An ar-hud visual focus intervention control method against cognitive tunneling

Abstract

The present application relates to the technical field of vehicle control, and relates to an AR-HUD visual focus intervention control method for preventing cognitive tunnel effect. Perception data of a target vehicle is acquired. According to the perception data, each risk target object is determined. Each virtual layer and each risk target object is projected to an equivalent imaging plane. According to the perception data and parameters of each virtual layer, a competition index between each virtual layer and each risk target object is determined, and an environmental risk level of each risk target object is determined. According to the competition index and the environmental risk level of each risk target object, a cognitive tunnel risk index of each risk target object is determined. According to the cognitive tunnel risk index, a virtual layer intervention level of each risk target object is determined. According to the competition index between the risk target object and each virtual layer, a virtual layer in a strong competitive relationship with the risk target object is determined. According to the virtual layer intervention level, intervention control is performed to prevent the driver from having a cognitive tunnel problem.

Description

Technical Field

[0001] This specification relates to the field of vehicle control technology, and in particular to an AR-HUD visual focus intervention control method to prevent cognitive tunneling effect. Background Technology

[0002] With the widespread application of Augmented Reality Head-Up Displays (AR-HUDs) in smart cars, virtual graphics such as navigation arrows, lane guidance, forward target markings, and speed limit warnings can be overlaid on the driver's field of vision. This technology effectively reduces the frequency with which drivers look down at the instrument panel or central control screen, but it also brings a complex visual environment where virtual layers and real road scenes are presented simultaneously in the same field of vision.

[0003] Existing AR-HUD systems typically render highlighted, dynamic, or spatially rigged 3D graphics continuously in the center of the driver's field of view or near the planned trajectory, based on the needs of navigation, assisted driving, and cockpit operations. When the brightness, contrast, edge complexity, flicker frequency, or spatial position of the virtual graphics are set inappropriately, the virtual graphics may overlap, be adjacent to, or visually compete with high-risk targets in the real road within the field of view, leading to a delay in the driver's perception of real-world risks. This problem is not simply a matter of "looking ahead," but rather a competition for visual priority between real-world risks and virtual graphics within the forward field of view. This can manifest as the driver's continued reliance on virtual graphics, insufficient awareness of changes in the real environment, and delayed responses to braking and steering, ultimately creating a cognitive tunneling effect.

[0004] In existing technologies, the main approach relies on a Driver Monitoring System (DMS) to detect and identify driver states such as fatigue, distraction, eye closure, head posture deviation, or field of vision. The focus of this assessment is the driver themselves. However, even if this method can detect that the driver's gaze is on the area ahead, it struggles to distinguish whether the driver is focused on actual road targets or the AR-HUD virtual layer. This makes it difficult to address the cognitive tunneling problem caused by competition between the AR-HUD virtual layer and real-world risk targets.

[0005] Therefore, this specification provides an AR-HUD visual focus intervention and control method to prevent cognitive tunneling effect. Summary of the Invention

[0006] This specification provides an AR-HUD visual focus intervention control method to prevent cognitive tunneling effect, thereby solving the aforementioned problems existing in the prior art.

[0007] The following technical solution is adopted in this specification:

[0008] This manual provides a method for controlling the visual focus of an AR-HUD to prevent cognitive tunneling, including:

[0009] S1. Obtain the perception data collected by the vehicle perception system of the target vehicle, and obtain each virtual layer of the AR-HUD projection and the parameters of each virtual layer;

[0010] S2. Based on the perceived data, determine each risk target in the central driving corridor of the target vehicle, wherein the central driving corridor is the area in the target vehicle that the driver can observe through the windshield and that the target vehicle can drive in;

[0011] S3. Project each virtual layer and each risk target onto the equivalent imaging plane in the driver's eye box;

[0012] S4. Based on the perceived data and the parameters of each virtual layer, determine the competition index between each virtual layer and each risk target, and determine the environmental risk level of each risk target;

[0013] S5. Determine the cognitive tunnel risk index for each risk target based on each competition index and the environmental risk level of each risk target;

[0014] S6. For each risk target, determine the virtual layer intervention level corresponding to the risk target based on the cognitive tunnel risk index of the risk target; and determine the virtual layers that form a strong competitive relationship with the risk target based on the competition index between the risk target and each virtual layer respectively.

[0015] S7. Based on the intervention level of the virtual layer corresponding to the risk target, and according to the preset intervention control operation, intervene and control the virtual layer that forms a strong competitive relationship with the risk target to reduce the visual competition between the virtual layer and the risk target and prevent the driver from experiencing cognitive tunneling problems.

