A driving safety assessment method and system for dynamic routes

Abstract

The present invention provides a driving safety assessment method and system for a dynamic route, which compares the restriction information of each road section on the route to be assessed with the vehicle parameters of the transport vehicle to determine the value of the first safety index; uses an average vehicle speed prediction model to predict the predicted average vehicle speed of each road section in multiple future time slices; determines the predicted time and reference time of passing the route to be assessed according to the predicted average vehicle speed and the reference speed, and determines the value of the second safety index; collects negative events and their duration on the route to be assessed at the current moment; predicts the estimated arrival time of driving to the location of the negative event, and determines the value of the third safety index based on the estimated arrival time and the duration of the negative event; and determines the value of the total safety index by combining the values ​​of the first safety index, the second safety index and the third safety index. The route to be assessed is safety assessed by the value of the total safety index to ensure the reliability, timeliness and efficiency of transportation.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic route safety situation assessment, and in particular to a driving safety assessment method and system for a dynamic route. Background Art

[0002] Road transport is currently the main mode of transportation. Road transport has stronger adaptability than other modes of transportation. It can adapt to different sizes of freight volumes by forming a fleet. Road transport is of special significance for cargo transportation during emergency rescue and disaster relief.

[0003] Given the complex road network and restrictions on the transportation of oversized and oversized goods, taking different driving routes has a huge impact on the reliability, timeliness and efficiency of transportation. Therefore, it is necessary to conduct a safety assessment of the driving route before driving. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a driving safety assessment method and system for a dynamic route to perform safety assessment on the driving route.

[0005] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] A first aspect of an embodiment of the present invention discloses a method for evaluating driving safety of a dynamic route, the method comprising:

[0007] Obtain restriction information for each road segment on the route to be evaluated;

[0008] Comparing the restriction information with vehicle parameters of the transport vehicle, and determining a value of the first safety index according to the comparison result;

[0009] Collecting the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to a plurality of consecutive time slices at the current moment to form an input vector;

[0010] Inputting the input vector into a pre-built average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain the predicted average vehicle speed of each road section on the route to be evaluated in multiple future time slices, wherein the average vehicle speed prediction model is obtained based on a preset model trained with sample data, and the sample data at least includes road network data, historical road network speed monitoring data for N consecutive days, and historical weather data;

[0011] Determining the predicted time taken for the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and determining the reference time taken for the transport vehicle to pass through the route to be evaluated according to a preset reference speed;

[0012] Determine a value of a second safety indicator by using the predicted time consumption and the reference time consumption;

[0013] Collecting negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events;

[0014] Predicting an estimated arrival time of the transport vehicle at the location of the negative event according to the predicted average vehicle speed, and determining a value of a third safety indicator based on the estimated arrival time and the duration of the negative event;

[0015] The value of the total safety index is determined by combining the value of the first safety index, the value of the second safety index, and the value of the third safety index.

[0016] Preferably, the process of training a preset model based on sample data to obtain an average vehicle speed prediction model includes:

[0017] Based on the sample data, the sample average vehicle speed and sample weather conditions of each road section in each time slice are counted, and the sample period information and sample date information corresponding to each time slice are counted;

[0018] Constructing a sample input vector using the sample average vehicle speed, the sample weather condition, the sample time period information and the sample date information;

[0019] A preset model is trained based on the sample input vector until the preset model converges to obtain an average vehicle speed prediction model, wherein the preset model is constructed based on a Transformer structure.

[0020] Preferably, comparing the restriction information with vehicle parameters of the transport vehicle, and determining the value of the first safety index according to the comparison result, includes:

[0021] Comparing the restriction information of each road section on the route to be evaluated with the vehicle parameters of the transport vehicle respectively;

[0022] If the vehicle parameters satisfy the restriction information of each road section on the route to be evaluated, determining the value of the first safety index to be a first preset value;

[0023] If the vehicle parameters do not satisfy the restriction information of any road section on the route to be evaluated, the value of the first safety index is determined to be a second preset value.

