A snow tourism traffic dynamic path optimization method based on multi-source data fusion

Abstract

This invention relates to the field of route optimization technology and discloses a dynamic route optimization method for ice and snow tourism transportation based on multi-source data fusion. This method addresses the problem of intermittent congestion during dynamic route planning. The method includes: constructing a candidate route evaluation dataset based on real-time traffic status data corresponding to each road segment; identifying the intermittent congestion status of each road segment based on a standardized traffic dataset and real-time traffic status data corresponding to each road segment, and correcting the road segment traffic evaluation data to obtain a corrected candidate route evaluation dataset; and performing route optimization analysis on each candidate route between a preset starting point and a target endpoint based on the corrected candidate route evaluation dataset to determine the target dynamic optimal route. This effectively improves the accuracy, stability, and practical reliability of the dynamic route optimization process for ice and snow tourism transportation.

Description

Technical Field

[0001] This invention relates to the field of route optimization technology, and more specifically to a method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion. Background Technology

[0002] With the continuous development of smart tourism systems, intelligent traffic management technologies, and multi-source traffic data collection technologies, traffic operations in ice and snow tourism scenarios are increasingly characterized by complex road conditions, frequent environmental changes, and dynamic fluctuations in traffic status. Roads surrounding scenic areas, dedicated tourist roads, and scenic access roads continuously generate a large amount of road traffic data, vehicle trajectory data, environmental monitoring data, and visitor flow data. This data is typically acquired by different collection devices or business systems and then uniformly analyzed and processed through traffic management platforms, navigation service platforms, or route planning systems. Traffic management systems usually need to analyze the traffic status of various road segments within the target area based on the aforementioned multi-source traffic data and generate corresponding route planning results and dynamic traffic guidance information accordingly. Therefore, how to achieve accurate traffic status analysis and improve the rationality of route optimization results under the condition of coexisting multi-source traffic data has become an important technological direction of continuous focus in the field of smart transportation for ice and snow tourism.

[0003] In existing technologies, dynamic route optimization systems based on multi-source traffic data fusion have been widely applied. These systems typically acquire relevant traffic data from road monitoring equipment, vehicle terminals, environmental monitoring equipment, and scenic area visitor flow monitoring equipment. They then comprehensively analyze the current traffic conditions of various road segments within a target area and generate corresponding route planning schemes based on the analysis results. By leveraging multi-source data fusion and dynamic route calculation technologies, these systems can adjust the route planning results according to real-time traffic condition changes, thereby improving the real-time performance and intelligence of traffic route recommendations.

[0004] In existing traffic route optimization mechanisms, the traffic status of each road segment is typically quantitatively evaluated based on average road speed, overall congestion level, or fixed-period statistical results, and these evaluation results are used as the basis for subsequent route planning analysis. During route optimization, the system generally uses the current traffic evaluation results of each road segment as a judgment benchmark to comprehensively calculate different candidate routes, thereby determining the target traffic route. This approach is widely used in practice, and its basic idea is to maintain consistency in route optimization logic through a unified road traffic evaluation mechanism, facilitating unified quantitative analysis of the traffic status of different road segments.

[0005] However, the above-mentioned technologies have at least the following technical problems:

[0006] In the actual operation of transportation scenarios in ice and snow tourism, some roads around scenic spots, sightseeing roads, or special tourist roads often experience intermittent congestion due to tourists' temporary stops, waiting for photos, short-term parking, or passing oncoming traffic. Although these roads may appear to have normal traffic flow in terms of overall average speed or macro-level congestion, their actual traffic capacity is already reduced or localized congestion occurs frequently. In this situation, if the corresponding road segments are uniformly evaluated based solely on average road speed, overall congestion level, or fixed-period statistical results, the aforementioned intermittent congestion may not be effectively identified. Road segments that appear to be clear but actually have intermittent congestion may still be judged as preferred routes, causing a discrepancy between the route optimization results and the actual road capacity. This, in turn, affects the accuracy, stability, and reliability of the dynamic route optimization process for ice and snow tourism transportation. Summary of the Invention

[0007] To overcome the aforementioned deficiencies in the prior art, this invention provides a dynamic route optimization method for ice and snow tourism transportation based on multi-source data fusion, in order to solve the problems existing in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A method for dynamic route optimization of ice and snow tourism traffic based on multi-source data fusion includes the following steps: Step 1: Obtain multi-source traffic data within the target ice and snow tourism area and construct an original traffic dataset; the multi-source traffic data includes vehicle speed data collected by road monitoring equipment, vehicle trajectory data uploaded by vehicle terminals, environmental monitoring data collected by meteorological monitoring equipment, and passenger flow distribution data collected by scenic area passenger flow monitoring equipment; Step 2: Preprocess the original traffic dataset to obtain a standardized traffic dataset; Step 3: Perform multi-source fusion analysis on the standardized traffic dataset, and input the fusion analysis results into a preset traffic state analysis model for state identification to obtain real-time traffic state data corresponding to each road segment. The real-time traffic state data includes the average speed of the road segment, traffic density of the road segment, congestion level of the road segment, snow / ice accumulation status of the road segment, weather impact level of the road segment, and traffic risk level of the road segment; Step 4: Based on the real-time traffic state data corresponding to each road segment, calculate the corresponding traffic speed of each road segment. The process involves: 1) Obtaining road segment traffic evaluation data and constructing a candidate path evaluation dataset; 2) Obtaining road segment traffic evaluation data including road segment travel time data, road segment congestion level data, road segment safety risk data, and road segment environmental impact data; 3) Obtaining a revised candidate path evaluation dataset based on the standardized traffic dataset and real-time traffic status data corresponding to each road segment; 4) Obtaining a revised candidate path evaluation dataset based on the revised candidate path evaluation dataset; 5) Obtaining a revised candidate path evaluation dataset based on the standardized traffic dataset and real-time traffic status data corresponding to each road segment, identifying intermittent congestion states for each road segment, and correcting the road segment traffic evaluation data when intermittent congestion states are identified, thus obtaining the revised target dynamic optimal path; 6) Obtaining a revised candidate path evaluation dataset based on the revised candidate path evaluation dataset, performing path optimization analysis on each candidate path between the preset starting point and the target endpoint, and determining the target dynamic optimal path; 7) Obtaining a revised candidate path evaluation dataset based on newly added multi-source traffic data, updating the real-time traffic status data, intermittent congestion states, and the revised candidate path evaluation dataset, and re-executing the path optimization analysis based on the updated candidate path evaluation dataset to output the updated target dynamic optimal path.

[0010] Preferably, the step of identifying the intermittent congestion state corresponding to each road segment is as follows: acquiring vehicle traffic speed data corresponding to each road segment within a preset time window, constructing a short-period speed fluctuation sequence based on the vehicle traffic speed data; constructing a stop-and-go switching direction sequence based on the short-period speed fluctuation sequence, and calculating the stop-and-go switching frequency parameter based on the stop-and-go switching direction sequence; calculating the low-speed continuous discrete parameter and the fluctuation recovery anomaly parameter based on the short-period speed fluctuation sequence; normalizing the stop-and-go switching frequency parameter, the low-speed continuous discrete parameter, and the fluctuation recovery anomaly parameter, and calculating the intermittent congestion judgment index based on the normalized stop-and-go switching frequency parameter, the low-speed continuous discrete parameter, and the fluctuation recovery anomaly parameter; acquiring the average traffic speed and the congestion level of each road segment, and obtaining the surface smooth traffic condition judgment result based on the average traffic speed and the congestion level; and determining whether the road segment is in an intermittent congestion state based on the surface smooth traffic condition judgment result and the intermittent congestion judgment index.