[0016] Based on the aforementioned technical means, this solution quantifies the competition index between virtual layers and risky targets, and combines this with the environmental risk level to quantify the risk of driver attention being "captured" by a single piece of virtual information. When the cognitive tunnel risk index reaches a threshold, proactive intervention is made on virtual layers with strong competitive relationships to reduce the risk of driver attention being captured by non-critical virtual information, thereby guiding the driver's attention back to real road risks and preventing attention fixation from the source. This solution focuses on real risky targets within the "central driving corridor" and dynamically adjusts the display priority of virtual layers according to their real-time risk level. This ensures that key warning information is effectively highlighted in high-risk scenarios, while non-critical virtual information is suppressed, reducing reaction delays caused by information overload or visual occlusion. By uniformly mapping virtual layers and risky targets to the equivalent imaging plane of the driver's eye box, precise competitive analysis is achieved under the same visual coordinate system. This avoids misjudgments caused by misalignment between virtual and real spaces, allowing intervention control to precisely target virtual layers that have a "strong competitive relationship" with risky targets, rather than blindly adjusting all displayed content, maintaining the availability of necessary information.

[0017] Furthermore, in S4, based on the perceived data and the parameters of each virtual layer, the competition index between each virtual layer and each risk target is determined, specifically including:

[0018] In the equivalent imaging plane, the projection area of ​​each virtual layer and each risk target is determined;

[0019] Based on the perceived data and the parameters of each virtual layer, calculate the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection area of ​​each virtual layer and the projection area of ​​each risk target.

[0020] Based on the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection area of ​​each virtual layer and the projection area of ​​each risk target, the competition index between each virtual layer and each risk target is determined.

[0021] Based on the aforementioned technical means, by introducing multi-dimensional quantitative indicators to calculate the competition index between virtual layers and real risk targets, the AR-HUD system can more delicately, proactively, and in accordance with human factors engineering identify which virtual layers are competing with real risk targets for the driver's visual attention. This provides a high-precision and interpretable quantitative basis for subsequent intervention and control, significantly reducing the probability of false intervention and missed intervention, and ultimately more effectively preventing the cognitive tunneling effect.

[0022] Furthermore, the parameters of each virtual layer include the refresh rate, displacement direction vector, and depth plane distance of each virtual layer; the perception data includes the optical flow direction vector and depth data of each risk target.

[0023] Based on the perceived data and the parameters of each virtual layer, the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection area of ​​each virtual layer and the projection area of ​​each risk target are calculated, specifically including:

[0024] Calculate the spatial overlap between the projection area of ​​each virtual layer and the projection area of ​​each risk target;

[0025] Calculate the distance between the center of the projection area of ​​each virtual layer and the center of the projection area of ​​each risk target, and normalize the distance to obtain the proximity.

[0026] The dynamic conflict level is determined based on the refresh rate and displacement direction vector of each virtual layer and the optical flow direction vector of each risk target.

[0027] Depth conflict level is determined based on the depth plane distance of each virtual layer and the depth data of each risky target.

[0028] The cumulative time that the projection area of ​​each virtual layer is in contact with the projection area of ​​each risk target within a preset time period is determined, and the continuous occupancy is determined based on the cumulative time.

[0029] Furthermore, the perception data includes the longitudinal relative distance, longitudinal relative speed, lateral offset, lateral distance from the center point, and lateral speed component of each risk target between the risk target and the target vehicle;

[0030] In S4, based on the perceived data, the environmental risk level of each risk target is determined, specifically including:

[0031] For each risk target, the collision risk between the risk target and the target vehicle is determined based on the longitudinal relative distance and longitudinal relative speed of the risk target relative to the target vehicle.

[0032] The closing speed risk of the risk target is determined based on its longitudinal relative velocity with respect to the target vehicle.

[0033] The proximity of the risk target to the target vehicle is determined based on the lateral offset of the risk target relative to the target vehicle.

[0034] Based on the lateral distance of the risk object relative to the center point of the target vehicle and the lateral velocity component of the risk object, the lane conflict risk between the risk object and the target vehicle is determined;

[0035] Based on the type of the risk target, determine the preset risk coefficient for the risk target;

[0036] The environmental risk level of the target object is determined based on the collision risk, the closing speed risk, the proximity, the lane conflict risk, and the risk coefficient.

[0037] Furthermore, S5 specifically includes:

[0038] For each risk target, the competition index with the highest value is determined from the competition index between the risk target and each virtual layer;

[0039] The cognitive tunnel risk index of the target object is determined based on its environmental risk level and the competition index with the highest value.

[0040] Furthermore, based on the environmental risk level of the target object and the competition index with the highest value, the cognitive tunnel risk index of the target object is determined, specifically including:

[0041] Determine the preset semantic priority weight of the virtual layer corresponding to the competition index with the largest value;

[0042] The cognitive tunnel risk index of the target object is determined based on its environmental risk level, the highest competition index, and the semantic priority weight.

[0043] Furthermore, in S6, based on the competition index between the risk target and each virtual layer, the virtual layers that form a strong competitive relationship with the risk target are determined, specifically including:

[0044] For each risk target, determine the competition index that exceeds the preset competition threshold from the competition index between the risk target and each virtual layer;

[0045] The virtual layer corresponding to the competition index that exceeds the preset competition threshold is determined as the virtual layer that forms a strong competitive relationship with the risk target.