[0024] Preferably, determining the value of the third safety indicator based on the estimated arrival time and the duration of the negative event includes:

[0025] For each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable is accumulated by 1, and the value of the designated variable is used to represent the number of times the negative event is encountered;

[0026] A value of a third safety indicator is determined based on the value of the designated variable.

[0027] Preferably, determining the value of the total safety index by combining the value of the first safety index, the value of the second safety index and the value of the third safety index comprises:

[0028] The product of the value of the first safety index, the value of the second safety index and the value of the third safety index is calculated to obtain the value of the total safety index.

[0029] Preferably, it also includes:

[0030] Determining, from a plurality of preset value intervals, a preset value interval in which the value of the total safety index lies;

[0031] Output the driving safety assessment level corresponding to the preset value interval in which the value of the total safety index is located, wherein each of the preset value intervals is set with a corresponding driving safety assessment level.

[0032] A second aspect of an embodiment of the present invention discloses a driving safety assessment system for a dynamic route, the system comprising:

[0033] An acquisition unit, used for acquiring restriction information of each road section on the route to be evaluated;

[0034] A first processing unit, configured to compare the restriction information with vehicle parameters of the transport vehicle, and determine a value of a first safety index according to the comparison result;

[0035] A first collection unit is used to collect the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to a plurality of consecutive time slices at the current moment to form an input vector;

[0036] A prediction unit, configured to input the input vector into a pre-built average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain a predicted average vehicle speed for each road section on the route to be evaluated in multiple future time slices, wherein the average vehicle speed prediction model is obtained by training a preset model based on sample data, and the sample data at least includes road network data, historical road network speed monitoring data for N consecutive days, and historical weather data;

[0037] A second processing unit, configured to determine a predicted time taken for the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and to determine a reference time taken for the transport vehicle to pass through the route to be evaluated according to a preset reference speed;

[0038] A first determining unit, configured to determine a value of a second safety indicator by using the predicted time consumption and the reference time consumption;

[0039] A second collecting unit is used to collect negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events;

[0040] a third processing unit, configured to predict an estimated arrival time of the transport vehicle at the location of the negative event according to the predicted average vehicle speed, and determine a value of a third safety indicator based on the estimated arrival time and the duration of the negative event;

[0041] The second determining unit is used to determine a value of a total safety index by combining a value of the first safety index, a value of the second safety index, and a value of the third safety index.

[0042] Preferably, the prediction unit comprises:

[0043] The statistical module is used to count the sample average vehicle speed and sample weather conditions of each road section in each time slice based on the sample data, and to count the sample period information and sample date information corresponding to each time slice;

[0044] A construction module, used to construct a sample input vector using the sample average vehicle speed, the sample weather condition, the sample time period information and the sample date information;

[0045] A training module is used to train a preset model based on the sample input vector until the preset model converges to obtain an average vehicle speed prediction model, wherein the preset model is constructed based on a Transformer structure.

[0046] Preferably, the first processing unit is specifically used to: compare the restriction information of each section on the route to be evaluated with the vehicle parameters of the transport vehicle; if the vehicle parameters meet the restriction information of each section on the route to be evaluated, determine the value of the first safety indicator to be a first preset value; if the vehicle parameters do not meet the restriction information of any section on the route to be evaluated, determine the value of the first safety indicator to be a second preset value.

[0047] Preferably, the third processing unit for determining the value of the third safety indicator based on the estimated arrival time and the duration of the negative event is specifically configured to:

[0048] For each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable is accumulated by 1, and the value of the designated variable is used to represent the number of times the negative event is encountered;

[0049] A value of a third safety indicator is determined based on the value of the designated variable.

