Vehicle-mounted and road scene traffic information quantity collaborative threshold determination method

Abstract

The invention discloses a vehicle-mounted and road scene traffic information quantity collaborative threshold determination method, which comprises the following steps of: respectively dividing a traffic information source of a road scene and display elements of a vehicle-mounted interactive interface into a plurality of categories, calculating information quantity of each category based on an information entropy model, and determining weight of each category, respectively obtaining the total information amount of the road scene and the total information amount of the vehicle-mounted interaction interface; constructing a function of the traffic information amount and the gaze entropy based on a Wundt curve; acquiring eye movement gazing data of a driver in a combined driving scene with different total information amounts of a road scene and a vehicle-mounted interaction interface, and calculating a gazing entropy value; and by taking the traffic information amount as input and the gaze entropy value as output, performing parameter fitting on the function to obtain a minimum threshold value and a maximum threshold value of the traffic information amount. According to the method, the gaze probability distribution differences corresponding to different traffic information amounts are reflected through the gaze entropy values, so that the information amount change can be realized through quantitative indexes, and the influence of information complexity on driving behaviors is truly reflected.

Description

Technical Field

[0001] This invention relates to the field of intelligent transportation technology, and in particular to a method for determining the collaborative threshold of traffic information in vehicle and road scenarios. Background Technology

[0002] In driving, drivers continuously acquire external information to reduce environmental uncertainty, thereby completing a series of actions including perception, decision-making, and operation. With the development of vehicle-to-everything (V2X) technology, in-vehicle interfaces can provide rich information about people, vehicles, roads, and the environment. However, a driver's information processing capacity has a natural limit. When road scenes and in-vehicle interfaces simultaneously present a large amount of information, it triggers information competition and resource consumption, leading to problems such as insufficient information resulting in inadequate perception or information overload causing distraction. Currently, at least the following problems exist: 1. Road scene information and vehicle interface information belong to two independent systems, namely "road-side" and "vehicle-side," lacking a unified measurement standard. Existing research usually only determines the information threshold for one side of the vehicle interface or the road scene independently, without considering both as a joint information source in driving tasks for unified evaluation. This one-sided threshold method cannot reflect the synergistic effect between road scene information and vehicle interface information, and it is difficult to comprehensively characterize the overall information load faced by drivers when performing driving tasks, exhibiting significant limitations in real-world driving environments.

[0003] 2. Existing research focuses more on information overload and less on information shortage. There is a lack of models that can simultaneously describe the two-sided effects of "information shortage" and "information overload," thus making it impossible to determine the optimal range of information content. Summary of the Invention

[0004] To address the technical problems existing in the background art, this invention proposes a method for determining the collaborative threshold of traffic information in vehicle and road scenarios, which can provide drivers with a suitable information range and is suitable for engineering design guidance.

[0005] This invention proposes a method for determining the collaborative threshold of traffic information in vehicle and road scenarios, comprising: Traffic information sources and display elements of the in-vehicle interface in road scenes are divided into multiple categories. The information content of each category is calculated based on the information entropy model, and the weight of each category is determined. The total information content of the road scene is obtained by weighted summation. Total information content of the in-vehicle interface ; Constructing a function of traffic information content and gaze entropy based on Wundt curves The traffic information volume is the total information volume of the road scene. Total information content of the in-vehicle interface sum; Obtain the total amount of information about the driver in the road scene. Total information content of the in-vehicle interactive interface Eye-tracking gaze data under different driving scenarios, and gaze entropy value calculated based on eye-tracking gaze data; Using the traffic information as input and the gaze entropy value as output, the function... By performing parameter fitting, the minimum threshold of the traffic information is obtained. With the maximum threshold .

[0006] Furthermore, functions constructed based on Wundt curves for: in, It is the reward function for information. It is a penalty function for information; It is a constant, indicating the maximum reward value. It is a constant, indicating the maximum penalty value; This represents the minimum information threshold. This represents the maximum information content threshold; , It's the slope.

[0007] Furthermore, the calculation step of the gaze entropy value includes: dividing the driver's gaze point into multiple regions of interest, calculating the gaze entropy information contained in each region of interest, and the formula for calculating the total gaze entropy value is as follows:

[0008] in, Indicates the number of regions of interest; , Representative at the The gaze entropy information contained in each region of interest The driver's gaze point falls on the first The probability of each region of interest; , Represented as the maximum gaze entropy value; For the driver in the Average fixation time for each region of interest.