[0046] Furthermore, in S7, according to preset intervention and control operations, intervention and control are performed on virtual layers that have a strong competitive relationship with the risk target, specifically including:

[0047] When the intervention level of the virtual layer corresponding to the risk target is Level 1, reduce the brightness and contrast of the virtual layer that is in strong competition with the risk target.

[0048] Furthermore, in S7, according to preset intervention and control operations, intervention and control are performed on virtual layers that have a strong competitive relationship with the risk target, specifically including:

[0049] When the intervention level of the virtual layer corresponding to the risk target is level two, the virtual layer that has a strong competitive relationship with the risk target will be transferred to the edge of the AR-HUD projection area for two-dimensional display.

[0050] Furthermore, in S7, according to preset intervention and control operations, intervention and control are performed on virtual layers that have a strong competitive relationship with the risk target, specifically including:

[0051] When the intervention level of the virtual layer corresponding to the risk target is level three, the virtual layer that forms a strong competitive relationship with the risk target is hidden.

[0052] The above-mentioned technical solutions adopted in this specification can achieve the following beneficial effects:

[0053] This solution quantifies the competition index between virtual layers and risky targets, combining it with environmental risk levels to quantify the risk of driver attention being "captured" by a single piece of virtual information. When the cognitive tunnel risk index reaches a threshold, proactive intervention is made on virtual layers with strong competition, reducing the risk of driver attention being captured by non-critical virtual information, thereby guiding driver attention back to real road risks and preventing attention fixation at its source. This solution focuses on real risky targets within the "central driving corridor" and dynamically adjusts the display priority of virtual layers based on their real-time risk levels. This ensures that key warning information is effectively highlighted in high-risk scenarios, while non-critical virtual information is suppressed, reducing reaction delays caused by information overload or visual occlusion. By uniformly mapping virtual layers and risky targets to the equivalent imaging plane of the driver's eye box, precise competition analysis is achieved under the same visual coordinate system. This avoids misjudgments caused by misalignment between virtual and real spaces, allowing intervention control to precisely target virtual layers that have a "strong competition" relationship with risky targets, rather than blindly adjusting all displayed content, maintaining the availability of necessary information. Attached Figure Description

[0054] The accompanying drawings, which are included to provide a further understanding of this specification and form part of this specification, illustrate exemplary embodiments and are used to explain this specification, but do not constitute an undue limitation thereof. In the drawings:

[0055] Figure 1 A flowchart illustrating an AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect, provided as an embodiment of this specification;

[0056] Figure 2 This specification provides a corresponding Figure 1 A schematic diagram of the structure of an electronic device. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of this specification clearer, the technical solutions of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in this specification without creative effort are within the scope of protection of this application.

[0058] In embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0059] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0060] Figure 1 A flowchart illustrating an AR-HUD visual focus intervention and control method to prevent cognitive tunneling effect, provided in this specification, includes the following steps:

[0061] S1: Obtain the perception data collected by the vehicle perception system of the target vehicle, and obtain the virtual layers of the AR-HUD projection and the parameters of each virtual layer.

[0062] This specification describes the process of preventing cognitive tunneling effects in AR-HUD visual focus intervention control. In the embodiments described herein, this process can be executed by the AR-HUD system. However, this specification does not limit the type of device or platform used to implement this process; for example, a personal computer, mobile terminal, or other such device or platform can also be used. For ease of description, the AR-HUD system will be used as the executing entity in the following description.

[0063] In one or more embodiments of this specification, the environmental perception system (hereinafter referred to as the perception system) on the target vehicle can acquire perception data such as the forward road scene, information on surrounding traffic participants, and the operating status of the main vehicle (i.e., the target vehicle). The perception data acquired by the perception system can be obtained from sources such as cameras, millimeter-wave radar, lidar, ultrasonic sensors, navigation systems, vehicle bus, and vehicle-to-everything (V2X) information. Therefore, the AR-HUD system can acquire the perception data collected by the perception system. Simultaneously, it can also acquire the various virtual layers projected by the AR-HUD and the parameters of each virtual layer. These virtual layers include, but are not limited to, navigation arrows, route guidance strips, intersection magnification images, point-of-information (POI) prompts, preceding vehicle markings, speed limit information, incoming call prompts, multimedia information, and safety warning icons. It is worth noting that the target vehicle refers to the vehicle undergoing AR-HUD visual focus intervention control to prevent the cognitive tunneling effect.

[0064] S2: Based on the perception data, identify each risk target in the central driving corridor of the target vehicle, wherein the central driving corridor is the area in the target vehicle that the driver can observe through the windshield and that the target vehicle can drive in.