[0011] Preferably, the step of constructing a short-period speed fluctuation sequence based on vehicle traffic speed data is as follows: arranging the vehicle traffic speed data corresponding to each road segment within a preset time window in chronological order to construct a corresponding vehicle traffic speed time series; calculating the speed difference between adjacent time points based on the vehicle traffic speed data corresponding to adjacent time points in the vehicle traffic speed time series to obtain a speed difference sequence; and taking the absolute value of each speed difference in the speed difference sequence to obtain the corresponding short-period speed fluctuation sequence.

[0012] Preferably, the step of constructing the stop-start switching direction sequence based on the short-period speed fluctuation sequence is as follows: obtaining a speed difference sequence, and comparing each speed difference in the speed difference sequence with zero to determine the direction of change corresponding to each speed difference; when the corresponding speed difference is greater than zero, the corresponding direction of change is determined to be speed increase; when the corresponding speed difference is less than zero, the corresponding direction of change is determined to be speed decrease; when the corresponding speed difference is equal to zero, the corresponding direction of change is determined to be speed maintenance; and arranging each direction of change in sequence according to the order of the speed differences in the speed difference sequence to construct the corresponding stop-start switching direction sequence.

[0013] Preferably, the step of obtaining the stop-and-go switching frequency parameter is as follows: Obtain the stop-and-go switching direction sequence corresponding to the target road segment, and numerically label each direction item in the stop-and-go switching direction sequence; wherein, if the corresponding direction item indicates an increase in vehicle speed, the direction label value corresponding to that direction item is recorded as 1; if the corresponding direction item indicates a decrease in vehicle speed, the direction label value corresponding to that direction item is recorded as -1; if the corresponding direction item indicates that vehicle speed remains constant, the direction label value corresponding to that direction item is recorded as 0; remove each direction item with a direction label value of 0 from the stop-and-go switching direction sequence to obtain the stop-and-go switching frequency parameter for the target road segment. The corresponding valid direction sequence is determined; the number of direction items in the valid direction sequence is counted to obtain the number of valid direction items; when the number of valid direction items is not greater than 1, the stop-go switching frequency parameter corresponding to the target road segment is set to 0; when the number of valid direction items is greater than 1, the two adjacent direction items in the valid direction sequence are traversed, and the direction reversal identifier value corresponding to each adjacent direction item is calculated; the direction reversal identifier values ​​corresponding to all adjacent direction items are summed to obtain the number of valid reverse switching corresponding to the target road segment; the number of valid reverse switching is divided by the number of valid direction items minus 1 to obtain the stop-go switching frequency parameter corresponding to the target road segment.

[0014] Preferably, the step of obtaining the low-speed continuous discrete parameters is as follows: obtaining the short-period speed fluctuation sequence corresponding to the target road segment and the vehicle traffic speed time series corresponding to the short-period speed fluctuation sequence; calculating the median value of each vehicle traffic speed data in the vehicle traffic speed time series to obtain the low-speed reference speed corresponding to the target road segment; and calculating the average value of each fluctuation value in the short-period speed fluctuation sequence to obtain the fluctuation reference amplitude corresponding to the target road segment; comparing each vehicle traffic speed data with the low-speed reference speed, and comparing each corresponding fluctuation value with the fluctuation reference amplitude; when a vehicle traffic speed data is not greater than the low-speed reference speed, and its corresponding fluctuation value is not less than the fluctuation reference amplitude, it is determined that the position is in a low-speed disturbance state, and the corresponding state identifier value is recorded as 1; otherwise, it is determined that the position is not in a low-speed disturbance state, and the corresponding state identifier value is recorded as 0, thereby constructing the low-speed disturbance state sequence corresponding to the target road segment; and processing the low-speed disturbance state... In the state sequence, segments with consecutive values ​​of 1 are identified, and each segment with consecutive values ​​of 1 is defined as a low-speed continuous segment. The number of consecutive state identifier values ​​of 1 in each low-speed continuous segment is counted to obtain the segment duration length corresponding to each low-speed continuous segment. The number of low-speed continuous segments corresponding to the target road segment is counted to obtain the segment count. The segment duration lengths corresponding to each low-speed continuous segment are summed to obtain the total segment duration length. When the segment count is greater than 0, the total segment duration length is divided by the segment count to obtain the average segment duration length. The absolute values ​​of the length differences between the segment duration lengths corresponding to each low-speed continuous segment and the average segment duration length are calculated to obtain the total length dispersion. When the segment count is greater than 1, the total length dispersion is divided by the product of the total segment duration length and the segment count minus 1 to obtain the low-speed continuous dispersion parameter corresponding to the target road segment. When the segment count is not greater than 1, the low-speed continuous dispersion parameter corresponding to the target road segment is set to 0.

[0015] Preferably, the step of obtaining the fluctuation recovery anomaly parameter is as follows: obtaining the short-period speed fluctuation sequence corresponding to the target road segment, sorting each fluctuation value in the short-period speed fluctuation sequence according to its numerical value, calculating the upper quartile value corresponding to the sorting result, and obtaining the high-level reference amplitude of the fluctuation corresponding to the target road segment; comparing each fluctuation value in the short-period speed fluctuation sequence with the high-level reference amplitude; when a certain fluctuation value is not less than the high-level reference amplitude, and the fluctuation value corresponding to its previous sampling position is less than the high-level reference amplitude, determining that sampling position as a high-level fluctuation trigger position; when a certain fluctuation value is less than the high-level reference amplitude, the sampling position is determined as a high-level fluctuation trigger position; when a certain fluctuation value is less than the high-level reference amplitude, the sampling position is determined as a high-level fluctuation trigger position. When the first sampling position is reached, and the fluctuation value is not less than the high-level reference amplitude, the sampling position is determined as the high-level fluctuation trigger position. For each high-level fluctuation trigger position, starting from the corresponding high-level fluctuation trigger position, the fallback completion position is searched one by one along the subsequent sampling direction of the short-period velocity fluctuation sequence. When the fluctuation value corresponding to a subsequent sampling position is less than the high-level reference amplitude, and the fluctuation values ​​corresponding to that sampling position and the next sampling position are both less than the high-level reference amplitude, the sampling position is determined as the fallback completion position of the corresponding high-level fluctuation trigger position. When the sampling position is the end position of the short-period velocity fluctuation sequence... If the fluctuation value corresponding to the sampling position is less than the high-level reference amplitude of the fluctuation, it is directly determined as the fallback completion position. If no corresponding fallback completion position is found within the preset time window for a certain high-level fluctuation trigger position, the sampling position corresponding to the end of the short-period velocity fluctuation sequence is determined as the fallback completion position corresponding to the high-level fluctuation trigger position. For each high-level fluctuation trigger position, the number of samples between the corresponding high-level fluctuation trigger position and the corresponding fallback completion position is counted to obtain the corresponding recovery length. The difference between the fluctuation value corresponding to the high-level fluctuation trigger position and the high-level reference amplitude of the fluctuation is obtained to obtain the corresponding... The overamplitude is calculated as follows: For each high-level fluctuation trigger position, when the corresponding overamplitude is greater than 0, the corresponding recovery length is divided by the corresponding overamplitude to obtain the corresponding recovery anomaly coefficient; when the corresponding overamplitude is not greater than 0, the recovery anomaly coefficient corresponding to the high-level fluctuation trigger position is not calculated; the number of recovery anomaly coefficients corresponding to the target road segment is counted to obtain the number of effective recovery samples; when the number of effective recovery samples is greater than 0, the median value of all recovery anomaly coefficients is calculated to obtain the fluctuation recovery anomaly parameter corresponding to the target road segment; when the number of effective recovery samples is 0, the fluctuation recovery anomaly parameter corresponding to the target road segment is determined to be 0.

