitive tunnel risk index of the risk target, and determining a virtual layer forming a strong competition relationship with the risk target according to a competition index of the risk target and each virtual layer respectively; s7, performing intervention control on the virtual layer forming a strong competition relationship with the risk target object according to the intervention grade of the virtual layer corresponding to the risk target object and preset intervention control operation so as to reduce the visual competition between the virtual layer and the risk target object and prevent the driver from generating a cognitive tunnel problem. According to the technical means, the risk that the attention of the driver is captured by single virtual information is quantified by quantifying the