[0065] In one or more embodiments of this specification, the AR-HUD system can identify various risk targets within the central driving corridor of a target vehicle based on perception data. Specifically, it uses data sensed by the perception system to filter for vehicles ahead, pedestrians, non-motorized vehicles, static obstacles, intervening targets, construction areas, or abnormal road events, extracting candidate risk targets located within or about to enter the central driving corridor of the main vehicle. The central driving corridor is the area within the target vehicle that the driver can observe through the windshield and is accessible for the target vehicle to travel.

[0066] S3: Project each virtual layer and each risk target onto the equivalent imaging plane in the driver's eye box.

[0067] In one or more embodiments of this specification, the AR-HUD system projects each virtual layer and each risk target onto an equivalent imaging plane in the driver's eye box.

[0068] The contour points of each risk target surface are transformed from the sensor coordinate system of the sensor collecting data in the perception system to the vehicle coordinate system, and then from the vehicle coordinate system to the driver's eye-view coordinate system, and finally mapped to the equivalent imaging plane via a projection matrix. Similarly, for each virtual layer, the three-dimensional virtual layers projected by each AR-HUD are also transformed to the same equivalent imaging plane. After each virtual layer and each risk target is projected onto the equivalent imaging plane in the driver's eye-view, it can be expressed in various combinations such as bounding boxes and polygons. The equivalent imaging plane is the reference projection plane corresponding to the display distance of the AR-HUD virtual image from the driver's eye-view perspective. In this specification, the technical content of projecting each virtual layer and each risk target onto the equivalent imaging plane in the driver's eye-view can refer to existing coordinate transformation techniques for mapping three-dimensional elements to a two-dimensional plane, and is not protected by this specification, nor is it described in detail here.

[0069] S4: Based on the perceived data and the parameters of each virtual layer, determine the competition index between each virtual layer and each risk target, and determine the environmental risk level of each risk target.

[0070] In one or more embodiments of this specification, the AR-HUD system can determine the competition index between each virtual layer and each risk target, and determine the environmental risk level of each risk target, based on the sensing data and the parameters of each virtual layer.

[0071] Specifically, the method for determining the competition index is as follows: In the equivalent imaging plane, the projection areas of each virtual layer and each risk target are determined. Then, based on the perceived data and the parameters of each virtual layer, the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection areas of each virtual layer and each risk target are calculated. Finally, based on the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection areas of each virtual layer and each risk target, the competition index between each virtual layer and each risk target is determined.

[0072] The parameters for each virtual layer include its refresh rate, displacement direction vector (representing the direction of position change of the virtual layer at each moment), and depth plane distance (representing the depth distance of the virtual layer relative to the driver's eye-box reference viewpoint). The perception data includes the optical flow direction vector (representing the relative motion direction of the risk object) and depth data for each risk target.

[0073] Therefore, the methods for determining spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy can be as follows:

[0074] Calculate the projection area (B) of each virtual layer. j ) and the projection area (H) of each risk target. i Spatial overlap between (O) ij O ij = Area(H i ∩B j ) / Area(H i The overlapping area (H) represents the degree to which virtual elements occlude the risk target area, and is determined by the percentage of the overlapping area in H. i Represented by the proportion of area.

[0075] Calculate the distance between the center of the projection area of ​​each virtual layer and the center of the projection area of ​​each risk target, and normalize this distance to obtain the proximity. Let dist(H) i B j ) represents H i With B j The distance from the center of the projected region is the proximity N. ij = max(0, 1- dist(H i B j ) / d ref ), where d ref This is the nearest reference distance. When two regions overlap, dist(H) is used. i B j When N = 0, N ij =1.

[0076] Based on the refresh rate f of each virtual layer j Displacement direction vector u j The optical flow direction vector v of each risk target object i Determine the dynamic conflict degree M ij M ij = ν1·min(f j / f ref ,1) + ν2·(1-cosθ ij ) / 2, where cosθ ij =(u j ·v i ) / (|u j ||v i |). f ref This is the preset reference refresh rate. When the virtual layer animation is frequent and its dynamic direction differs significantly from the movement direction of the actual risk target, M... ij Increase.

[0077] Based on the depth plane distance z of each virtual layer j and depth data z for each risk target i Determine the degree of conflict D ij D ij = max(0, 1 - |z i -z j | / z ref ), z ref This is a preset depth conflict reference value. When the depth plane of the virtual layer approaches the depth of the actual risk target, D... ij High. Deep conflict reference value z ref The calibration depends on the virtual image distance of the AR-HUD system installed in the target vehicle and the depth-of-field characteristics of human vision. For example, when the depth-of-field plane distance of the AR-HUD is z j When the depth is between 7.5 meters and 15 meters, the depth conflict reference value z ref It can be marked as 15 meters to 30 meters (for example, take z). ref =20 meters), to cover the physical space range where the human eye's retina cannot effectively separate the real and virtual layers through focus adjustment.

