Intelligent traffic management system based on driver way-giving willingness analysis

Abstract

The invention discloses an intelligent traffic management system based on driver way-giving intention analysis, which belongs to the technical field of intelligent traffic management and comprises a data acquisition module, a data processing module, a model construction module, an analysis and verification module and a decision output module. Through the technical scheme of data fusion-model construction-dynamic response, the three problems of data one-sidedness, model limitation and intervention passivity in the prior art are directly solved, and accurate analysis and active control of the way-giving willingness of the driver at the non-signalized intersection are realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent traffic management, and in particular relates to an intelligent traffic management system based on analysis of drivers' willingness to give way. Background Art

[0002] Existing research has largely focused on macro-level analysis of pedestrian crossing safety and pedestrian-vehicle conflicts, using quantitative models to explore the factors influencing drivers' yielding behavior. For example, studies based on logit models have shown that vehicle speed, traffic density, and the number of pedestrians significantly influence drivers' yielding decisions. Bayesian multi-level logistic regression models have revealed the dynamic characteristics of pedestrian crossing behavior over time. Furthermore, improvements to road infrastructure design (such as trapezoidal speed humps and self-luminous crosswalks) have been shown to reduce vehicle speeds and cut-over through physical intervention. Some studies have also used logistic regression models to analyze the impact of individual driver characteristics (such as gender and personality type) on waiting time thresholds, providing data support for pedestrian safety crossing. These findings lay a theoretical foundation for understanding the mechanisms of human-vehicle interaction and optimizing traffic management strategies.

[0003] Despite some progress, research still faces significant limitations. First, existing analyses have largely focused on urban roads or signalized intersections. Research on the unique scenario of unsignalized intersections on rural roads is severely lacking. Unsignalized intersections on rural roads, characterized by mixed traffic, complex road conditions, and weak infrastructure, present unique driver behavior patterns and conflict mechanisms. Second, while existing models can identify the influence of external environmental variables (such as the number of lanes and pedestrian behavior), their exploration of drivers' subjective psychological factors (such as hazard perception, sensation seeking, and moral constraints) is extremely limited. In particular, there is a lack of systematic analysis incorporating psychological theories (such as the theory of planned behavior and deterrence theory). Furthermore, existing road infrastructure optimization schemes (such as speed humps) often rely on passive interventions and fail to form a closed-loop linkage with drivers' dynamic decision-making processes, limiting their effectiveness. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes an intelligent traffic management system based on the analysis of the driver's willingness to give way, so as to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above objectives, the present invention provides an intelligent traffic management system based on driver willingness analysis, comprising:

[0006] A data acquisition module is used to obtain raw data that affects the driver's willingness to yield through the vehicle terminal, mobile terminal, and roadside sensing equipment. The raw data includes: driver behavior data, pedestrian dynamic information, and environmental parameters;

[0007] A data processing module, used to clean and standardize the raw data to generate a structured data set;

[0008] A model building module, for building a structural equation model of the driver's willingness to yield based on latent variables and the structured data set;

[0009] An analysis and verification module is used to perform reliability testing, validity testing, and goodness of fit verification on the structural equation model, and output key influencing factors that affect the driver's willingness to yield;

[0010] The decision output module is used to generate dynamic warning instructions based on the key influencing factors and send them to the roadside interactive equipment at the unsignalized intersection.

[0011] Preferably, the data acquisition module includes:

[0012] The vehicle's OBD interface is used to collect real-time driver behavior data, including vehicle speed, braking frequency, and steering angle data;

[0013] Roadside infrared cameras are used to monitor pedestrian dynamic information, including: crossing density, distracted behavior, and number of pedestrians;

[0014] Environmental sensors are used to collect environmental parameters, including road grade, weather conditions, and traffic density.

