Electronic devices and methods for configuring head-up displays

Abstract

This invention provides an electronic device and method for configuring a head-up display (HUD). The method includes: obtaining a first coordinate value corresponding to a driver's line of sight, and obtaining a statistical value corresponding to the first coordinate value; inputting the first coordinate value and the statistical value into a quantum machine learning model to obtain a first predicted statistical value; and configuring the display area of ​​the HUD according to the first coordinate value and the first predicted statistical value. This invention can dynamically adjust the display area of ​​the HUD.

Description

Technical Field

[0001] This invention relates to an electronic device and method for configuring a head-up display. Background Technology

[0002] Currently, head-up display (HUD) systems have become standard equipment in vehicles such as airplanes and automobiles. HUDs project instrument information into the driver's line of sight, preventing distraction caused by frequent glances down at the dashboard. With the advent of augmented reality (AR) technology, the information provided by HUD systems has become more diverse. For example, HUD systems can use devices such as cameras to acquire information about the vehicle's external environment, generate navigation information based on this information, and project it onto the windshield.

[0003] While HUD technology has become increasingly complex and plays an increasingly important role, it still faces a persistent problem: the projected image from the HUD may obstruct the driver's view, reducing their awareness of road conditions. Therefore, preventing the HUD's projected image from obstructing the driver's view is one of the goals that researchers in this field are dedicated to. Summary of the Invention

[0004] The present invention provides an electronic device and method for configuring a head-up display, which can dynamically adjust the display area of ​​the head-up display.

[0005] An electronic device for configuring a head-up display (HUD) according to the present invention includes a processor, a storage medium, and a transceiver. The storage medium stores a quantum machine learning model. The processor is coupled to the storage medium and the transceiver, wherein the processor is configured to perform: obtaining a first coordinate value corresponding to a driver's line of sight via the transceiver, and obtaining a statistical value corresponding to the first coordinate value; inputting the first coordinate value and the statistical value into the quantum machine learning model to obtain a first predicted statistical value; and configuring the display area of ​​the HUD according to the first coordinate value and the first predicted statistical value.

[0006] In one embodiment of the invention, the processor is further configured to perform: obtaining the speed of the vehicle via a transceiver; and configuring the size of the display area according to the speed.

[0007] In one embodiment of the present invention, the speed described above is negatively correlated with the size.

[0008] In one embodiment of the invention, the processor is further configured to perform: obtaining the average gaze time corresponding to the driver's gaze via a transceiver; and configuring the display content of the head-up display according to the average gaze time, wherein the amount of information in the display content is inversely proportional to the average gaze time.

[0009] In one embodiment of the invention, the processor is further configured to perform: obtaining the speed of the vehicle via a transceiver; and determining the amount of information based on the speed, wherein the amount of information is negatively correlated with the speed.

[0010] In one embodiment of the present invention, the processor is further configured to perform: determining the writing direction based on the content displayed on the head-up display; and determining the position of the displayed content in the display area based on the writing direction.

[0011] In one embodiment of the invention, the processor is further configured to perform: acquiring training data of a quantum machine learning model via a transceiver, wherein the training data includes historical coordinate values, historical statistics, historical speeds corresponding to vehicles, and historical relative traffic flow corresponding to vehicles; calculating ideal statistics corresponding to the driver's line of sight based on the historical coordinate values, historical statistics, historical speeds, and historical relative traffic flow; and training the quantum machine learning model according to a loss function, wherein the loss function is related to the error between the output statistics of the quantum machine learning model and the ideal statistics.

[0012] In one embodiment of the invention, the processor is further configured to perform: obtaining multiple probabilities, each corresponding to a plurality of statistical values, output by a quantum machine learning model based on a first coordinate value and a statistical value; and selecting, based on the multiple probabilities, a candidate statistical value corresponding to the maximum probability from a plurality of candidate statistical values ​​as a first predicted statistical value.

[0013] In one embodiment of the present invention, the processor is further configured to perform: obtaining a second coordinate value corresponding to the line of sight via a transceiver; inputting the second coordinate value and a first predicted statistical value into a quantum machine learning model to obtain a second predicted statistical value; and configuring a display area based on the second coordinate value and the second predicted statistical value.