[0078] Determine the projection area of ​​each virtual layer over a preset time length T. w The cumulative time t of contact between the projected area of ​​each risk target and the target area. occ,ij And based on the cumulative time t occ,ij Determine the degree of persistent occupancy P ij Preset time length T w The preferred value range is 0.5 seconds to 3.0 seconds. In the embodiment, let T be... w = 2.0 seconds, used to calculate the cumulative percentage of time the virtual layer comes into contact with the risk target within those 2 seconds (i.e., the virtual layer enters the projection area of ​​the risk target), in order to filter out false triggers caused by vehicle bumps or momentary line-of-sight shifts. P ij = t occ,ij / T w .

[0079] Therefore, the method for determining the competition index between each risk target and each virtual layer can be as follows: when the perception system perceives n risk targets and the AR-HUD system determines m virtual layers, C ij = α1·O ij +α2·N ij + α3·M ij + α4·D ij + α5·P ij, where α1 to α5 are preset weight coefficients, satisfying α1 + α2 + α3 + α4 + α5 = 1, i ∈ [1, n], j ∈ [1, m], C ij is the competition index between the i-th risk target and the j-th virtual layer.

[0080] In this specification, the perception data at least includes the longitudinal relative distance, longitudinal relative speed, lateral offset, center point lateral distance between each risk target and the target vehicle, and the lateral speed component of each risk target.

[0081] Thus, the method for determining the environmental risk level can be, for each risk target, according to the longitudinal relative distance d of the risk target (assuming the risk target is the i-th risk target) relative to the target vehicle i , longitudinal relative speed Δv i , determine the time to collision (TTC) of the risk target and the target vehicle i . TTC i = d i / max(Δv i , ε), where ε is a preset positive parameter to prevent the denominator from being 0.

[0082] Of course, further, TTC can also be determined in the following way i .

[0083] When Δv i ≤ 0, TTC i = 0.

[0084] When Δv i > 0, first calculate TTC i = d i / max(Δv i , ε).

[0085] When TTC i ≤ T1, TTC i takes 1;

[0086] When T1 < TTC i < T2, TTC i takes (T2 - TTC i ) / (T2 - T1).

[0087] When TTCi ≥ T2, TTC i takes 0, where T1 is a preset emergency risk threshold, T2 is a preset general risk threshold, and T1 < T2.

[0088] After that, according to the longitudinal relative speed Δv of the risk target relative to the target vehiclei Determine the closure velocity risk f of the target object. v The following method shall be adopted:

[0089] When Δv i When ≤0, f v Take 0.

[0090] When 0 < Δv i <V ref When, f v Take Δv i / V ref V ref This is the preset relative reference speed.

[0091] When Δv i ≥V ref At that time, f v Take 1.

[0092] Then, based on the lateral offset y of the risk target relative to the target vehicle... i Determine the proximity f between the risk target and the target vehicle. y Lateral offset is the distance the risky object deviates from the centerline of the central driving corridor in the lateral direction. f is determined as follows: y :

[0093] When |y i |≥Y ref At that time, f y Take 0.

[0094] When |y i | <Y ref When, f y Take 1- |y i | / Y ref Y ref This is the preset reference value for the half-width of the central driving corridor.

[0095] It can also be based on the lateral distance q of the risk target relative to the center point of the target vehicle. dist,i and the lateral velocity component q of the target object vy,i Determine the lane conflict risk p between the target object and the target vehicle. i,in .

[0096] p i,in =ρ1·q1+ ρ2·q2. ρ1+ρ2=1, where ρ1 and ρ2 are preset weight parameters.

[0097] q1 = max(0, 1 - q) dist,i / q corr,ref ), q corr,refThis is a preset lateral distance quantization parameter.

[0098] When q vy,i ≤0, q2=0.

[0099] When q vy,i >0, q2 = min (1, q vy,i / q vy,ref ), q vy,ref These are preset lateral velocity quantization parameters.

[0100] Subsequently, based on the type of the risk target, the preset risk coefficient k for the risk target is determined. i_type .

[0101] Risk coefficient k i_type The risk target can be pre-marked according to its type: 1.0 for pedestrians, 0.9 for non-motorized vehicles, 0.8 for vehicles cutting in from the front, 0.7 for vehicles in front in the same lane, and 0.6 for general static obstacles.

[0102] Finally, the environmental risk level (Risk) of the target object can be determined based on collision risk, closing speed risk, proximity risk, lane conflict risk, and risk coefficient. i .

[0103] Risk i = λ1·TTC i + λ2·f v + λ3·f y + λ4·p i,in + λ5·k i_type λ1 to λ6 are preset weighting coefficients, and λ1+λ2+λ3+λ4+λ5 =1.

[0104] S5: Determine the cognitive tunnel risk index for each risk target based on each competition index and the environmental risk level of each risk target.

[0105] In one or more embodiments of this specification, the AR-HUD system can determine the cognitive tunnel risk index of each risky target based on each competition index and the environmental risk level of each risky target.

[0106] Specifically, for each risk target, the competition index with the highest value is determined from the competition indices between the risk target and each virtual layer. Then, based on the environmental risk level of the risk target and the highest competition index, the cognitive tunnel risk index of the risk target is determined.