[0015] Preferably, the data processing module includes:

[0016] Normalization processing unit, used to normalize the original data and eliminate dimensional differences;

[0017] An abnormal data elimination unit is used to detect and eliminate abnormal driving behavior data through the isolation forest algorithm to obtain cleaned data;

[0018] The data mapping unit is used to map the cleaned data to preset latent variable dimensions, where the latent variable dimensions include pedestrians' willingness to cross the street, external environment perception, behavioral intention, subjective norms, travel experience, punishment mechanism, perceived behavioral control, and behavioral attitude.

[0019] Preferably, the model building module includes:

[0020] The measurement model unit is used to define the correspondence between observed variables and latent variables;

[0021] The structural model unit is used to define the path hypothesis between latent variables and calculate the path coefficients through maximum likelihood estimation.

[0022] Preferably, the analysis and verification module:

[0023] Reliability verification unit, used to calculate Cronbach's α coefficient and combined reliability, and complete the reliability test;

[0024] Confirmatory factor analysis unit, used to obtain discriminant validity and convergent validity through confirmatory factor analysis;

[0025] The goodness of fit analysis unit is used to compare the GFI, RMSEA and CFI indicators with the preset thresholds to verify the model adaptability.

[0026] Preferably, the decision output module includes:

[0027] The first decision-making unit is used to trigger the roadside LED warning screen to display dynamic text prompts when pedestrian distraction behavior is detected;

[0028] The second decision-making unit is used to control the flashing frequency of the LED floor tiles of the self-luminous crosswalk according to the density of pedestrians crossing the street;

[0029] The third decision-making unit is used to push voice warning information to the driver's vehicle terminal through V2X communication.

[0030] Preferably, the system further comprises:

[0031] Edge computing nodes, deployed near the intersection of the village road, are used to process sensor data in real time and run lightweight structural equation models;

[0032] Cloud management platform for storing historical data, optimizing model parameters, and generating security assessment reports.

[0033] Preferably, the edge computing node includes:

[0034] The first optimization unit is used to adjust the weight of the latent variable according to the real-time pedestrian crossing behavior data;

[0035] The second optimization unit is used to update the path coefficient of the penalty mechanism through an incremental learning algorithm.

[0036] Preferably, the system further comprises:

[0037] A self-luminous crosswalk device whose luminous intensity is dynamically adjusted based on the predicted results of the driver's willingness to yield;

[0038] The trapezoidal speed hump is linked to the decision output module and increases the height of the hump to force a speed reduction when a low willingness to yield is detected.

[0039] Preferably, the closed-loop control of the system includes:

[0040] The coefficient correction unit is used to correct the path coefficient of the structural equation model according to the actual execution rate of the driver's yielding behavior;

[0041] Dynamic optimization unit, used to optimize dynamic warning strategies through reinforcement learning algorithms.

[0042] Compared with the prior art, the present invention has the following advantages and technical effects:

[0043] The present invention provides an intelligent traffic management system based on analysis of driver's willingness to give way, comprising: a data acquisition module for acquiring original data that affects the driver's willingness to give way through an on-board terminal, a mobile terminal and a roadside sensing device, wherein the original data includes driver behavior data, pedestrian dynamic information and environmental parameters; a data processing module for cleaning and standardizing the original data to generate a structured data set; a model construction module for constructing a structural equation model of the driver's willingness to give way based on latent variables and in combination with the structured data set; an analysis and verification module for performing reliability testing, validity testing and fit verification on the structural equation model, and outputting key influencing factors that affect the driver's willingness to give way; and a decision output module for generating dynamic warning instructions based on the key influencing factors and sending them to the roadside interactive equipment at the unsignalized intersection.

[0044] To address the technical problem that "the existing technology lacks the comprehensive collection of drivers' subjective psychological factors and external environmental variables, resulting in a single analysis dimension", the present invention integrates on-board terminals, roadside sensing devices and mobile terminals through a data acquisition module. The system realizes the simultaneous collection of drivers' psychological characteristics (such as attitudes and norms) and environmental parameters (such as pedestrian distraction and road grade) for the first time, providing a data foundation for multi-dimensional analysis.