[0009] Furthermore, methods for dividing gaze points into regions of interest include mechanical partitioning, stepwise statistical analysis, dynamic clustering, and K-Means clustering.

[0010] Furthermore, the optimization algorithm used for parameter fitting is either gradient descent or the Adam optimizer.

[0011] Furthermore, methods for determining the weights of each category include the analytic hierarchy process (AHP), entropy weighting, best-worst method, or the Delphi expert method.

[0012] Furthermore, the categories of traffic information sources in the road scene include: meaning-based, motion-based, physical-based, and environmental-based; the categories of display elements in the in-vehicle interactive interface include: Chinese characters, English letters, Arabic numerals, characters, colors, and icons.

[0013] Furthermore, the information entropy model includes the Shannon entropy model, the Renyi entropy model, or the Tsallis entropy model.

[0014] This invention achieves the same-source calculation of information content in the vehicle interface and road scene, and reflects the difference in gaze probability distribution corresponding to different traffic information content through gaze entropy value, so that the change in information content can be realized through quantitative indicators, and truly reflects the impact of information complexity on driving behavior. Attached Figure Description

[0015] Figure 1 A graph showing the relationship between traffic information volume and information utility; Figure 2 For the present invention, the function A diagram illustrating the training steps of the parameter fitting algorithm; Figure 3 This is a flowchart illustrating a specific embodiment of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will now be described in detail with reference to the accompanying drawings and embodiments. The embodiments are implemented based on the technical solution of the present invention, providing detailed implementation methods and specific operating procedures. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them, and the scope of protection of the present invention is not limited to the following embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0017] This invention proposes a method for determining the collaborative threshold of traffic information in vehicle and road scenarios, comprising: Traffic information sources and display elements of the in-vehicle interface in road scenes are divided into multiple categories. The information content of each category is calculated based on the information entropy model, and the weight of each category is determined. The total information content of the road scene is obtained by weighted summation. Total information content of the in-vehicle interface ; Constructing a function of traffic information content and gaze entropy based on Wundt curves Traffic information volume is the total information volume of the road scene. Total information content of the in-vehicle interface sum; Obtain the total amount of information about the driver in the road scene Total information content of the in-vehicle interactive interface Eye-tracking gaze data under different driving scenarios, and gaze entropy value calculated based on eye-tracking gaze data; Using traffic information as input and gaze entropy as output, the function... By performing parameter fitting, the minimum threshold of traffic information is obtained. With the maximum threshold .

[0018] like Figure 1 As shown, in the function In the corresponding curve, if the traffic information volume is lower than the minimum threshold... If the information obtained by the driver is insufficient, such as exceeding the maximum threshold, then the driver's information is inadequate. Drivers will experience information overload, leading to distraction, and will only be able to focus on the minimum threshold. With the maximum threshold Within the appropriate range, drivers can obtain a balance between traffic information and attention. This application provides a quantifiable optimal range for the collaborative design of road scene and vehicle interface information.

[0019] like Figure 3 The specific implementation steps will be described below.

[0020] S1, Calculate the total information content of the road scene. Total information content of the in-vehicle interface The details are as follows: S1.1 The traffic information sources in the road scene are systematically classified. In this embodiment, traffic information sources are divided into meaning-based, motion-based, physical, and environmental categories. In other embodiments, the categories can be appropriately expanded or merged according to system requirements without affecting the unified calculation method of information volume. The meanings of each category are as follows: Meaningful information: Traffic signs, road markings, traffic lights, and other information that has indicative meaning; Movement category: Dynamic traffic participants such as vehicles, pedestrians, and non-motorized vehicles; Physical components: Road traffic facilities such as lane markings, curbs, and guardrails; Environmental categories: Information related to environmental disturbances such as greening, building facades, and commercial advertising.

[0021] Constructing traffic information sources Ω in road scenarios: (1) In the formula: For traffic information sources of significance; For sports-related traffic information sources; For physical traffic information sources; It is a source of traffic information related to the environment.

[0022] S1.1.1 Calculate the amount of traffic information source information of significance category.

[0023] (1) Calculate the information content of a single meaning-based traffic information source: Based on Shannon's information theory, the first... The amount of information in each type of meaning : (2) in, For meaning-based information The amount of information; For meaning-based information The probability of occurrence; the unit of information is bit.