[0014] In one embodiment of the present invention, the above statistical values ​​include standard deviation and positive or negative sign.

[0015] A method for configuring a head-up display according to the present invention includes: obtaining a first coordinate value corresponding to a driver's line of sight, and obtaining a statistical value corresponding to the first coordinate value; inputting the first coordinate value and the statistical value into a quantum machine learning model to obtain a first predicted statistical value; and configuring the display area of ​​the head-up display according to the first coordinate value and the first predicted statistical value.

[0016] Based on the above, the electronic device of the present invention can dynamically adjust the display area of ​​the head-up display according to the driver's line of sight. In addition to preventing the projected image of the head-up display from continuously obscuring a specific area, thus reducing the driver's understanding of road conditions, the electronic device can also assist the driver in understanding the information displayed on the head-up display more quickly and accurately. Attached Figure Description

[0017] Figure 1 A schematic diagram of an electronic device for configuring a head-up display is shown according to an embodiment of the present invention.

[0018] Figure 2 A flowchart illustrating a method for configuring a head-up display is shown according to an embodiment of the present invention.

[0019] Figure 3 A schematic diagram of set Q is shown according to an embodiment of the present invention.

[0020] Figure 4 A schematic diagram illustrating the display content of a head-up display according to an embodiment of the present invention is shown.

[0021] Figure 5 A flowchart illustrating a method for configuring a head-up display is shown according to an embodiment of the present invention.

[0022] Explanation of symbols in the attached drawings:

[0023] 100. Electronic devices;

[0024] 110. Processor;

[0025] 120. Storage medium;

[0026] 130. Transceiver;

[0027] 200. Quantum machine learning models;

[0028] 40. Center point;

[0029] 400. Display area;

[0030] 41. Perspective;

[0031] 42. Field of vision;

[0032] P, any point in set Q;

[0033] Q, set;

[0034] S201, S202, S203, S204, S205, S206, S501, S502, S503: Steps. Detailed Implementation

[0035] For drivers, head-up displays (HUDs) often distract them, so the projection position and displayed information must be carefully arranged to avoid affecting the driver's ability to view road conditions or assess driving status. To facilitate viewing the information displayed on the HUD, the display area needs to be aligned with the driver's line of sight. However, a driver's line of sight is influenced by multiple factors. Predicting a driver's line of sight using machine learning techniques would require a very large amount of training samples and computational resources. Therefore, this invention proposes a method for configuring HUDs based on quantum mechanical visual flow theory. This method improves the communication process between the driver and the vehicle by using parameters such as gaze duration, vehicle speed, and traffic flow, intelligently adjusting the HUD's display area and content. This invention avoids HUDs distracting the driver, thereby improving driving safety.

[0036] Figure 1 A schematic diagram of an electronic device 100 for configuring a head-up display is illustrated according to an embodiment of the present invention. The electronic device 100 includes a processor 110, a storage medium 120, and a transceiver 130.

[0037] Processor 110 may be, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose microcontroller (MCU), microprocessor, digital signal processor (DSP), programmable controller, application-specific integrated circuit (ASIC), graphics processing unit (GPU), image signal processor (ISP), image processing unit (IPU), arithmetic logic unit (ALU), complex programmable logic device (CPLD), field-programmable gate array (FPGA), or other similar elements or combinations thereof. Processor 110 may be coupled to storage medium 120 and transceiver 130, and access and execute multiple modules and various applications stored in storage medium 120.

[0038] Storage medium 120 may be any type of fixed or removable random access memory (RAM), read-only memory (ROM), flash memory, hard disk drive (HDD), solid-state drive (SSD), or similar elements or combinations thereof, used to store multiple modules or various applications executable by processor 110. In this embodiment, storage medium 120 may store multiple modules containing quantum machine learning model 200, the functions of which will be described later.