[0045] To address the technical problem that "existing models ignore drivers' psychological mechanisms (such as subjective norms and perceived behavioral control), resulting in insufficient explanatory power for their willingness to give way," the model construction module in this invention defines latent variables (such as behavioral attitudes and punishment mechanisms) and their path relationships, and for the first time introduces a psychological framework into the analysis of unsignaled intersections on through-village roads, significantly improving the model's ability to depict drivers' decision-making logic.

[0046] To address the technical problem that "traditional facilities (such as speed humps) rely on passive intervention and cannot adapt to dynamic traffic scenarios", the decision-making output module in the present invention generates warning instructions in real time based on the model analysis results, actively intervenes in high-risk scenarios directly, and reduces the probability of human-vehicle conflicts.

[0047] The present invention directly solves the three major problems of the existing technology, namely, data one-sidedness, model limitations, and passive intervention, through the technical solution of data fusion-model construction-dynamic response, and realizes the accurate analysis and active control of the willingness of drivers to give way at unsignalized intersections. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings:

[0049] Figure 1 This is a structural equation model diagram based on Amos in an embodiment of the present invention;

[0050] Figure 2 Schematic diagram of a system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0051] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0052] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0053] Example 1

[0054] like Figure 1-2 As shown, this embodiment provides an intelligent traffic management system based on driver yielding willingness analysis, which includes five modules: data acquisition module, data processing module, model building module, analysis and verification module, and decision output module. Specifically, it includes:

[0055] A data acquisition module is used to obtain raw data that affects the driver's willingness to yield through the vehicle terminal, mobile terminal, and roadside sensing equipment. The raw data includes: driver behavior data, pedestrian dynamic information, and environmental parameters;

[0056] Furthermore, the data acquisition module includes:

[0057] The vehicle's OBD interface is used to collect real-time driver behavior data, including vehicle speed, braking frequency, and steering angle data;

[0058] Roadside infrared cameras are used to monitor pedestrian dynamic information, including: crossing density, distracted behavior, and number of pedestrians;

[0059] Environmental sensors are used to collect environmental parameters, including road grade, weather conditions, and traffic density.

[0060] A data processing module, used to clean and standardize the raw data to generate a structured data set;

[0061] Furthermore, the data processing module includes:

[0062] Normalization processing unit, used to normalize the original data and eliminate dimensional differences;

[0063] An abnormal data elimination unit is used to detect and eliminate abnormal driving behavior data through the isolation forest algorithm to obtain cleaned data;

[0064] The data mapping unit is used to map the cleaned data to preset latent variable dimensions, where the latent variable dimensions include pedestrians' willingness to cross the street, external environment perception, behavioral intention, subjective norms, travel experience, punishment mechanism, perceived behavioral control, and behavioral attitude.

[0065] Specifically, the eight latent variables used in the structural equation model are behavioral intention (BI), pedestrian willingness to cross the street (WTC), subjective norm (SN), external environment perception (SP), travel experience or habit intensity (TE), penalty mechanism (PM), perceived behavioral control (PBC), and behavioral attitude (AB). These latent variables and their corresponding observed variables are listed in Table 1 as a structured data set.

[0066] Table 1

[0067]

[0068] In Table 1, subjective norm refers specifically to the expectations and pressures drivers perceive from their surroundings (such as family and friends), traffic rules, and social ethics when deciding whether to yield to pedestrians. Perceived behavioral control, on the other hand, reflects the driver's subjective assessment of the difficulty of yielding, an assessment that incorporates personal driving skills and experience as well as external traffic conditions (such as road conditions, traffic signals, and pedestrian movements).

[0069] A model building module, for building a structural equation model of the driver's willingness to yield based on latent variables and the structured data set;

[0070] Furthermore, the model building module includes:

[0071] The measurement model unit is used to define the correspondence between observed variables and latent variables;

[0072] The structural model unit is used to define the path hypothesis between latent variables and calculate the path coefficients through maximum likelihood estimation.