[0016] Preferably, the step of obtaining the surface traffic flow condition judgment result based on the average traffic speed and congestion level of the road segment is as follows: Classify each road segment in the target ice and snow tourism area according to road type to obtain multiple sets of similar road segments; within each set of similar road segments, obtain the average traffic speed corresponding to each road segment, and calculate the median value of the average traffic speed corresponding to each road segment to obtain the regional average speed benchmark parameter for the corresponding set of similar road segments; within each set of similar road segments, obtain the congestion level corresponding to each road segment, and calculate the median value of the congestion level corresponding to each road segment to obtain the regional congestion level benchmark parameter for the corresponding set of similar road segments; The average traffic speed of the target road segment is compared with the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is compared with the regional congestion level benchmark parameter of the set of similar road segments. If the average traffic speed of the target road segment is not less than the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is not greater than the regional congestion level benchmark parameter of the set of similar road segments, the surface traffic flow condition of the target road segment is determined to be satisfied; otherwise, the surface traffic flow condition of the target road segment is determined to be unsatisfactory.

[0017] Preferably, the step of determining whether a road segment is in an intermittent congestion state based on the surface unobstructed condition judgment result and the intermittent congestion judgment index is as follows: obtain the surface unobstructed condition judgment result corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments to which it belongs; when the judgment result is satisfied and the intermittent congestion judgment index corresponding to the target road segment is greater than the intermittent congestion benchmark index corresponding to the set of similar road segments to which it belongs, the target road segment is determined to be in an intermittent congestion state; otherwise, the target road segment is determined not to be in an intermittent congestion state.

[0018] Preferably, the step of correcting the road segment traffic evaluation data corresponding to the road segment to obtain the corrected candidate path evaluation dataset includes: obtaining the intermittent congestion judgment index corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments to which the target road segment belongs; obtaining the road segment travel time data, road segment congestion level data, road segment safety risk data, and road segment environmental impact data corresponding to the target road segment; calculating the intermittent congestion correction coefficient corresponding to the target road segment based on the intermittent congestion judgment index and the intermittent congestion benchmark index corresponding to the set of similar road segments to which the target road segment belongs; correcting the road segment travel time data corresponding to the target road segment based on the intermittent congestion correction coefficient to obtain the corrected road segment travel time data; and correcting the road segment congestion level data corresponding to the target road segment based on the intermittent congestion correction coefficient. The process involves several steps: First, the congestion level data of the target road segment is corrected. Second, the safety risk data of the target road segment is corrected based on the intermittent congestion correction coefficient. Third, the environmental impact data of the target road segment is retained as the corrected environmental impact data. Finally, the corrected travel time data, congestion level data, safety risk data, and environmental impact data are collectively used to determine the corrected traffic evaluation data of the target road segment. Fourth, the original traffic evaluation data of the target road segment in the candidate path evaluation dataset is replaced with the corrected traffic evaluation data. The original traffic evaluation data of the remaining road segments not identified as intermittently congested in the candidate path evaluation dataset is retained, thus obtaining the corrected candidate path evaluation dataset.

[0019] The technical effects and advantages of this invention are as follows:

[0020] Based on real-time traffic status data for each road segment, a candidate path evaluation dataset is constructed. Using a standardized traffic dataset and real-time traffic status data for each road segment, intermittent congestion is identified, and the road segment traffic evaluation data is corrected to obtain a corrected candidate path evaluation dataset. Based on this corrected candidate path evaluation dataset, path optimization analysis is performed on each candidate path between the preset starting point and the target endpoint to determine the target dynamic optimal path, effectively improving the accuracy, stability, and practical reliability of the dynamic path optimization process for ice and snow tourism transportation. Attached Figure Description

[0021] Figure 1 A flowchart illustrating a dynamic route optimization method for ice and snow tourism transportation based on multi-source data fusion, provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The method for dynamic path optimization of ice and snow tourism transportation based on multi-source data fusion involved in the present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention provides a method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion, such as... Figure 1 As shown, it includes the following steps:

[0024] Step 1: Obtain multi-source traffic data within the target ice and snow tourism area and construct the original traffic dataset; the multi-source traffic data includes vehicle speed data collected by road monitoring equipment, vehicle trajectory data uploaded by vehicle terminals, environmental monitoring data collected by meteorological monitoring equipment, and visitor flow distribution data collected by scenic area visitor flow monitoring equipment.

[0025] Step 2: Preprocess the original traffic dataset to obtain a standardized traffic dataset; preprocessing includes outlier removal, missing data completion, time alignment, and data format standardization.

[0026] Step 3: Perform multi-source fusion analysis on the standardized traffic dataset, and input the fusion analysis results into the preset traffic state analysis model for state identification to obtain real-time traffic state data corresponding to each road segment. The real-time traffic state data includes, but is not limited to, the average traffic speed of the road segment, the traffic density of the road segment, the congestion level of the road segment, the snow / ice accumulation status of the road segment, the weather impact level of the road segment, and the traffic risk level of the road segment. The real-time traffic state data is used to characterize the traffic status, congestion status, and environmental impact status of the corresponding road segment.

[0027] It should be noted that when performing multi-source fusion analysis on standardized traffic datasets, existing multi-source heterogeneous data fusion methods can be used. Specifically, traffic data from different sources can be uniformly fused through methods such as time alignment, spatial mapping, and feature association to generate fusion analysis results that characterize the real-time state of roads. This embodiment does not limit this approach.

[0028] It should be noted that the traffic state analysis model can be constructed using existing traffic state recognition models. The traffic state recognition model can be a traffic state prediction model trained based on machine learning or deep learning, such as a random forest model, a long short-term memory network model, or a graph convolutional traffic prediction model. This embodiment does not limit its specific structure.

[0029] Step 4: Based on the real-time traffic status data corresponding to each road segment, calculate the road segment traffic evaluation data corresponding to each road segment, and construct a candidate path evaluation dataset; the road segment traffic evaluation data includes road segment travel time data, road segment congestion level data, road segment safety risk data, and road segment environmental impact data;

[0030] It should be noted that in this embodiment, the existing road segment evaluation calculation method can be used to quantitatively analyze the road segment travel time, congestion level, safety risks and environmental impact based on the real-time traffic status data corresponding to each road segment, so as to obtain the corresponding road segment traffic evaluation data. This embodiment does not limit this.