[0024] (2) Calculate the total information content of traffic information sources of significance. : (3) in The number of traffic information sources of significance.

[0025] S1.1.2 Calculate the amount of information from motion-related traffic information sources.

[0026] (1) Calculate the information content of a single sports-related traffic information source: Based on Shannon's information theory, the first... Information content of each sports category : (4) in, For sports information The amount of information; For sports information The probability of occurrence; the unit of information is bit.

[0027] (2) Calculate the total information content of sports-related traffic information sources : (5) in, This refers to the number of traffic information sources related to sports.

[0028] S1.1.3 Calculate the amount of information from physical traffic information sources.

[0029] (1) Calculate the information content of a single physical traffic information source: Based on Shannon's information theory, the first... Information content of each physical category : (6) in, For physical information The amount of information; For physical information The probability of occurrence; the unit of information is bit.

[0030] (2) Calculate the total information content of physical traffic information sources : (7) in, This refers to the number of physical traffic information sources.

[0031] S1.1.4 Calculate the amount of environmental traffic information sources.

[0032] (1) Calculate the information content of a single environmental traffic information source: Based on Shannon's information theory, the first... Information content of each environmental category : (8) in, For environmental information The amount of information; For environmental information The probability of occurrence; the unit of information is bit.

[0033] (2) Calculate the total amount of information from environmental traffic information sources. : (9) in, This refers to the number of environmental traffic information sources.

[0034] S1.1.5 Determine the weights of each category of traffic information sources in the road scene and assign values ​​to the weights of different categories of information.

[0035] During driving tasks, drivers perceive different types of traffic information in road scenarios. In this embodiment, the importance of meaning-based, motion-based, physical, and environmental traffic information can be weighted using the analytic hierarchy process.

[0036] (1) Expert scoring: Using the 9-quartile scale, 20 drivers were invited to compare the importance of traffic information in terms of meaning, motion, physics and environment in pairs, and a judgment matrix was established for each driver.

[0037] (2) Summary judgment matrix: Take the geometric mean of each corresponding judgment value of the 20 drivers to obtain the summary judgment matrix.

[0038] (3) The weights are calculated using the eigenvalue method, and the principal eigenvectors are normalized to obtain the weights of the traffic information of the meaning class. Weighting of sports-related traffic information Weight of physical traffic information Weighting of environmental traffic information ; (4) Matrix consistency is analyzed and judged by calculating the consistency ratio (CR) value. If This indicates that the consistency of the judgment matrix is ​​acceptable. Otherwise, the judgment matrix needs to be adjusted and the weights recalculated.

[0039] S1.1.6 Calculate the total traffic information volume of the road scene : (10) S1.2 The display elements of the in-vehicle interactive interface are systematically classified into Chinese characters, English letters, Arabic numerals, symbols, colors, and icons. In other embodiments, the categories can be appropriately expanded or merged according to system requirements without affecting the unified calculation method of information volume.

[0040] S1.2.1 Calculate the information content of Chinese character elements.

[0041] (1) Calculate the information content of a single Chinese character element based on Shannon's information theory. : (11) in, The number of commonly used Chinese characters is 3500; the unit of information is bits.

[0042] (2) Calculate the total amount of Chinese characters on the vehicle interactive interface. : (12) in The number of Chinese characters on the in-vehicle interactive interface.

[0043] S1.2.2 Calculate the information content of English letter elements.

[0044] (1) Calculate the information content of a single English letter based on Shannon's information theory. : (13) in, The total number of English letters is 26; the unit of information is bits.

[0045] (2) Calculate the total amount of English letters on the in-vehicle interactive interface. : (14) in The number of English letters on the in-vehicle interactive interface.

[0046] S1.2.3 Calculate the information content of Arabic numeral elements.

[0047] (1) Calculate the information content of a single Arabic numeral based on Shannon's information theory. : (15) in, The total number of Arabic numerals is 10; the unit of information is bits.

[0048] (2) Calculate the total amount of Arabic numeral information on the vehicle interface. : (16) in The number of Arabic numerals on the in-vehicle interactive interface.

[0049] S1.2.4 Calculate the information content of character elements.

[0050] (1) Calculate the information content of a single character based on Shannon's information theory. : (17) in, The total number of characters is determined based on the actual vehicle infotainment system settings; the unit of information volume is bits.