[0039] Transceiver 130 transmits and receives signals wirelessly or via a wired connection. Transceiver 130 can also perform operations such as low-noise amplification, impedance matching, mixing, up or down frequency conversion, filtering, amplification, and similar functions.

[0040] Figure 2 A flowchart illustrating a method for configuring a head-up display according to an embodiment of the present invention is shown, wherein the method may be performed by, for example Figure 1 The electronic device 100 shown is implemented. It should be noted that... Figure 2 The flowchart is not intended to restrict the execution order of steps S203 to S205. For example, step S205 may be executed before or after step S204.

[0041] Electronic device 100 can be communicatively connected to a head-up display (HUD) via transceiver 130, and the HUD is configured to project the image displayed on the HUD onto a display area on a medium (e.g., a windshield). In step S201, processor 110 can obtain the coordinate values ​​(x(t), y(t)) corresponding to the driver's line of sight via transceiver 130, and obtain the statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) corresponding to the coordinate values, where x(t) is the X-coordinate of the position the driver's line of sight is focused on at time t, y(t) is the Y-coordinate of the position the driver's line of sight is focused on at time t, Δx(t-ΔT) is the standard deviation of the driver's line of sight on the X-axis at time (t-ΔT), and Δy(t-ΔT) is the standard deviation of the driver's line of sight on the Y-axis at time (t-ΔT), where ΔT is the sampling time interval. The aforementioned X-axis and Y-axis can correspond to a Cartesian coordinate system. The units for the coordinate values ​​mentioned above may include, but are not limited to, pixels.

[0042] In one embodiment, processor 110 can be communicatively connected to a camera via transceiver 130, wherein the camera is used to capture an image of the driver's face. Processor 110 can perform iris tracking on the image to determine the position of the driver's gaze within the windshield, thereby obtaining coordinate values ​​(x(t), y(t)) based on the position. That is, the coordinate values ​​(x(t), y(t)) are related to the driver's fixation or refixation of the windshield.

[0043] A driver's two eyes may gaze at different positions on the windshield. Accordingly, processor 110 can perform binocular vision-related corrections on the coordinate values ​​(x(t), y(t)). For example, processor 110 can obtain two coordinate values ​​corresponding to the gazes of the two eyes respectively, and calculate the average of the two coordinate values ​​to produce the corrected coordinate values ​​(i.e., coordinate values ​​((x(t), y(t))). Taking coordinate value x(t) as an example, as shown in equation (1), x1(t) is the X-coordinate value of the position gazed at by the driver's left eye at time point t, and x2(t) is the X-coordinate value of the position gazed at by the driver's right eye at time point t.

[0044]

[0045] Having obtained the coordinate values ​​(x(t), y(t)), the processor 110 can further obtain the statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) corresponding to time point t. The statistical values ​​may include the standard deviation and a plus or minus sign. The statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) can be obtained from two sources. Specifically, the quantum machine learning model 200 can generate statistical values ​​corresponding to the current time point based on the statistical values ​​corresponding to previous time points. For example, if the processor 110 detects that the quantum machine learning model 200 has output statistical values ​​(±Δx(t-2ΔT), ±Δy(t-2ΔT)), the processor 110 can input (±Δx(t-2ΔT), ±Δy(t-2ΔT)) into the quantum machine learning model 200 to obtain the statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)), where ΔT is the sampling time interval. On the other hand, if the processor 110 does not detect any statistical values ​​output by the quantum machine learning model 200, it means that the processor 110 cannot obtain statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) using the quantum machine learning model 200. Accordingly, the processor 110 can calculate statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) based on historical data sampled before time point (t-ΔT). For example, the processor 110 can calculate statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) based on multiple historical coordinate values ​​such as coordinate values ​​(x(tn), y(tn)), coordinate values ​​(x(t-n+1), y(t-n+1)), ..., coordinate values ​​(x(t-3ΔT), y(t-3ΔT)) and coordinate values ​​(x(t-2ΔT), y(t-2ΔT)). In short, if the statistical value (±Δx(t-ΔT), ±Δy(t-ΔT)) is the initial statistical value, then the processor 110 calculates the statistical value (±Δx(t-ΔT), ±Δy(t-ΔT)) based on the historical coordinate values. If the statistical value (±Δx(t-ΔT), ±Δy(t-ΔT)) is not the initial statistical value, then the processor 110 obtains the statistical value (±Δx(t-ΔT), ±Δy(t-ΔT)) using the quantum machine learning model 200.