[0031] It should be noted that the road segment congestion level and road segment congestion degree data in this embodiment are parameters at different levels. The road segment congestion level is derived from real-time traffic status data and is used to characterize the congestion category or level information of the target road segment at the current moment, mainly for determining surface smooth traffic conditions and identifying traffic conditions. The road segment congestion degree data, on the other hand, is a quantitative evaluation result further calculated based on real-time traffic status data and is used as a component of path cost calculation during the path optimization analysis process.

[0032] Step 5: Based on the standardized traffic dataset and the real-time traffic status data corresponding to each road segment, identify the intermittent congestion status corresponding to each road segment. When an intermittent congestion status is identified, the road segment traffic evaluation data corresponding to the road segment is corrected to obtain the corrected candidate path evaluation dataset. The intermittent congestion status is used to characterize the short-cycle traffic disturbance state of alternating stops and starts on the road segment when the overall traffic status appears smooth.

[0033] In this embodiment, it should be specifically explained that the step of identifying the intermittent congestion status corresponding to each road segment is as follows:

[0034] Obtain vehicle speed data for each road segment within a preset time window, and construct a short-period speed fluctuation sequence based on the vehicle speed data;

[0035] It should be noted that the preset time window is used to limit the time range for statistical analysis of vehicle speed data for each road segment. The preset time window is a continuous time interval selected during the traffic condition analysis process. Within this time interval, multiple sets of vehicle speed data corresponding to each road segment are collected to reflect the overall traffic status and changes of the target road segment during this time period.

[0036] A stop-start switching direction sequence is constructed based on a short-period velocity fluctuation sequence, and the stop-start switching frequency parameter is calculated based on the stop-start switching direction sequence.

[0037] Low-speed continuous discrete parameters and fluctuation recovery anomaly parameters are calculated based on short-period velocity fluctuation sequences.

[0038] The stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter are normalized. Specifically, in this embodiment, the normalization of these parameters can be performed using existing vector normalization methods. Specifically, the stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter are combined to form a three-dimensional vector. The norm value is obtained by calculating the square root of the sum of the squares of each component of this three-dimensional vector. Each parameter is then divided by this norm value to achieve normalization under a unified dimension. This normalization method eliminates the unbalanced impact of differences in the numerical range and variation amplitude of the stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter on the subsequent calculation of the intermittent blockage judgment index. This ensures that each parameter has a consistent scale basis in the comprehensive judgment process, preventing any single parameter from having an excessively dominant effect on the judgment result due to a large value range, thereby improving the stability and accuracy of the intermittent blockage state judgment result. Since the vector normalization method is existing technology, this embodiment will not describe its specific calculation process in detail. The intermittent blocking judgment index is calculated based on the normalized stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter. The specific steps for obtaining the index are as follows:

[0039] ;

[0040] In the formula, This is represented as the intermittent blockage judgment index. This is expressed as the normalized stop-and-go frequency parameter. This is represented as the normalized low-speed continuous discrete parameter. This is represented as the normalized fluctuation recovery anomaly parameter. , , The weighting coefficients are represented as the normalized stop-and-go frequency parameter, the normalized low-speed continuous discrete parameter, and the normalized fluctuation recovery anomaly parameter. , , , The weight coefficients of each factor are obtained through the Analytic Hierarchy Process (AHP), an analytical method that decomposes complex decision problems into multiple hierarchical structures and calculates the weight coefficients of each factor by constructing a judgment matrix and comparing the relative importance of each evaluation factor pairwise. In practical implementation, a corresponding comparison relationship can be constructed based on the influence of each parameter in the intermittent blocking determination, and a reasonable weight allocation result can be obtained through consistency checks. Since the AHP is existing technology, its specific calculation process will not be described in detail in this embodiment.

[0041] Obtain the average traffic speed and congestion level of each road segment, and obtain the surface smooth traffic condition judgment result based on the average traffic speed and congestion level of each road segment;

[0042] Based on the surface unobstructed condition assessment results and the intermittent congestion assessment index, it is determined whether the road segment is in an intermittent congestion state.

[0043] In this embodiment, it should be specifically explained that the steps for constructing a short-period speed fluctuation sequence based on vehicle traffic speed data are as follows:

[0044] Arrange the vehicle speed data of each road segment in chronological order within a preset time window to construct the corresponding vehicle speed time series.

[0045] Based on the vehicle speed data corresponding to adjacent time points in the vehicle speed time series, the speed difference between adjacent time points is calculated to obtain the speed difference sequence.

[0046] The absolute value of each speed difference in the speed difference sequence is taken to obtain the corresponding short-period speed fluctuation sequence, which is used to characterize the short-period change characteristics of vehicle traffic speed within a preset time window.

[0047] In this embodiment, it should be specifically explained that the steps for constructing the stop-go switching direction sequence based on the short-period velocity fluctuation sequence are as follows:

[0048] Obtain the velocity difference sequence, and compare each velocity difference in the velocity difference sequence with zero to determine the direction of change corresponding to each velocity difference;

[0049] When the corresponding speed difference is greater than zero, the corresponding direction of change is determined to be an increase in speed; when the corresponding speed difference is less than zero, the corresponding direction of change is determined to be a decrease in speed; when the corresponding speed difference is equal to zero, the corresponding direction of change is determined to be a maintenance of speed.

[0050] Based on the order of the speed differences in the speed difference sequence, the directions of change are arranged sequentially to construct the corresponding stop-go switching direction sequence.

[0051] In this embodiment, it should be specifically explained that the steps for obtaining the stop-go switching frequency parameter are as follows:

[0052] Obtain the stop-go switching direction sequence corresponding to the target road segment, and assign a numerical label to each direction item in the stop-go switching direction sequence; if the corresponding direction item indicates that the vehicle speed is increasing, then the direction label value corresponding to the direction item is recorded as 1; if the corresponding direction item indicates that the vehicle speed is decreasing, then the direction label value corresponding to the direction item is recorded as -1; if the corresponding direction item indicates that the vehicle speed is maintaining, then the direction label value corresponding to the direction item is recorded as 0.

[0053] The effective direction sequence corresponding to the target road segment is obtained by removing the direction items with a direction identifier value of 0 from the stop-go switching direction sequence. The effective direction sequence is used to characterize the continuous switching relationship of the actual direction change of vehicle speed in the target road segment within a preset time window.

[0054] The number of direction terms in the effective direction sequence is counted to obtain the number of effective direction terms; when the number of effective direction terms is not greater than 1, the stop-go switching frequency parameter corresponding to the target road segment is set to 0;

[0055] When the number of valid direction items is greater than 1, iterate through adjacent direction items in the valid direction sequence and calculate the direction reversal identifier value corresponding to each adjacent direction item. The specific steps are as follows:

[0056] ;

[0057] In the formula, This is represented by the direction reversal identifier value corresponding to the j-th and (j+1)-th adjacent direction items. This represents the direction identifier value corresponding to the j-th direction item in the valid direction sequence. This represents the direction identifier value corresponding to the (j+1)th direction item in the valid direction sequence; when two direction items are opposite, the direction reversal identifier value is 1; when two direction items are the same, the direction reversal identifier value is 0.