[0073] Specifically, this embodiment proposes the path hypothesis between latent variables:

[0074] H1: SN→AB subjective norms positively influence behavioral attitudes;

[0075] H2: TE→AB travel experience or habit intensity positively influences behavioral attitude;

[0076] H3: AB→BI behavioral attitude positively affects behavioral intention;

[0077] H4: PBC→AB perceived behavioral control positively affects behavioral attitude;

[0078] H5: PM→TE penalty mechanism positively affects travel experience and habits;

[0079] H6: PM→SN punishment mechanism positively affects subjective norms;

[0080] H7: PM→BI penalty mechanism positively affects behavioral intention;

[0081] H8: SP→SN external environment perception positively affects subjective norms;

[0082] H9: SP→PBC external environment perception positively affects perceived behavioral control;

[0083] H10: SP→BI external environment perception positively affects behavioral intention;

[0084] H11: WTC→AB pedestrians' willingness to cross the street positively affects drivers' attitudes toward yielding;

[0085] H12: WTC→SN pedestrians’ willingness to cross the street positively influences subjective norms.

[0086] Structural equation models, such as Figure 1 As shown in the figure, pedestrian willingness to cross the street (WTC), penalty mechanism (PM), and external environment perception (SP) are clearly defined as exogenous latent variables. They serve as the starting point / independent variables of the model and have an initial impact on the internal dynamics of the system. The remaining latent variables, including behavioral intention (BI), subjective norm (SN), travel experience or habit intensity (TE), perceived behavioral control (PBC), and behavioral attitude (AB), are considered endogenous latent variables. They are effect / dependent variables, and their status is influenced and constrained by other variables in the model.

[0087] An analysis and verification module is used to perform reliability testing, validity testing, and goodness of fit verification on the structural equation model, and output key influencing factors that affect the driver's willingness to yield;

[0088] Furthermore, the analysis and verification module:

[0089] Reliability verification unit, used to calculate Cronbach's α coefficient and combined reliability, and complete the reliability test;

[0090] Specifically, to ensure the reliability and stability of the structured dataset, rigorous reliability testing was conducted. The data presented in Table 2 show that the Cronbach's α coefficient and composite reliability (CR) for each dimension exceeded the threshold of 0.7. Furthermore, the average variance extracted (AVE) was also above 0.5, meeting recognized standards in the academic community. This result strongly demonstrates the good reliability of the structured dataset across all dimensions, laying a solid foundation for subsequent factor analysis.

[0091] Table 2

[0092]

[0093] Confirmatory factor analysis unit, used to obtain discriminant validity and convergent validity through confirmatory factor analysis;

[0094] Specifically, before conducting factor analysis, we conducted a comprehensive test of the dataset's suitability. Using the Bartlett test of sphericity and the Kaiser-Meyer-Olkin (KMO) measure, the results presented in Table 3 clearly demonstrate that the dataset exhibits a high degree of suitability for factor analysis (specifically, the KMO value exceeds the benchmark of 0.6, reaching the "excellent" level; concurrently, the Bartlett test's p-value is less than 0.05, a statistically significant result). This finding not only reveals the possibility of shared factors among the variables but also strongly demonstrates the adequacy of the sample size to effectively conduct the subsequent factor analysis.

[0095] Table 3

[0096]

[0097] Factor analysis utilizes principal component analysis (PCA), a technique designed to deeply dissect the underlying structure behind questionnaire items, thereby simplifying the data and improving analytical efficiency. PCA minimizes data variance by reducing its dimensionality. It strictly adheres to the Kaiser criterion, retaining only factors with eigenvalues exceeding 1 to ensure analytical validity. To further enhance the interpretability of the results, the Varimax rotation method was used to optimize the factor loadings. The factor loading matrix clearly demonstrates the distribution of variables across different factors, providing a strong basis for factor naming. Ultimately, the results of factor analysis not only clarify the measurement dimensions but also lay a solid foundation for subsequent hypothesis testing and in-depth interpretation of the phenomenon.