[0050] A driving safety assessment method and system for a dynamic route based on the above-mentioned embodiment of the present invention is provided. The method comprises: obtaining restriction information of each road section on the route to be assessed; comparing the restriction information with vehicle parameters of the transport vehicle, and determining the value of the first safety index according to the comparison result; collecting the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be assessed corresponding to a plurality of consecutive time slices at the current moment to form an input vector; inputting the input vector into a pre-constructed average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain the predicted average vehicle speed of each road section on the route to be assessed in a plurality of future time slices; and determining the value of the first safety index according to the predicted average vehicle speed. Determine the predicted time for the transport vehicle to pass through the route to be evaluated, and determine the reference time for the transport vehicle to pass through the route to be evaluated based on the preset reference speed; determine the value of the second safety index using the predicted time and the reference time; collect the negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events; predict the estimated arrival time of the transport vehicle to the location of the negative event based on the predicted average speed, and determine the value of the third safety index based on the estimated arrival time and the duration of the negative event; determine the value of the total safety index by combining the values ​​of the first safety index, the second safety index, and the third safety index. In this solution, the value of the first safety index is determined by comparing the vehicle parameters of the transport vehicle and the restriction information of each section on the route to be evaluated. The predicted time for passing through the route to be evaluated is determined using the average speed prediction model, and the value of the second safety index is determined based on the predicted time and the reference time. The value of the third safety index is determined using the duration of the negative event and the predicted estimated arrival time of the transport vehicle to the location of the negative event. The value of the total safety index is calculated by combining the values ​​of the first safety index, the second safety index and the third safety index. The value of the total safety index is used to conduct a safety assessment on the route to be assessed to ensure the reliability, timeliness and efficiency of transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0052] Figure 1 A flow chart of a method for evaluating driving safety on a dynamic route provided by an embodiment of the present invention;

[0053] Figure 2 A flowchart of an average vehicle speed prediction model obtained through training provided by an embodiment of the present invention;

[0054] Figure 3A structural block diagram of a dynamic route driving safety assessment system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] In this application, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element.

[0057] As can be seen from the background technology, given the complex road network and restrictions on the transportation of oversized and oversized goods, taking different driving routes has a huge impact on the reliability, timeliness and efficiency of transportation. Therefore, it is necessary to conduct a safety assessment of the driving route before driving.

[0058] Therefore, an embodiment of the present invention provides a method and system for evaluating driving safety of a dynamic route, which determines the value of a first safety index by comparing the vehicle parameters of the transport vehicle and the restriction information of each road section on the route to be evaluated. The predicted time required to pass through the route to be evaluated is determined using an average vehicle speed prediction model, and the value of a second safety index is determined based on the predicted time and the reference time. The value of a third safety index is determined using the duration of the negative event and the predicted estimated arrival time of the transport vehicle at the location of the negative event. The value of the total safety index is calculated by combining the values ​​of the first safety index, the second safety index, and the third safety index, and the route to be evaluated is safety evaluated using the value of the total safety index to ensure the reliability, timeliness, and efficiency of transportation.

[0059] See also Figure 1 , shows a flow chart of a driving safety assessment method for a dynamic route provided by an embodiment of the present invention, the driving safety assessment method comprising:

[0060] Step S101: Obtain restriction information of each road section on the route to be evaluated.

[0061] It should be noted that the route to be evaluated is a dynamic route that needs to be evaluated for driving safety; the route to be evaluated is one or more, and each route to be evaluated can adopt the embodiment of the present invention. Figure 1 The contents of each step are processed.

[0062] In the specific implementation of step S101, the restriction information of each road section on the route to be evaluated is obtained from the road network data (such as national road network data). The restriction information of the road section includes but is not limited to: height limit, weight limit, turning radius, slope, etc.

[0063] Step S102: Compare the restriction information with the vehicle parameters of the transport vehicle, and determine the value of the first safety index according to the comparison result.

[0064] It should be noted that the transport vehicles are vehicles that are about to travel on the routes to be evaluated; in order to ensure the reliability, timeliness and efficiency of transportation, before the transport vehicles transport through the routes to be evaluated, they need to use this solution to conduct a driving safety assessment on each route to be evaluated, and then determine whether each route to be evaluated is suitable for transportation.