[0058] The reverse direction identifier values ​​corresponding to all adjacent direction items are summed to obtain the number of effective reverse switching for the target road segment. The number of effective reverse switching is divided by the number of effective direction items minus 1 to obtain the stop-go switching frequency parameter for the target road segment.

[0059] It should be specifically noted that, in this embodiment, the target road segment under intermittent congestion typically exhibits alternating changes in vehicle speed between decreasing and increasing. Therefore, the number of direction reversal switches in the effective direction sequence can reflect the frequency of stop-and-go alternation behavior on the road segment within a preset time window. By removing the direction items corresponding to speed maintenance before statistically analyzing the direction reversal switches, the interference of speed maintenance status on the switching identification results can be avoided, thereby enabling the calculated stop-and-go switching frequency parameter to more accurately characterize the intermittent congestion characteristics of the target road segment.

[0060] In this embodiment, it should be specifically explained that the steps for obtaining the low-speed continuous discrete parameters are as follows:

[0061] Obtain the short-period speed fluctuation sequence corresponding to the target road segment and the vehicle traffic speed time series corresponding to the short-period speed fluctuation sequence; wherein, each fluctuation value in the short-period speed fluctuation sequence is matched one-to-one with the corresponding sampling position in the vehicle traffic speed time series when the fluctuation value is constructed, so as to establish the correspondence between each fluctuation value and each vehicle traffic speed data.

[0062] Calculate the median value of each vehicle speed data in the vehicle speed time series to obtain the low-speed baseline speed corresponding to the target road segment; and calculate the average value of each fluctuation value in the short-period speed fluctuation series to obtain the fluctuation baseline amplitude corresponding to the target road segment.

[0063] The vehicle speed data is compared with the low-speed reference speed, and the corresponding fluctuation values ​​are compared with the fluctuation reference amplitude. When a vehicle speed data is not greater than the low-speed reference speed and its corresponding fluctuation value is not less than the fluctuation reference amplitude, the location is determined to be in a low-speed disturbance state, and the corresponding state identifier value is recorded as 1. Otherwise, the location is determined not to be in a low-speed disturbance state, and the corresponding state identifier value is recorded as 0, thereby constructing the low-speed disturbance state sequence corresponding to the target road segment.

[0064] The segments with consecutive values ​​of 1 in the low-speed disturbance state sequence are identified, and each segment with consecutive values ​​of 1 is determined as a low-speed continuous segment. The number of consecutive state identifier values ​​of 1 in each low-speed continuous segment is counted to obtain the segment duration length corresponding to each low-speed continuous segment.

[0065] It should be specifically noted that, in this embodiment, the "segment with a continuous value of 1" in the low-speed disturbance state sequence refers to a segment in the low-speed disturbance state sequence where the state identifier values ​​corresponding to adjacent sampling positions are all 1 and there is no interval with a value of 0 in between.

[0066] The number of low-speed continuous segments corresponding to the target road segment is counted to obtain the number of segments; and the duration of each low-speed continuous segment is summed to obtain the total duration of the segments.

[0067] When the number of segments is greater than 0, the total duration of segments is divided by the number of segments to obtain the average duration of segments; and the sum of the absolute values ​​of the length differences between the duration of each low-speed continuous segment and the average duration of segments is calculated to obtain the total length dispersion.

[0068] When the number of segments is greater than 1, the total length discrete value is divided by the product of the total duration of the segments and the number of segments minus 1 to obtain the low-speed continuous discrete parameter corresponding to the target road segment; when the number of segments is not greater than 1, the low-speed continuous discrete parameter corresponding to the target road segment is set to 0.

[0069] In this embodiment, it should be specifically explained that the steps for obtaining the fluctuation recovery anomaly parameters are as follows:

[0070] Obtain the short-period speed fluctuation sequence corresponding to the target road segment, sort the fluctuation values ​​in the short-period speed fluctuation sequence according to their numerical values, calculate the upper quartile value corresponding to the sorting result, and obtain the high-level reference amplitude of the fluctuation corresponding to the target road segment.

[0071] It should be noted that the upper quartile value is used to characterize the statistical distribution position of higher fluctuation levels in a short-period velocity fluctuation sequence. Specifically, after sorting the fluctuation values ​​in the short-period velocity fluctuation sequence according to their numerical magnitude, the fluctuation value corresponding to the 75th quartile position in the sorting result is taken as the upper quartile value. By using the upper quartile value as the high-level baseline amplitude of fluctuations, outlier interference caused by directly using the maximum fluctuation value can be avoided, while stably reflecting the significant fluctuation level of the target road segment within a preset time window, thereby improving the accuracy and stability of high-level fluctuation trigger location identification.

[0072] Each fluctuation value in the short-period velocity fluctuation sequence is compared with the high-level fluctuation reference amplitude. When a fluctuation value is not less than the high-level fluctuation reference amplitude, and the fluctuation value corresponding to its previous sampling position is less than the high-level fluctuation reference amplitude, the sampling position is determined as a high-level fluctuation trigger position. When a fluctuation value corresponds to the first sampling position, and the fluctuation value is not less than the high-level fluctuation reference amplitude, the sampling position is determined as a high-level fluctuation trigger position.

[0073] For each high-level fluctuation trigger position, starting from the corresponding high-level fluctuation trigger position, search for the fallback completion position one by one along the subsequent sampling direction of the short-period velocity fluctuation sequence; when the fluctuation value corresponding to a subsequent sampling position is less than the high-level fluctuation reference amplitude, and the fluctuation values ​​corresponding to this sampling position and the next sampling position are both less than the high-level fluctuation reference amplitude, the sampling position is determined as the fallback completion position of the corresponding high-level fluctuation trigger position.

[0074] When the sampling position is at the end of the short-period velocity fluctuation sequence, if the fluctuation value corresponding to the sampling position is less than the high-level reference amplitude of the fluctuation, it is directly determined as the position where the fall is completed.

[0075] If a high-level fluctuation trigger position cannot be found within a preset time window, the sampling position corresponding to the end of the short-cycle speed fluctuation sequence will be determined as the pullback completion position corresponding to the high-level fluctuation trigger position.

[0076] For each high-level fluctuation trigger position, the number of samples between the corresponding high-level fluctuation trigger position and the corresponding fall-off completion position is counted to obtain the corresponding recovery length; and the difference between the fluctuation value corresponding to the high-level fluctuation trigger position and the high-level fluctuation reference amplitude is obtained to obtain the corresponding overamplitude.

[0077] For each high-level fluctuation trigger position, when the corresponding overamplitude is greater than 0, the corresponding recovery length is divided by the corresponding overamplitude to obtain the corresponding recovery anomaly coefficient; when the corresponding overamplitude is not greater than 0, the recovery anomaly coefficient corresponding to the high-level fluctuation trigger position is not calculated.