[0107] Furthermore, in this specification, each virtual layer also possesses a semantic priority weight that characterizes its importance to the driving task. This semantic priority weight does not represent the degree of environmental hazard, but rather whether the virtual layer should be preferentially suppressed in the event of competition. For the j-th virtual layer, its semantic priority weight w... j The value ranges from [0,1]. The larger the value, the more likely the virtual layer should be suppressed.

[0108] In this manual, virtual layers are divided into four categories: safety-critical, driving task, secondary information, and comfort / entertainment. Values ​​can be assigned as follows:

[0109] Comfort / Entertainment information is retrieved from w j =1.0;

[0110] Secondary information types (such as POI, incoming calls, media cards) are retrieved using w j =0.8.

[0111] For driving tasks (such as regular navigation guides, intersection zoom-in maps), take w. j =0.5.

[0112] Safety-critical categories (such as forward collision warning, lane departure warning, and emergency braking alert) should be selected using the w parameter. j =0.1.

[0113] Therefore, for each risk target, the competition index with the highest value among the competition indices between that risk target and each virtual layer can be determined. Then, the preset semantic priority weight of the virtual layer corresponding to the highest competition index is determined. Finally, based on the environmental risk level of the risk target, the highest competition index, and the semantic priority weight, the cognitive tunneling risk index of that risk target is determined.

[0114] For the i-th risk target, its cognitive tunnel risk index T i The following calculation method can be used as a reference:

[0115] T i = β1·Risk i + β2·max , , ..., + β3·max , , ..., β1, β2, and β3 are preset weighting coefficients. β1 characterizes the impact of the environmental risk level of the real risk target on the determination of the intervention level. β2 characterizes the impact of the competition intensity of the virtual layer that forms the strongest competitive relationship with the current risk target on the determination of the intervention level. β3 characterizes the impact of the virtual layer that forms the strongest competitive relationship and should be suppressed first on the determination of the intervention level. The equation satisfies β1 + β2 + β3 = 1. In this implementation, β1 = 0.45, β2 = 0.35, and β3 = 0.20 can be used. This scheme focuses more on the competitive state between the "most dangerous real target" and the "virtual element that causes the strongest interference to it," facilitating rapid intervention triggering in the real-time control link.

[0116] S6: For each risk target, determine the virtual layer intervention level corresponding to the risk target based on the cognitive tunnel risk index of the risk target; and determine the virtual layers that form a strong competitive relationship with the risk target based on the competition index between the risk target and each virtual layer.

[0117] In one or more embodiments of this specification, the AR-HUD system can determine the virtual layer intervention level corresponding to each risky target based on its cognitive tunnel risk index and according to preset triggering conditions. In this specification, the virtual layer intervention level is divided into three levels: Level 1, Level 2, and Level 3. Each level corresponds to a different intervention strategy.

[0118] The preset trigger conditions are as follows: when the cognitive tunnel risk index meets the first trigger threshold, a level 1 intervention is triggered; when the cognitive tunnel risk index meets the second trigger threshold, a level 2 intervention is triggered; and when the cognitive tunnel risk index meets the third trigger threshold, a level 3 intervention is triggered. The threshold values ​​are: first trigger threshold < second trigger threshold < third trigger threshold.

[0119] To suppress level fluctuations caused by short-term fluctuations in the cognitive tunnel risk index, a level switching strategy can be set. Only when the cognitive tunnel risk index continuously meets a certain trigger threshold for a preset duration or number of consecutive frames will it switch to the corresponding intervention level.

[0120] The AR-HUD system can also determine the virtual layers that strongly compete with the risk target based on the competition index between the risk target and each virtual layer. Virtual layers whose competition index reaches a certain threshold can be considered as those that strongly compete with the risk target.

[0121] Specifically, for each risk target, the competition index exceeding a preset competition threshold is determined from the competition index between the risk target and each virtual layer. The virtual layers corresponding to the competition indices exceeding the preset threshold are then identified as virtual layers that form a strong competitive relationship with the risk target.

[0122] S7: Based on the intervention level of the virtual layer corresponding to the risk target, and according to the preset intervention control operation, intervene and control the virtual layer that forms a strong competitive relationship with the risk target to reduce the visual competition between the virtual layer and the risk target and prevent the driver from experiencing cognitive tunnel problems.

[0123] In one or more embodiments of this specification, for each risky target, the AR-HUD system can, according to the intervention level of the virtual layer corresponding to the risky target, perform intervention control operations according to preset intervention control operations to intervene and control the virtual layer that forms a strong competitive relationship with the risky target, so as to reduce the visual competition between the virtual layer and the risky target and prevent the driver from experiencing cognitive tunnel problems.

[0124] Specifically, when the intervention level of the virtual layer corresponding to the risk target is Level 1, reduce the brightness and contrast of virtual layers that strongly compete with the risk target. Also, stop the flickering animation of virtual layers that strongly compete with the risk target and weaken texture and edge enhancement. Prioritize retaining virtual layers such as safety warnings, basic vehicle speed, and necessary navigation.