[0046] In one embodiment, the processor 110 can also determine the duration for which the driver's gaze lingers on a specific location on the windshield using iris tracking technology. Taking coordinate values ​​(x(t), y(t)) as an example, the processor 110 can determine the duration ΔD(t) for which the driver is looking at coordinate values ​​(x(t), y(t)) using iris tracking technology. If there is text projected by a head-up display on the coordinate values ​​(x(t), y(t)), the processor 110 can define the duration ΔD(t) as the gaze duration corresponding to time point t. The processor 110 can calculate the driver's average gaze duration based on multiple gaze durations corresponding to multiple different time points. The faster a driver reads, the shorter their average gaze time. Generally, the shorter the length.

[0047] The driver's two eyes may correspond to different gaze durations. Accordingly, the processor 110 may perform binocular vision-related corrections on the gaze duration ΔD(t). In one embodiment, the processor 110 may obtain two gaze durations corresponding to the lines of sight of the two eyes respectively, and calculate the average of the two gaze durations to produce a corrected gaze duration (i.e., gaze duration ΔD(t)), as shown in Equation (2), where ΔD1(t) is the gaze duration corresponding to the driver's left eye and ΔD2(t) is the gaze duration corresponding to the driver's right eye.

[0048]

[0049] In one embodiment, the processor 110 may select the longer of the two gaze times as the corrected gaze time (i.e., gaze time ΔD(t)), as shown in Equation (3), where ΔD1(t) is the gaze time corresponding to the driver's left eye and ΔD2(t) is the gaze time corresponding to the driver's right eye.

[0050] ΔD(t)=max(ΔD1(t),ΔD2(t))…(3)

[0051] After obtaining the coordinate values ​​(x(t), y(t)) and statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)), in step S202, the processor 110 inputs the coordinate values ​​(x(t), y(t)) and statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)) into the quantum machine learning model 200 to obtain the predicted statistical values ​​(±Δx(t), ±Δy(t)). Specifically, the quantum machine learning model 200 can output multiple probabilities corresponding to multiple candidate statistical values ​​based on the input coordinate values ​​(x(t), y(t)) and statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)). The processor 110 can select the candidate statistical value corresponding to the maximum probability from the multiple candidate statistical values ​​based on the multiple probabilities as the predicted statistical value (±Δx(t), ±Δy(t)).

[0052] Figure 3 A schematic diagram of a set Q is illustrated according to an embodiment of the present invention. A quantum machine learning model 200 can define a set Q based on the visual flow theory of quantum mechanics, according to the input coordinate values ​​(x(t), y(t)) and statistical values ​​(±Δx(t-ΔT), ±Δy(t-ΔT)), where the set Q contains all candidates for the predicted statistical value (±Δx(t), ±Δy(t)). The quantum machine learning model 200 also generates a probability corresponding to each candidate in the set Q. Taking the candidate statistical value (±Δx1(t), ±Δy1(t)) in the set Q as an example, the quantum machine learning model 200 can generate a probability Ω(±Δx1(t), ±Δy1(t)) corresponding to the candidate statistical value (±Δx1(t), ±Δy1(t)). If the candidate statistics (±Δx1(t), ±Δy1(t)) have the highest probability in set Q, then processor 110 can select the candidate statistics (±Δx1(t), ±Δy1(t)) from set Q as the predicted statistics (±Δx(t), ±Δy(t)). That is, the probability Ω (±Δx1(t), ±Δy1(t)) of the candidate statistics (±Δx1(t), ±Δy1(t)) must satisfy the constraint of the formula (4), where P is any point in set Q.