[0051] (2) Calculate the total amount of character information on the vehicle interface. : (18) in This refers to the number of characters on the in-vehicle interactive interface.

[0052] S1.2.5 Calculate the information content of color elements.

[0053] (1) Calculate the information content of a single color based on Shannon's information theory. : (19) in, The total number of colors is determined based on the actual vehicle infotainment system settings; the unit of information volume is bits.

[0054] (2) Calculate the total amount of color information on the vehicle interface. : (20) in The number of colors on the in-vehicle interactive interface.

[0055] S1.2.6 Calculate the information content of icon elements.

[0056] (1) Calculate the information content of a single icon based on Shannon's information theory. : (twenty one) in, The total number of icons is determined based on the actual vehicle infotainment system settings; the unit of information is bits.

[0057] (2) Calculate the total information content of icons on the vehicle interactive interface. : (twenty two) in This refers to the number of icons on the in-vehicle interface.

[0058] The above embodiments are based on Shannon's information theory for calculating information content. It should be noted that in other embodiments, other entropy or probability models can be used as equivalent substitutes, such as the Renyi entropy model and the Tsallis entropy model, without affecting the unified measurement system of this invention. These information entropy models only change the form of information content representation, but can still be used to quantify traffic information in road scenes and vehicle interfaces, thus maintaining the same measurement effect as the above embodiments.

[0059] S1.2.7 Determine the weights of each category of display elements in the vehicle interface and assign information weights to different categories of display elements.

[0060] During driving tasks, drivers perceive each element on the in-vehicle interactive interface differently. Methods such as entropy weighting, best-worst method, Delphi expert method, and analytic hierarchy process can be used to determine the weights of traffic information sources and display elements of the in-vehicle interactive interface in road scenarios. These methods can reflect the driver's perception of the importance of different information types.

[0061] This embodiment uses the Analytic Hierarchy Process (AHP) as an example to assign weights to the importance of Chinese characters, English letters, numbers, characters, icons, and colors.

[0062] (1) Expert scoring: Using the quartile scale, 20 drivers were invited to compare the importance of Chinese characters, English letters, Arabic numerals, characters, colors, and icons in pairs to establish a judgment matrix for each driver.

[0063] (2) Summary judgment matrix: Take the geometric mean of each corresponding judgment value of the 20 drivers to obtain the summary judgment matrix.

[0064] (3) The weights are calculated using the eigenvalue method, and the principal eigenvectors are normalized to obtain the weights of the Chinese character elements. Weight of English letter elements Arabic numeral element weights Character element weight Color element weight Icon element weight .

[0065] (4) Matrix consistency is analyzed and judged by calculating the consistency ratio (CR) value. If This indicates that the consistency of the judgment matrix is ​​acceptable. Otherwise, the judgment matrix needs to be adjusted and the weights recalculated.

[0066] S1.2.8 Calculate the total information content of the in-vehicle interactive interface : (twenty three) S2, Calculate traffic information volume Traffic information volume Total information content of the road scene Total information content of the in-vehicle interface sum: (twenty four) S3. Constructing a function of traffic information content and gaze entropy based on Wundt curves. .

[0067] like Figure 1 As shown, information utility is defined as the effect of information on reducing drivers' uncertainty about the environment. The Wundt curve is used to reflect the relationship between traffic information volume and information utility. The formula is as follows: in, It is the reward function for information. It is a penalty function for information; It is a constant representing the maximum reward value, typically set to 1. It is a constant that indicates the maximum penalty value, and is usually set to 1; This represents the maximum information content threshold. This represents the minimum information content threshold. , It's the slope.

[0068] S4. Construct the total information volume of the road scene. Total information content of the in-vehicle interactive interface Under different combinations of driving scenarios, obtain the driver's eye-tracking gaze data in the constructed driving scenario, and calculate the gaze entropy value based on the eye-tracking gaze data.

[0069] S4.1 Construct a combination of the total information content of the road scene and the in-vehicle interactive interface.

[0070] 1. Design the total traffic information volume for different road scenarios Total information content of the in-vehicle interface The combination of these elements is used to construct typical driving conditions under different information levels. Among them: in, This represents the maximum total amount of information that can be obtained from the road environment in the experimental scenario. This represents the maximum amount of information that an in-vehicle interactive interface can display.