[0098] Table 4

[0099]

[0100] The variance extraction ratio (VER) metric quantifies the extent to which the extracted factors explain the variance of the original dataset variables, thereby assessing the representativeness of the factors. As shown in Table 4, the variance extraction ratio obtained in this example analysis exceeded the 50% threshold, strongly demonstrating the effectiveness of the factor extraction process and its good representativeness of the dataset.

[0101] Table 5

[0102]

[0103] An overview of the rotated component matrix is shown in Table 5. After varimax rotation, the relationships between the factors and the research items are clearly defined. Notably, the commonalities for all research items exceed the 0.4 threshold, strongly demonstrating the close associations between the factors and the research items and validating the effectiveness of the information extraction process.

[0104] Confirmatory factor analysis (CFA) plays a crucial role in the construction and validation of SEM. Its core purpose is to rigorously examine whether the correspondence between observed indicators and latent variables aligns with the pre-defined theoretical framework, thereby ensuring the accuracy and validity of the measurement instrument. CFA encompasses two crucial dimensions of validity assessment: discriminant validity and convergent validity. Discriminant validity aims to verify that different latent variables maintain their conceptual independence and distinguishability, while convergent validity focuses on ensuring that a set of observed variables consistently and reliably reflects the core characteristics of their corresponding latent variables. Using CFA as an analytical tool, this study conducted a comprehensive and systematic assessment of the measurement model's fit, not only thoroughly examining the fit between the model and the data but also laying a solid and reliable foundation for subsequent structural model analysis. This process not only enhanced the reliability and validity of the research findings but also further strengthened the explanatory and predictive power of the theoretical model in practical applications.

[0105] Table 6

[0106]

[0107] The discriminant validity analysis results, shown in Table 6, show that the square root of the AVE for each dimension in the questionnaire, that is, the values on the diagonal, are consistently higher than the other off-diagonal values in their respective columns. This result strongly demonstrates the questionnaire's excellent discriminant validity, meaning that the different dimensions can be clearly distinguished from each other and independently measure different constructs, thus ensuring the validity and accuracy of the measurement tool.

[0108] Table 7

[0109]

[0110] Convergent validity analysis is a crucial step in assessing the internal consistency and intercorrelations of items within each dimension of a questionnaire. High correlations between items indicate a high degree of agreement in measuring the same underlying concept, thereby enhancing the reliability and robustness of the study. Conversely, low correlations require a careful review of the rationality of the item design and the measurement methodology employed to identify and correct potential biases or misinterpretations. In specific assessments, composite reliability (CR), a key indicator of the consistency and stability of a measurement instrument, should generally be no less than 0.6 to ensure measurement reliability. AVE, on the other hand, is used to assess internal consistency and detect common method variance. When AVE values reach or exceed 0.5, it indicates good internal consistency within the measurement dimension, thereby confirming the validity of the measurement instrument. The results of the convergent validity analysis, as shown in Table 7, show that the CR and AVE values for all dimensions in this study met the established standard thresholds. This result strongly demonstrates the excellent convergent validity of the questionnaire, indicating that the collected data are not only accurate and reliable, but also provide solid support for the study's conclusions.

[0111] The goodness of fit analysis unit is used to compare the GFI, RMSEA and CFI indicators with the preset thresholds to verify the model adaptability.

[0112] Specifically, we used Amos 26.0 software to conduct an in-depth analysis of the constructed structural equation model. The model fit analysis results, shown in Table 8, comprehensively consider the obtained results and observe that all model fit indices fell within acceptable ranges. This finding strongly demonstrates the rationality and effectiveness of the model construction. Based on this, the research was able to smoothly advance to the subsequent path analysis phase.

[0113] Table 8

[0114]

[0115] This example comprehensively tests the 12 pre-defined hypotheses. The path coefficients are shown in Table 9, which details the relationship framework of these hypotheses, the corresponding standardized regression coefficients β, key statistical indicators such as p-values, and a clear judgment based on these indicators on whether each hypothesis is supported by the data.

[0116] Table 9

[0117]

[0118] The results of the unstandardized coefficients showed that all path coefficients were greater than 0, there was a positive influence between the variables, and the P value was less than 0.05, indicating that the influence was significant.