[0053]

[0054] The quantum machine learning model 200 can be trained by the processor 110 based on, for example, a quantum annealing algorithm. Specifically, the processor 110 can connect to an external sensor (e.g., a camera, a LiDAR, or a speedometer) via a transceiver 130 to obtain training data for the quantum machine learning model 200 from the external sensor. The training data may include historical coordinate values ​​(x(k), y(k)), historical statistics (±Δx(k-ΔT), ±Δy(k-ΔT)), historical speed v(k) corresponding to the vehicle's speed, and historical relative traffic flow r(k) corresponding to the vehicle, where time point k precedes time point t. The historical relative traffic flow r(k) represents the number of other vehicles that approached the vehicle per unit time before time point k. The processor 110 can calculate the ideal statistics corresponding to the driver's line of sight (I(±Δx(k)), I(±Δy(k))) according to equation (5), where Cx and Cy are predefined constants.

[0055]

[0056] Processor 110 can train quantum machine learning model 200 according to loss function L, where loss function L is used to represent the error between the output statistics of quantum machine learning model 200 and the ideal statistics (I(±Δx(k)), I(±Δy(k))), as shown in equation (6), where Let I represent the output statistics of the quantum machine learning model 200, and let I represent the ideal statistics (i.e., (I(±Δx(k)), I(±Δy(k)))). Representative output statistics The error between the target value I and the ideal statistical value, wherein the error includes, but is not limited to, mean square error (MSE) or mean absolute error (MAE). Processor 110 can input historical coordinate values ​​(x(k), y(k)) and historical statistical values ​​(±Δx(k-ΔT), ±Δy(k-ΔT)) into the quantum machine learning model 200 during training to obtain the output statistical value. The quantum machine learning model 200 can be trained by minimizing the loss function L.

[0057]

[0058] Back Figure 2 In step S203, the processor 110 can determine the position of the display area of ​​the head-up display based on the coordinate values ​​(x(t), y(t)) and the predicted statistical values ​​(±Δx(t), ±Δy(t)). Figure 4A schematic diagram illustrating the display content of a head-up display is shown according to an embodiment of the present invention. In one embodiment, the processor 110 can calculate the coordinates (C1, C2) of the center point 40 of the display area 400 according to the program (7) to determine the position of the display area 400.

[0059]

[0060] The displayed content may include, but is not limited to, dashboard information, navigation information, or warning information. Taking the application of adaptive cruise control (ACC) or AR systems as an example, assuming that the vehicle is equipped with the electronic device 100 of this invention, when the vehicle encounters an obstacle ahead, in addition to the ACC system automatically reducing speed, the electronic device 100 may be equipped with a head-up display to project warning information onto the display area, thereby alerting the driver that the ACC system has intervened in the vehicle's operation and preventing the driver from panicking.

[0061] In step S204, the processor 110 can determine the size of the display area 400 and the display content to be projected onto the display area 400 by the head-up display. Specifically, the processor 110 can configure the size of the display area 400 according to the current vehicle speed v(t). The processor 110 can obtain the speed v(t) by connecting to the vehicle's speedometer via transceiver 130. On the other hand, the processor 110 can also determine the size of the display area 400 based on the driver's average gaze time. Configure the displayed content.

[0062] In one embodiment, the vehicle's speed (e.g., speed v(t)) is negatively correlated with the size of the display area 400. The angle of view (AoV) of the human eyes is related to the vehicle's speed. As the vehicle's speed increases, the driver's angle of view narrows, and the driver's field of view (FoV) over the windshield also narrows, as illustrated in Table 1. To ensure that all content of the display area 400 is readable by the driver, the processor 110 can refer to Table 1 to configure the length of the display area 400 in the horizontal direction (i.e., the X-axis) so that the display area 400 does not exceed the driver's field of view, as shown in Table 1. Figure 4 As shown, the driver's perspective 41 and the driver's field of vision 42 over the windshield can be clearly defined.