[0071] 2. Total information content of road scenes At fixed intervals In the interval Discrete values ​​within the inner range, i.e.:

[0072] 3. Total information content of the in-vehicle interactive interface At fixed intervals In the interval Discrete values ​​within the inner range, i.e.:

[0073] 4. Construct a road scene-vehicle interface information combination matrix using a full-factor approach:

[0074] S4.2 Construct driving scenarios to present different combinations of road scene and in-vehicle interface information, and collect eye-tracking data of drivers under different combinations of driving scenarios. Specifically, driving scenarios can be simulated, and test drivers can be recruited to conduct driving simulation experiments. During the test, eye-tracking data can be collected using an eye tracker.

[0075] Sample size Calculated using the following formula:

[0076] In the formula: The significance level is typically 0.05. This represents the probability of a Type II error, typically 0.2. Normal distribution quantiles; Normal distribution Quantiles. This represents the expected effect size.

[0077] S4.3, gaze entropy can be used to characterize the utility of traffic information. The steps for calculating gaze entropy include: dividing the driver's gaze point into different regions of interest using the K-Means clustering algorithm, and calculating the gaze entropy value based on the probability of the gaze point being in a region of interest. The gaze entropy information contained in each region of interest is calculated separately. The formula for calculating the total gaze entropy value is as follows:

[0078] in, Indicates the number of regions of interest; , Representative at the The gaze entropy information contained in each region of interest The driver's gaze point falls on the first The probability of each region of interest; , Represented as the maximum gaze entropy value; For the driver in the Average fixation time for each region of interest.

[0079] It should be noted that mechanical partitioning, step-by-step statistical methods, and dynamic clustering methods can also be used to divide the gaze point into regions of interest and use them to calculate gaze entropy, with the same effect.

[0080] S4.4 Pair the traffic information content of each combination with the corresponding gaze entropy value to obtain the information content-utility fitting dataset: ,in This represents the number of types of traffic information.

[0081] S5. Using traffic information as input and gaze entropy as output, the function... Perform parameter fitting to determine the threshold for traffic information volume.

[0082] Specifically, the widely used gradient descent method is used to minimize the loss function Loss to determine the fitting parameters in formula (25). , , , For fitting the dataset Define the loss function :

[0083] in, These are the fitted values. This is the actual value.

[0084] For the input dataset, through steps such as parameter initialization, loss function calculation, and iterative updates, the optimal parameter configuration is finally found, and the output is... , , , Parameter value.

[0085] like Figure 2 As shown, the algorithm training process is as follows: 1. Initialization , , , Learning rate ; 2. Input: Dataset ; 3. Output: ; 4. Loss function calculation: ; 5. Through Optimizer iteration, ; 6. :

[0086] Among them, the parameters obtained by fitting The maximum threshold for traffic information volume, parameter The minimum threshold for traffic information volume is [value], and the suitable threshold range for traffic information volume is [value]. .

[0087] An appropriate information threshold can prevent drivers from falling into a state of excessive or insufficient cognitive load, and improve the quality of driving decisions and ensure driving safety.

[0088] The present invention has the following technical effects: 1. To address the lack of a unified measurement for road scene information and vehicle interface traffic information in existing technologies, this invention, based on Shannon information theory, constructs information quantity calculation models for both road scene information sources and vehicle interface display elements. Both road scene traffic information sources and vehicle interface display elements are divided into multiple categories. Information quantity is defined based on the probability of occurrence of each category, and weighted integration yields the total information quantity of the road scene and the total information quantity of the vehicle interface. This achieves a unified definition of information quantity on both sides, resolving the comparability issues across interfaces and scenes.

[0089] 2. To address the problem that existing technologies cannot simultaneously describe the dual harms of "information insufficiency" and "information overload," this invention, based on the Wundt curve from psychology, introduces a reward function and a penalty function related to information quantity to construct a two-sided joint function that reflects the utility of traffic information. This model can simulate the real-world behavioral characteristics of drivers experiencing cognitive insufficiency (low utility) when information quantity is low, and information overload (increased penalty) when information quantity is high. It can provide a quantifiable optimal range for the collaborative design of information quantity in road scenarios and in-vehicle interfaces.