[0119] All latent variables influence drivers' willingness to yield to pedestrians at unsignalized intersections. Among these influencing factors, pedestrian distraction, the presence of a penalty mechanism, and drivers' behavioral attitudes were identified as the most critical. A strong correlation was observed between travel habits and behavioral attitudes, indicating that drivers' daily travel patterns are closely linked to their underlying behavioral attitudes. Similarly, a significant correlation was found between perception of the external environment and perceived behavioral control, revealing how environmental factors influence drivers' decision-making processes. Furthermore, the link between pedestrians' willingness to cross the street and drivers' behavioral attitudes is also not negligible, further highlighting the complexity of the behavioral interaction between the two parties.

[0120] The finding that subjective norms have a significant positive impact on behavioral attitudes strongly supports Hypothesis H1. Specifically, it reveals the important role of external pressure factors, such as moral norms, in shaping drivers' willingness to yield on transvillage roads. This suggests that when drivers perceive strong social expectations and norms, they are more likely to exhibit proactive yielding behavior. Therefore, to further improve traffic safety on transvillage roads, strengthening relevant publicity and education efforts is particularly important. By enhancing drivers' sense of social responsibility, they can be more conscious of complying with traffic rules, thereby fostering a more harmonious and safe traffic environment on transvillage roads.

[0121] The finding that travel experience or habit strength has a significant positive impact on behavioral attitudes strongly supports Hypothesis H2. Specifically, it reveals that drivers with good driving habits are more likely to demonstrate a proactive and responsible attitude toward yielding to pedestrians crossing the road. Therefore, strengthening driver training and education to improve driving skills and foster safer driving habits is a key strategy for enhancing traffic safety on inter-village roads. This includes teaching drivers how to strictly control their speed and avoid speeding when encountering pedestrians crossing the road, as well as enhancing their ability to observe dynamics at intersections and zebra crossings, and proactively slowing down to avoid them.

[0122] The finding that behavioral attitudes positively influence behavioral intentions provides strong support for Hypothesis H3. This conclusion is highly consistent with the core concepts of the Theory of Planned Behavior and further validates the theory's applicability in real-world situations. Specifically, it reveals that drivers' positive attitudes toward yielding significantly increase their willingness to yield during actual driving.

[0123] The finding that perceived behavioral control has a significant positive impact on behavioral attitudes provides strong support for Hypothesis H4. Specifically, when drivers' self-control is weak—for example, when they believe it's easy to cut in but difficult to avoid cutting in—their willingness to yield is significantly reduced. This phenomenon is highly consistent with Bandura's self-efficacy theory, which emphasizes the key role of an individual's belief in their own abilities in behavioral choices. Insufficient perceived behavioral control reflects drivers' low self-restraint, which further hinders their ability to form a positive attitude toward yielding. Therefore, to enhance drivers' willingness to yield, systematic education and training are necessary to strengthen their safe driving cognition and self-control. Such training should focus on improving drivers' self-discipline in the face of various temptations, enabling them to more firmly adhere to traffic rules.

[0124] As an important external constraint, the penalty mechanism plays a significant role in regulating drivers' yielding behavior. Analysis results show that the penalty mechanism has a significant positive impact on travel habits, subjective norms, and behavioral intentions. This finding provides strong support for Hypotheses H5, H6, and H7. Specifically, positive incentives can motivate drivers to repeatedly yield, while negative penalties can effectively deter violations.

[0125] Perception of the external environment significantly influences subjective norms, perceived behavioral control, and behavioral intentions. Specifically, when drivers' perception of the external environment decreases, their willingness to yield also decreases. This finding strongly supports Hypotheses H8, H9, and H10. Given the key role of environmental perception in shaping drivers' yielding behavior, improving the traffic environment is an important approach to enhancing drivers' willingness to yield and improving traffic safety. Specifically, this can be achieved through measures such as adding traffic signs and optimizing road infrastructure. For example, roadside guardrails can be installed in areas with heavy traffic to increase driver alertness and protect pedestrians. In areas with less traffic, natural barriers can be used for protection, along with dedicated crosswalks and clear stop and yield signs.