[0078] The number of recovery anomaly coefficients corresponding to the target road segment is counted to obtain the number of effective recovery samples. When the number of effective recovery samples is greater than 0, the median value of all recovery anomaly coefficients is calculated to obtain the fluctuation recovery anomaly parameter corresponding to the target road segment. When the number of effective recovery samples is 0, the fluctuation recovery anomaly parameter corresponding to the target road segment is set to 0.

[0079] In this embodiment, it should be specifically explained that the steps for determining the surface traffic flow condition based on the average traffic speed and congestion level of the road segment are as follows:

[0080] The road segments within the target ice and snow tourism area are classified according to road type, resulting in multiple sets of similar road segments;

[0081] It should be noted that in this embodiment, the classification of road segments within the target ice and snow tourism area can be based on the road grade, traffic function, or usage attributes of each road segment. Specifically, each road segment can be divided into different types such as main roads of the scenic area, secondary roads of the scenic area, scenic roads, connecting roads, or parking lot entrance and exit roads, and a corresponding set of similar road segments can be constructed within the scope of the same type of road segment.

[0082] By using the above classification method, road segments with similar traffic capacity and traffic operation characteristics can be statistically analyzed in the same set. This ensures that the road segments are comparable when calculating the regional average speed benchmark parameter and the regional congestion level benchmark parameter, avoids the interference of differences in traffic characteristics between different road types on the judgment results, and improves the rationality and accuracy of the surface traffic conditions judgment results.

[0083] Since the road type classification method is a conventional technical means in existing traffic management and traffic data processing, this embodiment will not specify its specific classification rules in detail.

[0084] Within each set of similar road segments, the average traffic speed of each road segment is obtained, and the median value of the average traffic speed of each road segment is calculated to obtain the regional average speed benchmark parameter for the set of similar road segments.

[0085] Within each set of similar road segments, obtain the congestion level of each road segment and calculate the median value of the congestion level of each road segment to obtain the regional congestion level benchmark parameter for the set of similar road segments.

[0086] The average traffic speed of the target road segment is compared with the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is compared with the regional congestion level benchmark parameter of the set of similar road segments.

[0087] If the average traffic speed of the target road segment is not less than the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is not greater than the regional congestion level benchmark parameter of the set of similar road segments, the surface traffic flow condition of the target road segment is determined to be satisfied; otherwise, the surface traffic flow condition of the target road segment is determined to be unsatisfactory.

[0088] It should be noted that, in this embodiment, the surface traffic flow condition determination result is used to characterize whether the target road segment is in a state of no significant congestion at the overall average traffic level, rather than to characterize whether the target road segment has intermittent congestion at the short-cycle traffic fluctuation level. The overall traffic status of the target road segment is macroscopically determined by the average traffic speed and congestion level of the segment, and then the short-cycle disturbance state of the target road segment is microscopically identified by combining the intermittent congestion judgment index, thereby achieving accurate identification of road segments that appear to be smooth but actually have intermittent congestion.

[0089] In this embodiment, it should be specifically explained that the step of determining whether a road segment is in an intermittent congestion state based on the surface unobstructed condition judgment result and the intermittent congestion judgment index is as follows:

[0090] The system obtains the surface traffic flow condition assessment result for the target road segment and the intermittent congestion benchmark index for the set of similar road segments. If the assessment result is satisfactory and the intermittent congestion assessment index for the target road segment is greater than the intermittent congestion benchmark index for the set of similar road segments, the target road segment is determined to be in an intermittent congestion state. Otherwise, the target road segment is determined not to be in an intermittent congestion state. It should be noted that the intermittent congestion benchmark index for the set of similar road segments is used to characterize the overall distribution level of the intermittent congestion assessment index for each road segment within the set of similar road segments to which the target road segment belongs, within a preset time window. Specifically, the intermittent congestion assessment index for each road segment within the set of similar road segments to which the target road segment belongs can be obtained, and the median value can be calculated based on the intermittent congestion assessment index for each road segment to obtain the intermittent congestion benchmark index for the set of similar road segments.

[0091] In this embodiment, it is necessary to specifically explain the steps for correcting the road segment traffic evaluation data corresponding to the road segment to obtain the corrected candidate path evaluation dataset:

[0092] Obtain the intermittent congestion judgment index corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments to which the target road segment belongs; and obtain the road segment travel time data, road segment congestion degree data, road segment safety risk data, and road segment environmental impact data corresponding to the target road segment.

[0093] Based on the intermittent congestion judgment index corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments, the intermittent congestion correction coefficient corresponding to the target road segment is calculated. The specific steps are as follows:

[0094] ;

[0095] In the formula, This is represented as the intermittent congestion correction factor. Since correction is only applied when the intermittent congestion judgment index is greater than the intermittent congestion benchmark index, the intermittent congestion correction factor is a positive value. This is represented as the intermittent blockage judgment index. Represented as the intermittent blockage benchmark index;

[0096] Based on the intermittent congestion correction coefficient, the travel time data of the target road segment is corrected to obtain the corrected travel time data. The specific steps are as follows:

[0097] ;

[0098] In the formula, This is represented as the corrected travel time data for the road segment. This is represented as the travel time data for the road segment. This is expressed as the intermittent blockage correction factor;

[0099] Based on the intermittent congestion correction coefficient, the congestion level data of the target road segment is corrected to obtain the corrected road segment congestion level data. The specific steps for obtaining this data are as follows:

[0100] ;

[0101] In the formula, This is represented as the corrected data on road congestion levels. This data represents the degree of traffic congestion on a road segment.

[0102] Based on the intermittent congestion correction coefficient, the road segment safety risk data corresponding to the target road segment is corrected to obtain the corrected road segment safety risk data. The specific steps for obtaining this data are as follows:

[0103] ;

[0104] In the formula, This is represented as the corrected road segment safety risk data. This is represented as road segment safety risk data;

[0105] The environmental impact data of the target road segment is retained as the corrected environmental impact data of the road segment. The corrected road segment travel time data, the corrected road segment congestion data, the corrected road segment safety risk data, and the corrected road segment environmental impact data are jointly determined as the corrected road segment traffic evaluation data corresponding to the target road segment.

[0106] The original road segment traffic evaluation data corresponding to the target road segment in the candidate path evaluation dataset is replaced with the corrected road segment traffic evaluation data, and the original road segment traffic evaluation data corresponding to the remaining road segments in the candidate path evaluation dataset that are not identified as intermittent congestion are retained, thus obtaining the corrected candidate path evaluation dataset.

[0107] It should be specifically noted that, in this embodiment, the road segment environmental impact data is used to characterize the impact of the current environmental state of the target road segment on traffic conditions. This type of data mainly comes from meteorological conditions or external environmental factors, and its changes have been reflected in the real-time traffic status data. Therefore, when correcting the road segment traffic evaluation data, the road segment environmental impact data will not be repeatedly corrected to avoid the environmental impact being repeatedly included.

[0108] Step 6: Based on the revised candidate path evaluation dataset, perform path optimization analysis on each candidate route between the preset starting point and the target endpoint to determine the target dynamic optimal path;

[0109] It should be noted that, in this embodiment, when performing path optimization analysis on each candidate travel path between the preset starting point and the target endpoint, a path search optimization algorithm in the prior art can be used. The path search optimization algorithm can be Dijkstra's algorithm, A* algorithm, or other graph search optimization algorithms.