[0063] Table 1

[0064] 40 100 70 65 100 40

[0065] In one embodiment, the driver's average gaze time The speed of the vehicle is inversely proportional to the amount of information displayed. The driver's reaction time (i.e., the time required from the driver's perception of danger to the driver's effective braking) is also inversely proportional. As the vehicle's speed increases, the driver's reaction time will decrease. Considering driving safety, the time it takes for the driver to read the displayed content must be less than the driver's reaction time. If the amount of information that the displayed content can contain is quantified as a maximum number of characters M, then the maximum number of characters M must satisfy the constraint of equation (8), where RT(v(t)) represents the driver's reaction time when the vehicle's speed is v(t).

[0066]

[0067] In one embodiment, storage medium 120 may pre-store a mapping table relating vehicle speed to driver reaction time. Processor 110 may determine the maximum number of words M based on the mapping table. For example, assuming the driver's average gaze time... The head-up display (HUD) is expected to contain 5 characters, with a speed of 30 milliseconds. The processor 110 determines the driver's reaction time RT(v(t)) to be 1600 milliseconds based on the vehicle's speed and the relationship mapping table. Since 30 * 5 < 1600, the processor 110 can determine that the 5 characters in the displayed content do not exceed the maximum number of characters M. Therefore, all 5 characters can be included in the displayed content. Equation (8) shows that the amount of information in the displayed content is positively correlated with the reaction time and negatively correlated with the vehicle's speed. That is, the faster the vehicle's speed, the less information the HUD can display. The slower the vehicle's speed, the more information the HUD can display.

[0068] In step S205, the processor 110 determines the position of the displayed content within the display area 400. Specifically, the processor 110 determines the writing direction of the content displayed on the head-up display and determines the position of the displayed content within the display area 400 based on the writing direction. Assuming a sentence is written from left to right, when a reader is reading the sentence, less attention is allocated to the left side of the sentence. Most attention is allocated to the right side, which contains text being read or not yet read. This right side is called the perceptual span. Accordingly, the processor 110 determines which part of the display area belongs to the perceptual span based on the writing direction of the displayed content, and then positions the displayed content within the display area 400 in a position that is more inclined towards the perceptual span.

[0069] by Figure 4For example, the display content of display area 400 includes six characters: "8", "0", "k", "m", " / ", and "h". Since the displayed content is written from left to right, processor 110 can determine that the driver's perceptual span is located in the right region of display area 400 (i.e., to the right of center point 40 or in the positive X direction of center point 40). Accordingly, processor 110 can arrange most of the displayed content to the right of center point 40 and a small portion of the displayed content to the left of center point 40 (i.e., in the negative X direction of center point 40). Figure 4 In the process, the processor 110 will place four characters from the displayed content to the right of the center point 40 and two characters from the displayed content to the left of the center point 40.

[0070] In step S206, the processor 110 can configure the head-up display via the transceiver 130 according to parameters such as the display content, the position of the display content, the size of the display area, and the position of the display area.

[0071] Figure 5 A flowchart illustrating a method for configuring a head-up display according to an embodiment of the present invention is shown, wherein the method may be performed by, for example Figure 1 The illustrated electronic device 100 is implemented. In step S501, a first coordinate value corresponding to the driver's line of sight is obtained, and a statistical value corresponding to the first coordinate value is also obtained. In step S502, the first coordinate value and the statistical value are input into a quantum machine learning model to obtain a first predicted statistical value. In step S503, the display area of ​​the head-up display is configured according to the first coordinate value and the first predicted statistical value.

[0072] In summary, the electronic device of this invention predicts the driver's gaze location based on their line of sight and a quantum machine learning algorithm, and positions the head-up display area at that location. By dynamically adjusting the position of the display area, specific areas of the windshield can be prevented from being continuously obscured. The electronic device can determine the size of the display area or the amount of displayed content based on factors such as the driver's gaze duration, vehicle speed, and traffic flow, avoiding displaying excessive information to the driver when their reaction time is short, thus preventing distraction. Depending on the writing direction of the text displayed on the head-up display, the electronic device can adjust the position of the displayed content within the display area based on perceptual span theory, assisting the driver in understanding the information displayed on the head-up display more quickly and accurately.