[0126] A decrease in pedestrians' willingness to cross the street significantly weakens drivers' willingness to yield, which in turn has a significant impact on their behavioral attitudes and subjective norms. This finding provides strong support for Hypotheses H11 and H12. Pedestrian distraction is a key factor that cannot be ignored, often stemming from a lack of awareness of the traffic environment or an overconfidence in their own safety. According to attention allocation theory, when pedestrians are distracted, their reaction time to key information such as traffic signals and approaching vehicles is significantly reduced, undoubtedly increasing the risk of traffic accidents. Therefore, to improve overall road traffic safety, it is crucial to prioritize the cultivation of pedestrian traffic safety awareness and the improvement of their concentration. Specifically, through enhanced publicity and education, pedestrians should be fully aware of the dangers of distracted crossing and be encouraged to develop good habits of focused crossing. Furthermore, regulations should be implemented to ensure that pedestrians strictly adhere to traffic rules when crossing the road, thereby reducing the probability of traffic accidents.

[0127] The decision output module is used to generate dynamic warning instructions based on the key influencing factors and send them to the roadside interactive equipment at the unsignalized intersection.

[0128] Furthermore, the decision output module includes:

[0129] The first decision-making unit is used to trigger the roadside LED warning screen to display dynamic text prompts when pedestrian distraction behavior is detected;

[0130] The second decision-making unit is used to control the flashing frequency of the LED floor tiles of the self-luminous crosswalk according to the density of pedestrians crossing the street;

[0131] The third decision-making unit is used to push voice warning information to the driver's vehicle terminal through V2X communication.

[0132] In this embodiment, the system also includes edge computing nodes and a cloud management platform;

[0133] Edge computing nodes, deployed near the intersection of the village road, are used to process sensor data in real time and run lightweight structural equation models;

[0134] Furthermore, the edge computing node includes:

[0135] The first optimization unit is used to adjust the weight of the latent variable according to the real-time pedestrian crossing behavior data;

[0136] The second optimization unit is used to update the path coefficient of the penalty mechanism through an incremental learning algorithm.

[0137] Cloud management platform for storing historical data, optimizing model parameters, and generating security assessment reports.

[0138] In this embodiment, the system further includes a self-luminous crosswalk device and a trapezoidal speed hump;

[0139] A self-luminous crosswalk device whose luminous intensity is dynamically adjusted based on the predicted results of the driver's willingness to yield;

[0140] The trapezoidal speed hump is linked to the decision output module and increases the height of the hump to force a speed reduction when a low willingness to yield is detected.

[0141] The closed-loop control of the system includes:

[0142] The coefficient correction unit is used to correct the path coefficient of the structural equation model according to the actual execution rate of the driver's yielding behavior;

[0143] Dynamic optimization unit, used to optimize dynamic warning strategies through reinforcement learning algorithms.

[0144] Beneficial effects of this embodiment:

[0145] This example constructs a structural equation model of drivers' willingness to yield at unsignalized intersections on rural roads. Through reliability and validity testing, confirmatory factor analysis, model fit assessment, and path analysis, the relationships among the model's variables and its overall validity were systematically verified. Furthermore, subjective norms, travel habits, behavioral attitudes, perceived behavioral control, and penalty mechanisms all significantly influence willingness to yield. Furthermore, environmental perception and pedestrian willingness to cross the street are also important factors, particularly pedestrian distraction, penalty mechanisms, and behavioral attitudes.

[0146] These improvement measures in this embodiment comprehensively consider multiple factors, including drivers' behavioral attitudes, subjective norms, perceived behavioral control, and environmental perception. By optimizing the traffic environment, strengthening publicity and education, and enhancing pedestrian traffic safety awareness and focus, these measures aim to comprehensively promote drivers' yielding behavior and create a safer and more orderly road traffic environment. This embodiment is expected to provide valuable reference and guidance for improving traffic safety on village roads.