[0065] In the specific implementation of step S102, the vehicle parameters of the transport vehicle are compared with the restriction information of each section on the route to be evaluated, and the value of the first safety index (denoted as K1) is determined according to the comparison result.

[0066] It should be noted that the vehicle parameters of the transport vehicle include but are not limited to: the vehicle height of the transport vehicle, the load capacity of the transport vehicle, the turning radius of the transport vehicle, and the maximum climbing data of the transport vehicle.

[0067] When comparing the vehicle parameters of the transport vehicle with the restriction information of each section on the route to be evaluated, the vehicle height, load, turning radius, and maximum climbing data of the transport vehicle are compared with the restriction information of each section on the route to be evaluated to obtain the corresponding comparison results; the comparison results corresponding to each section on the route to be evaluated can be used to indicate whether the transport vehicle meets the restriction information of the section; if it does, it means that the section is suitable for the transport vehicle to pass; if it does not, it means that the section is not suitable for the transport vehicle to pass. By comparing the vehicle parameters and restriction information, it can be determined whether there are sections on the route to be evaluated that the transport vehicle cannot pass, and then determine whether the route to be evaluated meets the rigid requirements such as vehicle height limit and vehicle weight limit.

[0068] In some embodiments, the restriction information of each section on the route to be evaluated is compared with the vehicle parameters of the transport vehicle; if the vehicle parameters meet the restriction information of each section on the route to be evaluated, the value of the first safety indicator is determined to be a first preset value (for example, 1); if the vehicle parameters do not meet the restriction information of any section on the route to be evaluated, the value of the first safety indicator is determined to be a second preset value (for example, 0).

[0069] That is to say, if the route to be evaluated contains a road section that is inaccessible to transport vehicles, K1=0; if all sections of the route to be evaluated are accessible to transport vehicles, K1=1.

[0070] Step S103: Collect the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to multiple consecutive time slices at the current moment to form an input vector.

[0071] In the process of implementing step S103, the average vehicle speed, weather conditions, time period information and date information of each section on the route to be evaluated corresponding to multiple consecutive time slices at the current moment are collected; for example: the average vehicle speed, weather conditions, time period information and date information of each section on the route to be evaluated corresponding to T consecutive time slices at the current moment are collected.

[0072] The input vector is constructed using the average vehicle speed, weather conditions, time period information and date information of each road section corresponding to T consecutive time slices at the current moment.

[0073] It should be noted that the input vector is composed of the average vehicle speed of a certain section of a road corresponding to a certain time slice of a certain day, the weather conditions of a certain section of the road, the time period information, and the date information. For example, taking the jth time slice of the i-th day of the section g on the route to be evaluated as an example, the constructed input vector is recorded as is the average vehicle speed corresponding to the jth time slice on the i-th day of the road section g on the route to be evaluated, is the weather condition corresponding to the jth time slice of the i-th day on the route to be evaluated, time i,j is the time period information corresponding to the jth time slice of the i-th day, day i,j The date information corresponding to the j-th time slice on the i-th day.

[0074] Step S104: inputting the input vector into a pre-built average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain the predicted average vehicle speed of each road section on the route to be evaluated in multiple future time slices.

[0075] It should be noted that the average vehicle speed prediction model is obtained by training a preset model based on sample data. The specific process of training the average vehicle speed prediction model can be found in the following embodiment of the present invention. Figure 2 It should be noted that the sample data at least includes road network data, historical road network vehicle speed monitoring data for N consecutive days, and historical weather data for the corresponding date.

[0076] Among them, the road network data includes the following fields: the latitude and longitude of the starting point of the road section, the latitude and longitude of the end point of the road section, the road grade, the height limit of the road section, the weight limit of the road section, the turning radius, and the slope. The historical road network speed monitoring data includes the following fields: timestamp, license plate number, and vehicle speed. The historical weather data includes the following fields: timestamp, temperature, and weather conditions.