[0110] Specifically, based on the road segment traffic evaluation data corresponding to the road segments included in each candidate path after correction, the cumulative cost of each candidate path can be calculated, and the candidate path with the minimum cumulative cost can be selected as the target dynamic optimal path.

[0111] Step 7: Upon receiving new multi-source traffic data, update the real-time traffic status data, intermittent congestion status, and the corrected candidate path evaluation dataset, and re-execute the path optimization analysis based on the updated candidate path evaluation dataset to output the updated target dynamic optimal path.

[0112] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0113] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion, characterized in that, Includes the following steps: Step 1: Obtain multi-source traffic data within the target ice and snow tourism area and construct the original traffic dataset; the multi-source traffic data includes vehicle speed data collected by road monitoring equipment, vehicle trajectory data uploaded by vehicle terminals, environmental monitoring data collected by meteorological monitoring equipment, and visitor flow distribution data collected by scenic area visitor flow monitoring equipment. Step 2: Preprocess the original traffic dataset to obtain a standardized traffic dataset; Step 3: Perform multi-source fusion analysis on the standardized traffic dataset, and input the fusion analysis results into the preset traffic state analysis model for state identification to obtain real-time traffic state data corresponding to each road segment. The real-time traffic state data includes the average traffic speed of the road segment, the traffic density of the road segment, the congestion level of the road segment, the snow / ice accumulation status of the road segment, the weather impact level of the road segment, and the traffic risk level of the road segment. Step 4: Based on the real-time traffic status data corresponding to each road segment, calculate the road segment traffic evaluation data corresponding to each road segment, and construct a candidate path evaluation dataset; the road segment traffic evaluation data includes road segment travel time data, road segment congestion level data, road segment safety risk data, and road segment environmental impact data; Step 5: Based on the standardized traffic dataset and the real-time traffic status data corresponding to each road segment, identify the intermittent congestion status corresponding to each road segment, and when an intermittent congestion status is identified, correct the road segment traffic evaluation data corresponding to the road segment to obtain the corrected candidate path evaluation dataset. Step 6: Based on the revised candidate path evaluation dataset, perform path optimization analysis on each candidate route between the preset starting point and the target endpoint to determine the target dynamic optimal path; Step 7: Upon receiving new multi-source traffic data, update the real-time traffic status data, intermittent congestion status, and the corrected candidate path evaluation dataset, and re-execute the path optimization analysis based on the updated candidate path evaluation dataset to output the updated target dynamic optimal path.

2. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 1, characterized in that: The steps for identifying the intermittent congestion status corresponding to each road segment are as follows: Obtain vehicle speed data for each road segment within a preset time window, and construct a short-period speed fluctuation sequence based on the vehicle speed data; A stop-start switching direction sequence is constructed based on a short-period velocity fluctuation sequence, and the stop-start switching frequency parameter is calculated based on the stop-start switching direction sequence. Low-speed continuous discrete parameters and fluctuation recovery anomaly parameters are calculated based on short-period velocity fluctuation sequences. The stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter are normalized. The intermittent blockage judgment index is calculated based on the normalized stop-go switching frequency parameter, low-speed continuous discrete parameter, and fluctuation recovery anomaly parameter. Obtain the average traffic speed and congestion level of each road segment, and obtain the surface smooth traffic condition judgment result based on the average traffic speed and congestion level of each road segment; Based on the surface unobstructed condition assessment results and the intermittent congestion assessment index, it is determined whether the road segment is in an intermittent congestion state.

3. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that, The steps for constructing a short-period speed fluctuation sequence based on vehicle traffic speed data are as follows: Arrange the vehicle speed data of each road segment in chronological order within a preset time window to construct the corresponding vehicle speed time series. Based on the vehicle speed data corresponding to adjacent time points in the vehicle speed time series, the speed difference between adjacent time points is calculated to obtain the speed difference sequence. The absolute value of each velocity difference in the velocity difference sequence is taken to obtain the corresponding short-period velocity fluctuation sequence.

4. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that, The steps for constructing the stop-and-go direction switching sequence based on the short-period velocity fluctuation sequence are as follows: Obtain the velocity difference sequence, and compare each velocity difference in the velocity difference sequence with zero to determine the direction of change corresponding to each velocity difference; When the corresponding speed difference is greater than zero, the corresponding direction of change is determined to be speed increase; when the corresponding speed difference is less than zero, the corresponding direction of change is determined to be speed decrease; when the corresponding speed difference is equal to zero, the corresponding direction of change is determined to be speed maintenance. Based on the order of the speed differences in the speed difference sequence, the directions of change are arranged sequentially to construct the corresponding stop-go switching direction sequence.

5. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that: The steps for obtaining the stop-go switching frequency parameter are as follows: Obtain the stop-go switching direction sequence corresponding to the target road segment, and assign numerical labels to each direction item in the stop-go switching direction sequence; if the corresponding direction item indicates that the vehicle speed is increasing, the direction label value corresponding to the direction item is recorded as 1; if the corresponding direction item indicates that the vehicle speed is decreasing, the direction label value corresponding to the direction item is recorded as -1; if the corresponding direction item indicates that the vehicle speed is maintaining, the direction label value corresponding to the direction item is recorded as 0. Remove all direction items with a direction identifier value of 0 from the stop-go switching direction sequence to obtain the effective direction sequence corresponding to the target road segment; The number of direction terms in the effective direction sequence is counted to obtain the number of effective direction terms; when the number of effective direction terms is not greater than 1, the stop-go switching frequency parameter corresponding to the target road segment is set to 0; When the number of valid direction items is greater than 1, traverse the two adjacent direction items in the valid direction sequence and calculate the direction reversal identifier value corresponding to each adjacent direction item. The reverse direction identifier values ​​corresponding to all adjacent direction items are summed to obtain the number of effective reverse switching for the target road segment. The number of effective reverse switching is divided by the number of effective direction items minus 1 to obtain the stop-go switching frequency parameter for the target road segment.

6. A method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2 or 3, characterized in that: The steps for obtaining the low-speed continuous discrete parameters are as follows: Obtain the short-period speed fluctuation sequence corresponding to the target road segment and the vehicle traffic speed time series corresponding to the short-period speed fluctuation sequence; Calculate the median value of each vehicle speed data in the vehicle speed time series to obtain the low-speed baseline speed corresponding to the target road segment; The average value of each fluctuation in the short-period velocity fluctuation sequence is calculated to obtain the reference amplitude of the fluctuation corresponding to the target road segment; Compare the speed data of each vehicle with the low-speed baseline speed, and compare the corresponding fluctuation values ​​with the fluctuation baseline amplitude. When the speed data of a certain vehicle is not greater than the low speed reference speed, and the corresponding fluctuation value is not less than the fluctuation reference amplitude, it is determined that the position is in a low speed disturbance state, and the corresponding state identifier value is recorded as 1. Otherwise, determine that the location is not in a low-speed disturbance state and record the corresponding state identifier value as 0, thereby constructing a low-speed disturbance state sequence corresponding to the target road segment; The segments with consecutive values ​​of 1 in the low-speed disturbance state sequence are identified, and each segment with consecutive values ​​of 1 is determined as a low-speed continuous segment. The number of consecutive state identifier values ​​of 1 in each low-speed continuous segment is counted to obtain the segment duration length corresponding to each low-speed continuous segment. The number of low-speed continuous segments corresponding to the target road segment is counted to obtain the number of segments; and the duration of each low-speed continuous segment is summed to obtain the total duration of the segments. When the number of segments is greater than 0, the total duration of segments is divided by the number of segments to obtain the average duration of segments; and the sum of the absolute values ​​of the length differences between the duration of each low-speed continuous segment and the average duration of segments is calculated to obtain the total length dispersion. When the number of segments is greater than 1, the total length discrete is divided by the product of the total duration of the segments and the number of segments minus 1 to obtain the low-speed continuous discrete parameter corresponding to the target road segment; when the number of segments is not greater than 1, the low-speed continuous discrete parameter corresponding to the target road segment is set to 0.

7. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that: The steps for obtaining the fluctuation recovery anomaly parameters are as follows: Obtain the short-period speed fluctuation sequence corresponding to the target road segment, sort the fluctuation values ​​in the short-period speed fluctuation sequence according to their numerical values, calculate the upper quartile value corresponding to the sorting result, and obtain the high-level reference amplitude of the fluctuation corresponding to the target road segment. Compare each fluctuation value in the short-period velocity fluctuation sequence with the high-level benchmark amplitude of the fluctuation; When a certain fluctuation value is not less than the high-level fluctuation reference amplitude, and the fluctuation value corresponding to its previous sampling position is less than the high-level fluctuation reference amplitude, the sampling position is determined as a high-level fluctuation trigger position; when a certain fluctuation value corresponds to the first sampling position, and the fluctuation value is not less than the high-level fluctuation reference amplitude, the sampling position is determined as a high-level fluctuation trigger position. For each high-level fluctuation trigger position, starting from the corresponding high-level fluctuation trigger position, the fallback completion position is found one by one along the subsequent sampling direction of the short-period velocity fluctuation sequence; when the fluctuation value corresponding to a subsequent sampling position is less than the high-level fluctuation reference amplitude, and the fluctuation values ​​corresponding to this sampling position and the next sampling position are both less than the high-level fluctuation reference amplitude, the sampling position is determined as the fallback completion position of the corresponding high-level fluctuation trigger position. When the sampling position is at the end of the short-period velocity fluctuation sequence, if the fluctuation value corresponding to the sampling position is less than the high-level reference amplitude of the fluctuation, it is directly determined as the position where the fall is completed. If a high-level fluctuation trigger position cannot be found within a preset time window, the sampling position corresponding to the end of the short-cycle speed fluctuation sequence will be determined as the pullback completion position corresponding to the high-level fluctuation trigger position. For each high-level fluctuation trigger position, the number of samples between the corresponding high-level fluctuation trigger position and the corresponding fall-off completion position is counted to obtain the corresponding recovery length; And obtain the difference between the fluctuation value corresponding to the high-level fluctuation trigger position and the high-level fluctuation reference amplitude to obtain the corresponding overamplitude; For each high-level fluctuation trigger position, when the corresponding overamplitude is greater than 0, the corresponding recovery length is divided by the corresponding overamplitude to obtain the corresponding recovery anomaly coefficient; When the corresponding overamplitude is not greater than 0, the recovery anomaly coefficient corresponding to the high-level fluctuation trigger position is not calculated; The number of recovery anomalies corresponding to the target road segment is counted to obtain the number of valid recovery samples. When the number of effective recovery samples is greater than 0, the median value of all recovery anomaly coefficients is calculated to obtain the fluctuation recovery anomaly parameters corresponding to the target road segment; When the number of effective recovery samples is 0, the fluctuation recovery anomaly parameter corresponding to the target road segment is set to 0.

8. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that: The steps for determining the surface traffic flow condition based on the average traffic speed and congestion level of the road segment are as follows: The road sections within the target ice and snow tourism area are classified according to road type, resulting in multiple sets of similar road sections; Within each set of road segments of the same type, obtain the average traffic speed of each road segment and calculate the median value of the average traffic speed of each road segment to obtain the regional average speed benchmark parameter of the corresponding set of road segments of the same type. Within each set of similar road segments, obtain the congestion level of each road segment and calculate the median value of the congestion level of each road segment to obtain the regional congestion level benchmark parameter for the corresponding set of similar road segments. The average traffic speed of the target road segment is compared with the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is compared with the regional congestion level benchmark parameter of the set of similar road segments. If the average traffic speed of the target road segment is not less than the regional average speed benchmark parameter of the set of similar road segments, and the congestion level of the target road segment is not greater than the regional congestion level benchmark parameter of the set of similar road segments, the surface traffic flow condition of the target road segment is determined to be satisfied; otherwise, the surface traffic flow condition of the target road segment is determined to be unsatisfactory.

9. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 2, characterized in that: The step of determining whether a road segment is in an intermittent congestion state based on the surface unobstructed condition judgment result and the intermittent congestion judgment index is as follows: Obtain the surface traffic flow condition judgment result corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments. If the judgment result is satisfied and the intermittent congestion judgment index corresponding to the target road segment is greater than the intermittent congestion benchmark index corresponding to the set of similar road segments, the target road segment is determined to be in an intermittent congestion state; otherwise, the target road segment is determined not to be in an intermittent congestion state.

10. The method for dynamic route optimization of ice and snow tourism transportation based on multi-source data fusion according to claim 1, characterized in that: The step of correcting the road segment traffic evaluation data corresponding to the road segment to obtain the corrected candidate path evaluation dataset is as follows: Obtain the intermittent congestion judgment index corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments to which the target road segment belongs; and obtain the segment travel time data, segment congestion degree data, segment safety risk data, and segment environmental impact data corresponding to the target road segment. Based on the intermittent congestion judgment index corresponding to the target road segment and the intermittent congestion benchmark index corresponding to the set of similar road segments, calculate the intermittent congestion correction coefficient corresponding to the target road segment; Based on the intermittent congestion correction coefficient, the travel time data of the target road segment is corrected to obtain the corrected travel time data of the road segment. Based on the intermittent congestion correction coefficient, the congestion level data of the target road segment is corrected to obtain the corrected congestion level data of the road segment. Based on the intermittent congestion correction coefficient, the road segment safety risk data corresponding to the target road segment is corrected to obtain the corrected road segment safety risk data; The environmental impact data of the target road segment is retained as the corrected environmental impact data of the road segment. The corrected road segment travel time data, the corrected road segment congestion data, the corrected road segment safety risk data, and the corrected road segment environmental impact data are jointly determined as the corrected road segment traffic evaluation data corresponding to the target road segment. The original road segment traffic evaluation data corresponding to the target road segment in the candidate path evaluation dataset is replaced with the corrected road segment traffic evaluation data, and the original road segment traffic evaluation data corresponding to the remaining road segments in the candidate path evaluation dataset that are not identified as intermittent congestion are retained, thus obtaining the corrected candidate path evaluation dataset.

Images