[0147] The above are merely preferred embodiments of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. An intelligent traffic management system based on analysis of drivers' willingness to yield, characterized by: include: A data acquisition module is used to obtain raw data that affects the driver's willingness to yield through the vehicle terminal, mobile terminal, and roadside sensing equipment. The raw data includes: driver behavior data, pedestrian dynamic information, and environmental parameters; A data processing module, configured to clean and standardize the raw data to generate a structured data set; A model building module, for building a structural equation model of the driver's willingness to yield based on latent variables and the structured data set; An analysis and verification module is used to perform reliability testing, validity testing, and goodness of fit verification on the structural equation model, and output key influencing factors that affect the driver's willingness to yield; The decision output module is used to generate dynamic warning instructions based on the key influencing factors and send them to the roadside interactive equipment at the unsignalized intersection.

2. The system according to claim 1, wherein: The data acquisition module includes: The vehicle's OBD interface is used to collect real-time driver behavior data, including vehicle speed, braking frequency, and steering angle data; Roadside infrared cameras are used to monitor pedestrian dynamic information, including: crossing density, distracted behavior, and number of pedestrians; Environmental sensors are used to collect environmental parameters, including road grade, weather conditions, and traffic density.

3. The system according to claim 1, wherein: The data processing module includes: Normalization processing unit, used to normalize the original data and eliminate dimensional differences; An abnormal data elimination unit is used to detect and eliminate abnormal driving behavior data through the isolation forest algorithm to obtain cleaned data; The data mapping unit is used to map the cleaned data to preset latent variable dimensions, where the latent variable dimensions include pedestrians' willingness to cross the street, external environment perception, behavioral intention, subjective norms, travel experience, punishment mechanism, perceived behavioral control, and behavioral attitude.

4. The system according to claim 1, wherein: The model building module includes: The measurement model unit is used to define the correspondence between observed variables and latent variables; The structural model unit is used to define the path hypothesis between latent variables and calculate the path coefficients through maximum likelihood estimation.

5. The system according to claim 1, wherein: The analysis and verification module: Reliability verification unit, used to calculate Cronbach's α coefficient and combined reliability, and complete the reliability test; Confirmatory factor analysis unit, used to obtain discriminant validity and convergent validity through confirmatory factor analysis; The goodness of fit analysis unit is used to compare the GFI, RMSEA and CFI indicators with the preset thresholds to verify the model adaptability.

6. The system according to claim 1, wherein: The decision output module includes: The first decision-making unit is used to trigger the roadside LED warning screen to display dynamic text prompts when pedestrian distraction behavior is detected; The second decision-making unit is used to control the flashing frequency of the LED floor tiles of the self-luminous crosswalk according to the density of pedestrians crossing the street; The third decision-making unit is used to push voice warning information to the driver's vehicle terminal through V2X communication.

7. The system according to claim 1, wherein: Also includes: Edge computing nodes, deployed near the intersection of the village road, are used to process sensor data in real time and run lightweight structural equation models; Cloud management platform for storing historical data, optimizing model parameters, and generating security assessment reports.

8. The system according to claim 7, characterized in that The edge computing node includes: The first optimization unit is used to adjust the weight of the latent variable according to the real-time pedestrian crossing behavior data; The second optimization unit is used to update the path coefficient of the penalty mechanism through an incremental learning algorithm.

9. The system according to claim 1, wherein: Also includes: A self-luminous crosswalk device whose luminous intensity is dynamically adjusted based on the predicted results of the driver's willingness to yield; The trapezoidal speed hump is linked to the decision output module and increases the height of the hump to force a speed reduction when a low willingness to yield is detected.

10. The system according to claim 1, wherein: The closed-loop control of the system includes: The coefficient correction unit is used to correct the path coefficient of the structural equation model according to the actual execution rate of the driver's yielding behavior; Dynamic optimization unit, used to optimize dynamic warning strategies through reinforcement learning algorithms.

Images