[0125] In this specification, when the intervention level of the virtual layer corresponding to the risk target is Level 2, the virtual layer that strongly competes with the risk target will be moved to the edge of the AR-HUD projection area for 2D display. The virtual layer that strongly competes with the risk target will be cropped, 3D elements will be switched to fixed 2D side prompts, and the update frequency will be reduced. Virtual layers such as collision warning, lane departure warning, and necessary path indication will be prioritized for retention.

[0126] In this manual, when the intervention level of the virtual layer corresponding to the risk target is Level 3, virtual layers that strongly compete with the risk target are hidden. All non-safety-critical layers in the central driving corridor are hidden, retaining only emergency safety warnings. Sound / tactile assistance may be overlaid if necessary. Virtual layers such as forward collision warning, emergency braking warning, and pedestrian / obstacle avoidance warning are prioritized for retention.

[0127] based on Figure 1This paper presents an AR-HUD visual focus intervention and control method to prevent cognitive tunneling. By quantifying the competition index between virtual layers and risky targets, and combining this with the environmental risk level, it quantifies the risk of driver attention being "captured" by a single piece of virtual information. When the cognitive tunneling risk index reaches a threshold, it proactively intervenes in virtual layers with strong competition, reducing the risk of driver attention being captured by non-critical virtual information. This guides the driver's attention back to real road risks, preventing attention fixation at its source. This solution focuses on real risky targets within the "central driving corridor" and dynamically adjusts the display priority of virtual layers based on their real-time risk level. This ensures that key warning information is effectively highlighted in high-risk scenarios, while non-critical virtual information is suppressed, reducing reaction delays caused by information overload or visual occlusion. By mapping virtual layers and risky targets uniformly to the equivalent imaging plane of the driver's eye box, precise competition analysis is achieved under the same visual coordinate system. This avoids misjudgments caused by misalignment between virtual and real spaces, allowing intervention control to precisely target virtual layers with "strong competition" with risky targets, rather than blindly adjusting all displayed content, maintaining the availability of necessary information.

[0128] Furthermore, in one or more embodiments of this specification, after the intervention control operation, the competition index and cognitive tunnel risk index can be continuously recalculated, and the current intervention level can be maintained, weakened, or upgraded based on the competition mitigation situation.

[0129] Furthermore, if the target vehicle is equipped with a driver behavior assistance module, the driver's gaze can be used as one of the criteria for closed-loop verification, whether the driver's gaze returns to the real road risk area or is still focused on the vicinity of the suppressed virtual layer; when the closed-loop verification indicates that the risk has not been eliminated, it can be automatically upgraded to a higher level of rendering suppression strategy.

[0130] When the virtual layer intervention level no longer meets the triggering conditions of the first, second, or third level intervention, the virtual layer enters the recovery phase. It adopts a progressive sequence opposite to the intervention to gradually restore parameters such as transparency, brightness, contrast, spatial binding, depth plane, and update frequency, avoiding instantaneous changes in the image.

[0131] This specification also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 This paper presents an AR-HUD visual focus intervention and control method to prevent cognitive tunneling effect.

[0132] This instruction manual also provides Figure 2 The diagram shows a schematic structural representation of the electronic device. Figure 2As shown, at the hardware level, this electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to achieve the above. Figure 1 This paper presents an AR-HUD visual focus intervention and control method to prevent cognitive tunneling effect.

[0133] Of course, in addition to software implementation, this specification does not exclude other implementation methods, such as logic devices or a combination of hardware and software. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0134] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must also be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0135] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0136] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0137] For ease of description, the above devices are described in terms of function, divided into various units. Of course, in implementing this specification, the functions of each unit can be implemented in one or more software and / or hardware components.

[0138] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0139] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0140] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0141] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0142] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0143] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0144] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic or disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0145] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0146] Those skilled in the art will understand that the embodiments of this specification can be provided as methods, systems, or computer program products. Therefore, this specification may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this specification may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0147] This specification can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This specification can also be practiced in distributed computing environments, where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0148] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0149] The above description is merely an embodiment of this specification and is not intended to limit this specification. Various modifications and variations can be made to this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of the claims of this specification.