[0077] In the specific implementation of step S104, the input vector constructed in step S103 is input into the average speed prediction model for speed prediction, thereby predicting the average speed of each section on the route to be evaluated in multiple future time slices. For example, the average speed of each section on the route to be evaluated in de_max_step future time slices is predicted.

[0078] It should be noted that “the predicted average vehicle speed of the road section in multiple future time slices” specifically refers to: the predicted average vehicle speed of the road section in each future time slice.

[0079] The predicted average speed can be expressed as It represents the predicted average vehicle speed of the road section g on the route to be evaluated in the mth time slice in the future, where m∈[1,de_max_step], g∈[1,G], and G is the total number of road sections on the route to be evaluated.

[0080] Step S105: determining the predicted time taken by the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and determining the reference time taken by the transport vehicle to pass through the route to be evaluated according to a preset reference speed.

[0081] In the process of implementing step S105, the predicted time taken for the transport vehicle to pass through the route to be evaluated is determined based on the predicted average speed (equivalent to the estimated time taken for the transport vehicle to pass through the route to be evaluated). The specific method of calculating the predicted time is as follows: assuming that the transport vehicle starts from the starting point of the route to be evaluated at time t0, whenever the transport vehicle travels to a new section on the route to be evaluated (or when the time is in a new time slice), the speed of the transport vehicle is updated to the predicted average speed in the time slice corresponding to the corresponding section (the section where the transport vehicle is traveling). Then, the distance traveled by the transport vehicle at each predicted average speed is divided by the corresponding predicted average speed to obtain the time consumed by the transport vehicle in each section of the driving process, that is, the time consumed in a certain section = the distance traveled by the transport vehicle at a certain predicted average speed / the predicted average speed; the time consumed in each section between the starting point and the end point of the route to be evaluated is accumulated to obtain the predicted time consumed by the transport vehicle through the route to be evaluated (denoted as T predict ).

[0082] The reference time taken by the transport vehicle to pass through the route to be evaluated is determined according to the preset reference speed. The reference time is calculated by calculating the straight-line distance between the starting point and the end point of the route to be evaluated (denoted as S) according to the longitude and latitude of the starting point and the longitude and latitude of the end point of the route to be evaluated, and then setting the reference speed V ref Using the straight-line distance and reference speed calculated above, the reference time (denoted as T) for the transport vehicle to pass through the route to be evaluated can be calculated. ref ).

[0083] It should be noted that the method of using longitude and latitude to calculate the straight-line distance between the starting point and the end point of the route to be evaluated can be referred to in the following embodiment of the present invention. Figure 2 The relevant contents of calculating the road section length in step S201.

[0084] Step S106: Determine the value of the second safety indicator by using the predicted time consumption and the reference time consumption.

[0085] In the specific implementation of step S106, the predicted time and the reference time are used to determine the value of the second safety index (denoted as K2). Specifically, the ratio between the reference time and the predicted time is calculated to obtain the value of the second safety index, that is, K2 = T ref / T predict .

[0086] Step S107: Collecting negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events.

[0087] In the specific process of implementing step S107, negative events that affect driving safety on the route to be evaluated at the current moment are collected, and the duration of each negative event is collected.

[0088] For example: collect negative events that seriously affect driving safety, such as traffic accidents, landslides, road collapses, and severe weather (heavy to torrential rain, heavy to blizzards, and heavy fog) on ​​the route to be evaluated at the current moment; record the duration of the traffic accident as TS1, the duration of the landslide as TS2, the duration of the road collapse as TS3, and the duration of the severe weather as TS4.

[0089] Step S108: predicting the estimated arrival time of the transport vehicle to the location of the negative event according to the predicted average vehicle speed, and determining the value of the third safety index based on the estimated arrival time and the duration of the negative event.