[0090] 3. To address the problem that existing technologies cannot quantify the impact of information content on cognitive state from the perspective of drivers' actual visual behavior, this invention introduces dimensionless gaze entropy as a quantitative indicator of traffic information utility. By collecting eye-tracking gaze data of drivers under different information content combinations, the K-Means clustering method is used to divide the gaze point into several regions of interest, and the gaze entropy value is calculated based on the gaze probability distribution of each region of interest. A higher gaze entropy indicates a more dispersed driver attention and a higher cognitive load; a lower gaze entropy indicates a more focused driver attention and more efficient information processing. By using gaze entropy as an objective output of traffic information content-information utility modeling, the information content utility function has a measurable and fittable quantitative basis.

[0091] The term "an embodiment" or "embodiment" as used in this invention refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the invention. In the description of this invention, it should be understood that the terms "first," "second," and "third," etc., in the specification, claims, and accompanying drawings are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0092] This specification provides method operation steps as shown in the embodiments or flowcharts, but based on conventional or non-inventive labor, more or fewer operation steps may be included. The order of steps listed in the embodiments is merely one of many possible execution orders and does not represent the only possible execution order. In actual system or server product execution, the method can be executed in the order shown in the embodiments or drawings, or in parallel (e.g., in a parallel processor or multi-threaded processing environment), or the execution order of steps without timing constraints can be adjusted.

[0093] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for determining the collaborative threshold of traffic information in vehicle and road scenarios, characterized in that, include: Traffic information sources and display elements of the in-vehicle interface in road scenes are divided into multiple categories. The information content of each category is calculated based on the information entropy model, and the weight of each category is determined to obtain the total information content of the road scene. Total information content of the in-vehicle interface ; Constructing a function of traffic information content and gaze entropy based on Wundt curves The traffic information volume is the total information volume of the road scene. Total information content of the in-vehicle interface sum; Obtain the total amount of information about the driver in the road scene. Total information content of the in-vehicle interactive interface Eye-tracking gaze data under different driving scenarios, and gaze entropy value calculated based on eye-tracking gaze data; Using the traffic information as input and the gaze entropy value as output, the function... By performing parameter fitting, the minimum threshold of the traffic information is obtained. With the maximum threshold .

2. The method for determining the collaborative threshold of traffic information in vehicle and road scenarios according to claim 1, characterized in that, Functions built based on Wundt curves for: in, It is the reward function for information. It is a penalty function for information; It is a constant, indicating the maximum reward value. It is a constant, indicating the maximum penalty value; This represents the minimum information threshold. This represents the maximum information content threshold; , It's the slope.

3. The method for determining the collaborative threshold of traffic information volume in vehicle and road scenarios according to claim 1, characterized in that, The steps for calculating the gaze entropy value include: dividing the driver's gaze point into multiple regions of interest, calculating the gaze entropy information contained in each region of interest, and the formula for calculating the total gaze entropy value is as follows: in, Indicates the number of regions of interest; , Representative at the The gaze entropy information contained in each region of interest The driver's gaze point falls on the first The probability of a region of interest; , Represented as the maximum gaze entropy value; For the driver in the Average fixation time for each region of interest.

4. The method for determining the collaborative threshold of traffic information in vehicle and road scenarios according to claim 3, characterized in that, Methods for dividing gaze points into regions of interest include mechanical partitioning, stepwise statistical analysis, and dynamic clustering. Clustering method for partitioning.

5. The method for determining the collaborative threshold of traffic information volume in vehicle and road scenarios according to claim 1, characterized in that, The optimization algorithm used for parameter fitting is either gradient descent or the Adam optimizer.

6. The method for determining the collaborative threshold of traffic information in vehicle and road scenarios according to claim 1, characterized in that, Methods for determining the weights of different categories include the analytic hierarchy process (AHP), entropy weighting, best-worst method, or the Delphi expert method.

7. The method for determining the collaborative threshold of traffic information in vehicle and road scenarios according to claim 1, characterized in that, The categories of traffic information sources in the road scene include: meaning-based, motion-based, physical, and environmental; the categories of display elements in the in-vehicle interactive interface include: Chinese characters, English letters, Arabic numerals, characters, colors, and icons.

8. The method for determining the collaborative threshold of traffic information volume in vehicle and road scenarios according to claim 1, characterized in that, The information entropy model includes the Shannon entropy model, the Renyi entropy model, or the Tsallis entropy model.

Images