Claims

1. An electronic device for configuring a head-up display, characterized in that, include: transceiver; Storage medium for storing quantum machine learning models; as well as A processor, coupled to the storage medium and the transceiver, wherein the processor is configured to perform: The transceiver obtains a first coordinate value corresponding to the driver's line of sight and a statistical value corresponding to the first coordinate value; wherein, the two coordinate values ​​of the lines of sight of the two eyes are averaged to obtain the first coordinate value; The first coordinate value and the statistical value are input into the quantum machine learning model to obtain the first predicted statistical value; Configure the display area of ​​the head-up display according to the first coordinate value and the first predicted statistical value; The transceiver acquires the training data of the quantum machine learning model, wherein the training data includes historical coordinate values, historical statistical values, historical speeds corresponding to the vehicle, and historical relative traffic flow corresponding to the vehicle; the historical relative traffic flow r(k) represents the number of other vehicles that approached the vehicle per unit time before time point k; Calculate the ideal statistical value corresponding to the driver's line of sight based on the historical coordinate values, the historical statistical values, the historical speed, and the historical relative traffic flow; as well as The quantum machine learning model is trained based on a loss function, wherein the loss function is related to the error between the output statistics of the quantum machine learning model and the ideal statistics.

2. The electronic device as claimed in claim 1, characterized in that, The processor is further configured to execute: The speed of the vehicle is obtained through the transceiver; and The size of the display area is configured according to the speed.

3. The electronic device as claimed in claim 2, characterized in that, The speed is negatively correlated with the size.

4. The electronic device as claimed in claim 1, characterized in that, The processor is further configured to execute: The transceiver obtains the average gaze duration corresponding to the driver's line of sight; and The head-up display is configured to display content based on the average gaze time, wherein the amount of information in the display content is inversely proportional to the average gaze time.

5. The electronic device as claimed in claim 4, characterized in that, The processor is further configured to execute: The speed of the vehicle is obtained through the transceiver; and The amount of information is determined based on the speed, wherein the amount of information is negatively correlated with the speed.

6. The electronic device as claimed in claim 1, characterized in that, The processor is further configured to execute: The writing direction is determined based on the content displayed on the head-up display; and The position of the displayed content in the display area is determined according to the writing direction.

7. The electronic device as claimed in claim 1, characterized in that, The processor is further configured to execute: The quantum machine learning model outputs multiple probabilities corresponding to multiple candidate statistical values ​​based on the first coordinate value and the statistical value. as well as Based on the plurality of probabilities, a candidate statistical value corresponding to the maximum probability is selected from the plurality of candidate statistical values ​​as the first predicted statistical value.

8. The electronic device as claimed in claim 1, characterized in that, The processor is further configured to execute: The transceiver obtains a second coordinate value corresponding to the line of sight. The second coordinate value and the first predicted statistical value are input into the quantum machine learning model to obtain the second predicted statistical value; as well as The display area is configured based on the second coordinate value and the second predicted statistical value.

9. The electronic device as claimed in claim 1, characterized in that, The statistical values ​​include the standard deviation and the sign.

10. A method for configuring a head-up display, characterized in that, include: Obtain the first coordinate value corresponding to the driver's line of sight, and obtain the statistical value corresponding to the first coordinate value; wherein, the two coordinate values ​​of the lines of sight of the two eyes are averaged to obtain the first coordinate value; The first coordinate value and the statistical value are input into the quantum machine learning model to obtain the first predicted statistical value; Configure the display area of ​​the head-up display according to the first coordinate value and the first predicted statistical value; Obtain training data for the quantum machine learning model, wherein the training data includes historical coordinate values, historical statistical values, historical speeds corresponding to vehicles, and historical relative traffic flow corresponding to vehicles; the historical relative traffic flow r(k) represents the number of other vehicles that approached the vehicle per unit time before time point k; Calculate the ideal statistical value corresponding to the driver's line of sight based on the historical coordinate values, the historical statistical values, the historical speed, and the historical relative traffic flow; as well as The quantum machine learning model is trained based on a loss function, wherein the loss function is related to the error between the output statistics of the quantum machine learning model and the ideal statistics.

Images