[0090] In the specific implementation of step S108, the time when the transport vehicle reaches the location of the negative event (this time is called the estimated arrival time) is predicted based on the predicted average vehicle speed; wherein, the "location of the negative event" specifically refers to the point where the negative event occurs.

[0091] For each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable (denoted as C) is accumulated by 1. The value of the designated variable is used to indicate the number of times the negative event is encountered.

[0092] Specifically, the estimated arrival time of the transport vehicle at the point where the negative event occurs is predicted; if the transport vehicle is still within the duration of a negative event when it reaches the point where the negative event occurs (that is, the estimated arrival time of the transport vehicle to the location where the negative event occurs is within the duration of the negative event), then the value of C is increased by 1; it should be noted that the initial value of C is set to 0 (the maximum value can be set to 5, the maximum value of C is only for example).

[0093] After the designated variable C is updated in the above manner, the value of the third safety index (denoted as K3) is determined based on the value of the designated variable; specifically, the value of the third safety index is calculated using formula (1).

[0094] K3=1-0.2C (1)

[0095] Step S109: Determine the value of the total safety index by combining the value of the first safety index, the value of the second safety index, and the value of the third safety index.

[0096] In the specific implementation of step S109, the product of the value of the first safety index, the value of the second safety index and the value of the third safety index is calculated to obtain the value of the total safety index (denoted as K) corresponding to the route to be evaluated.

[0097] Specifically, the value of the total safety index is calculated by formula (2).

[0098] K=K1×K2×K3 (2)

[0099] It should be noted that a plurality of preset numerical intervals are pre-set, and each preset numerical interval is provided with a corresponding driving safety assessment level; the driving safety assessment levels can be respectively divided into: excellent, good, medium, poor, and bad.

[0100] In some embodiments, after the value of the total safety index is calculated, the preset value interval of the total safety index is determined from multiple preset value intervals; and the driving safety assessment level corresponding to the preset value interval of the total safety index is output. The driving safety assessment level corresponding to the preset value interval of the total safety index is the driving safety assessment level of the route to be assessed.

[0101] For example: when K≥0.8, the driving safety assessment level corresponding to the value of the total safety index is excellent; when 0.6≤K<0.8, the driving safety assessment level corresponding to the value of the total safety index is good; when 0.4≤K<0.6, the driving safety assessment level corresponding to the value of the total safety index is medium; when 0.2≤K<0.4, the driving safety assessment level corresponding to the value of the total safety index is poor; when 0≤K<0.2, the driving safety assessment level corresponding to the value of the total safety index is poor.

[0102] It should be noted that the above method can be used to calculate the value of the corresponding total safety index for each route to be evaluated; the driving safety assessment level of the corresponding route to be evaluated can be determined based on the value of the total safety index.

[0103] In an embodiment of the present invention, the value of the first safety index is determined by comparing the vehicle parameters of the transport vehicle and the restriction information of each road section on the route to be evaluated. The average vehicle speed prediction model is used to determine the predicted time required to pass through the route to be evaluated, and the value of the second safety index is determined based on the predicted time and the reference time. The value of the third safety index is determined using the duration of the negative event and the predicted estimated arrival time of the transport vehicle at the location of the negative event. The value of the total safety index is calculated by combining the values ​​of the first safety index, the second safety index, and the third safety index, and the route to be evaluated is safety evaluated by the value of the total safety index to ensure the reliability, timeliness, and efficiency of transportation.

[0104] The above embodiments of the present invention Figure 1 The content of the average vehicle speed prediction model obtained by training in step S104 is shown in Figure 2 , shows a flow chart of training an average vehicle speed prediction model provided by an embodiment of the present invention, Figure 2 The following steps are involved:

[0105] Step S201: Based on the sample data, the sample average vehicle speed and sample weather conditions of each road section in each time slice are counted, and the sample time period information and sample date information corresponding to each time slice are counted.