Claims

1. A method for visual focus intervention and control in AR-HUD to prevent cognitive tunneling effect, characterized in that, include: S1. Obtain the perception data collected by the vehicle perception system of the target vehicle, and obtain each virtual layer of the AR-HUD projection and the parameters of each virtual layer; S2. Based on the perceived data, determine each risk target in the central driving corridor of the target vehicle, wherein the central driving corridor is the area in the target vehicle that the driver can observe through the windshield and that the target vehicle can drive in; S3. Project each virtual layer and each risk target onto the equivalent imaging plane in the driver's eye box; S4. Based on the perceived data and the parameters of each virtual layer, determine the competition index between each virtual layer and each risk target, and determine the environmental risk level of each risk target. The parameters of each virtual layer include the refresh rate, displacement direction vector, and depth plane distance of each virtual layer. The perceived data includes the optical flow direction vector and depth data of each risk target. S5. Determine the cognitive tunnel risk index for each risk target based on each competition index and the environmental risk level of each risk target; S6. For each risk target, determine the corresponding virtual layer intervention level based on the cognitive tunnel risk index of that risk target; And based on the competition index between the risk target and each virtual layer, determine the virtual layers that form a strong competitive relationship with the risk target; S7. Based on the intervention level of the virtual layer corresponding to the risk target, and according to the preset intervention control operation, intervene and control the virtual layer that forms a strong competitive relationship with the risk target to reduce the visual competition between the virtual layer and the risk target and prevent the driver from experiencing cognitive tunnel problems. In step S4, based on the perceived data and the parameters of each virtual layer, the competition index between each virtual layer and each risk target is determined, specifically including: S41. In the equivalent imaging plane, determine the projection area of ​​each virtual layer and each risk target; S42. Based on the perceived data and the parameters of each virtual layer, calculate the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection area of ​​each virtual layer and the projection area of ​​each risk target; S42 includes calculating the spatial overlap between the projection area of ​​each virtual layer and the projection area of ​​each risk target, 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, and normalizing the distance to obtain the proximity, determining the dynamic conflict based on the refresh frequency and displacement direction vector of each virtual layer and the optical flow direction vector of each risk target, determining the depth conflict based on the depth plane distance of each virtual layer and the depth data of each risk target, determining the cumulative time that the projection area of ​​each virtual layer is in contact with the projection area of ​​each risk target within a preset time length, and determining the persistent occupancy based on the cumulative time; S43. Based on the spatial overlap, proximity, dynamic conflict, depth conflict, and persistent occupancy between the projection area of ​​each virtual layer and the projection area of ​​each risk target, determine the competition index between each virtual layer and each risk target.

2. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, The perception data includes the longitudinal relative distance, longitudinal relative speed, lateral offset, lateral distance from the center point, and lateral speed component of each risk target and the target vehicle; In S4, based on the perceived data, the environmental risk level of each risk target is determined, specifically including: For each risk target, the collision risk between the risk target and the target vehicle is determined based on the longitudinal relative distance and longitudinal relative speed of the risk target relative to the target vehicle. The closing speed risk of the risk target is determined based on its longitudinal relative velocity with respect to the target vehicle. The proximity of the risk target to the target vehicle is determined based on the lateral offset of the risk target relative to the target vehicle. Based on the lateral distance of the risk object relative to the center point of the target vehicle and the lateral velocity component of the risk object, the lane conflict risk between the risk object and the target vehicle is determined; Based on the type of the risk target, determine the preset risk coefficient for the risk target; The environmental risk level of the target object is determined based on the collision risk, the closing speed risk, the proximity, the lane conflict risk, and the risk coefficient.

3. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, S5 specifically includes: For each risk target, the competition index with the highest value is determined from the competition index between the risk target and each virtual layer; The cognitive tunnel risk index of the target object is determined based on its environmental risk level and the competition index with the highest value.

4. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 3, characterized in that, Based on the environmental risk level of the target object and the competition index with the highest value, the cognitive tunnel risk index of the target object is determined, specifically including: Determine the preset semantic priority weight of the virtual layer corresponding to the competition index with the largest value; The cognitive tunnel risk index of the target object is determined based on its environmental risk level, the highest competition index, and the semantic priority weight.

5. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, S6 determines the virtual layers that have a strong competitive relationship with the risk target based on the competition index between the risk target and each virtual layer, specifically including: For each risk target, determine the competition index that exceeds the preset competition threshold from the competition index between the risk target and each virtual layer; The virtual layer corresponding to the competition index that exceeds the preset competition threshold is determined as the virtual layer that forms a strong competitive relationship with the risk target.

6. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, S7 implements preset intervention and control operations to intervene and control virtual layers that strongly compete with the risk target, specifically including: When the intervention level of the virtual layer corresponding to the risk target is Level 1, reduce the brightness and contrast of the virtual layer that is in strong competition with the risk target.

7. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, S7 implements preset intervention and control operations to intervene and control virtual layers that strongly compete with the risk target, specifically including: When the intervention level of the virtual layer corresponding to the risk target is level two, the virtual layer that has a strong competitive relationship with the risk target will be transferred to the edge of the AR-HUD projection area for two-dimensional display.

8. The AR-HUD visual focus intervention and control method for preventing cognitive tunneling effect as described in claim 1, characterized in that, S7 implements preset intervention and control operations to intervene and control virtual layers that strongly compete with the risk target, specifically including: When the intervention level of the virtual layer corresponding to the risk target is level three, the virtual layer that forms a strong competitive relationship with the risk target is hidden.

Images