[0106] In the specific implementation of step S201, sample data is obtained, and the sample data at least includes road network data, historical road network vehicle speed monitoring data for N consecutive days, and historical weather data for corresponding dates.

[0107] The sample data is preprocessed. The specific preprocessing method at least includes the following contents: data cleaning of the sample data, deletion of invalid data in the sample data, and completion of missing fields in the sample data; and correction processing of the historical road network vehicle speed monitoring data. If the historical road network vehicle speed monitoring data monitored on a section is greater than the speed limit of the section, the historical road network vehicle speed monitoring data that is greater than the speed limit needs to be lowered to the maximum speed limit.

[0108] In some embodiments, after the sample data is preprocessed, the preprocessed sample data is statistically analyzed to calculate the section length of each road section; specifically, the section length is calculated based on the longitude and latitude of the starting point and the longitude and latitude of the ending point of the section in combination with formula (3) and formula (4).

[0109] θ=arccos(sinS lat ×sinE lat +cosS lat ×cosE lat ×cos(E lon -S lon )) (3)

[0110] l SE =R×θ (4)

[0111] In formula (3) and formula (4), S lon is the longitude of the starting point of the road segment, S lat is the latitude of the starting point of the road segment, E lon is the longitude of the end point of the section, E lat is the latitude of the end point of the section, R is the radius of the earth, l SE It is the length of the section between the starting point (point S) and the end point (point E) of the section.

[0112] After preprocessing the sample data, the historical road network speed monitoring data for N consecutive days is divided into cnt time slices with an interval of τ0 (such as 30 minutes) from 0:00 to 23:50; and the sample average speed of each road section in each time slice is counted. Specifically, the sample average speed of each road section in each time slice is calculated by formula (5).

[0113]

[0114] In formula (5), is the total number of vehicles passing through the road section g during the j-th time slice on the i-th day, is the speed of the kth vehicle passing through the road section g in the jth time slice on the i-th day, is the sample average vehicle speed corresponding to the road section g in the jth time slice on the i-th day.

[0115] Then, the sample weather conditions of each road section in each time slice are counted. The sample weather conditions can be divided into five categories: heavy to torrential rain, moderate snow, heavy to blizzard, heavy fog, and others. The five categories of sample weather conditions are represented by 1, 2, 3, 4, and 5 respectively. The sample weather conditions are recorded as Represents the sample weather conditions corresponding to the road section g in the jth time slice on the i-th day.

[0116] Next, the sample period information corresponding to each time slice is counted. Specifically, starting from 0:00 a.m., a day is divided into a sample period every 3 hours, and a total of 8 sample periods are obtained. The 8 sample periods are represented by 1, 2, 3, 4, 5, 6, 7, and 8 respectively; the sample period information here is recorded as time i,j , time i,j Indicates the sample period information corresponding to the j-th time slice on the i-th day.

[0117] Next, the sample date information corresponding to each time slice is counted. Specifically, the date information of each day in the "N consecutive days" mentioned above is counted, and each day is divided into rest days (legal holidays, weekends, etc.) and working days. Rest days are represented by 0 and working days are represented by 1. The sample date information here is recorded as day i,j , day i,j Indicates the sample date information corresponding to the j-th time slice on the i-th day.

[0118] Step S202: construct a sample input vector using the sample average vehicle speed, sample weather conditions, sample time period information and sample date information.

[0119] In the process of implementing step S202, after obtaining the sample average vehicle speed, sample weather conditions, sample time period information and sample date information through the above step S201, a preset model (which is the average vehicle speed prediction model after training convergence) is constructed. The preset model is constructed based on the Transformer structure, that is, the preset model (which can be called the Transformer model) is constructed based on the Transformer structure.

[0120] Set the main parameters of the constructed preset model. For example, the main parameters of the preset model are set as follows: for the encoder part, embedding_size is set to 128, en_layers_num is set to 2, en_seq_len is set to 16, en_hidden_size is set to 128, and en_head_num is set to 4.

[0121] For the decoder part, the decoder adopts an autoregressive structure to achieve multi-step prediction, de_layers_num is set to 4, de_max_step is set to 128, de_hidden_size is set to 128, and de_head_num is set to 4. The input of each step of the autoregressive structure is the predicted average vehicle speed predicted in the previous time slice.

[0122] After setting up the preset model, use the sample average speed Sample weather conditions Sample period information time i,j and sample date information day i,j , construct the sample input vector for inputting the preset model. Taking the jth time slice of the i-th day of road section g as an example, the corresponding sample input vector is

[0123] Step S203: training a preset model based on the sample input vector until the preset model converges to obtain an average vehicle speed prediction model.

[0124] In the specific implementation process of step S203, after constructing the sample input vector, the preset model is trained based on the sample input vector until the preset model converges to obtain the average vehicle speed prediction model.

[0125] Specifically, each time the data consisting of the sample input vectors of T consecutive time slices are input into the preset model, the preset model outputs the predicted result of the average vehicle speed of the road section in the future de_max_step time slices (that is, the output predicted average vehicle speed), and the predicted result is compared with the actual monitoring data (that is, the actual average vehicle speed) and the error is back-propagated, so as to train the final average vehicle speed prediction model.

[0126] The above content is about the relevant instructions for training the average vehicle speed prediction model.

[0127] Through the above-mentioned embodiments of the present invention Figure 1 and Figure 2 As can be seen from the content, this solution has the following advantages:

[0128] (1) This solution comprehensively analyzes the factors that may affect the safety of driving routes, and builds an average speed prediction model based on road network data, speed monitoring data, weather, time period, and date. Based on the average speed prediction model, the predicted time required for transport vehicles to pass through the route to be evaluated can be calculated, thereby determining the traffic efficiency of the route to be evaluated.

[0129] (2) In view of the situation where the average vehicle speed of a road section has temporal coupling, this scheme specifically adopts the Transformer network and uses its self-attention structure to capture the temporal correlation of the average vehicle speed of the road section, thereby improving the accuracy of the predicted average vehicle speed.

[0130] (3) This solution comprehensively analyzes the reliability, efficiency and safety of the route to be evaluated from the three aspects of passability, traffic efficiency and traffic safety, and ultimately obtains a more accurate and comprehensive driving safety assessment level of the route to be evaluated. This provides effective guidance for decision makers on whether to choose the route to be evaluated. In addition, safety assessments can be performed on multiple routes to be evaluated to assist decision makers in selecting the best driving route.

[0131] Corresponding to the driving safety assessment method for a dynamic route provided by the above embodiment of the present invention, see Figure 3 , the embodiment of the present invention also provides a structural block diagram of a driving safety assessment system for a dynamic route, the driving safety assessment system comprising: an acquisition unit 301, a first processing unit 302, a first acquisition unit 303, a prediction unit 304, a second processing unit 305, a first determination unit 306, a second acquisition unit 307, a third processing unit 308, and a second determination unit 309;

[0132] The acquisition unit 301 is used to acquire restriction information of each road section on the route to be evaluated.

[0133] The first processing unit 302 is used to compare the restriction information with the vehicle parameters of the transport vehicle and determine the value of the first safety index according to the comparison result.

[0134] In a specific implementation, the first processing unit 302 is specifically used to: compare the restriction information of each section on the route to be evaluated with the vehicle parameters of the transport vehicle; if the vehicle parameters meet the restriction information of each section on the route to be evaluated, determine the value of the first safety indicator to be a first preset value; if the vehicle parameters do not meet the restriction information of any section on the route to be evaluated, determine the value of the first safety indicator to be a second preset value.

[0135] The first collection unit 303 is used to collect the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to multiple consecutive time slices at the current moment to form an input vector.

[0136] The prediction unit 304 is used to input the input vector into a pre-constructed average vehicle speed prediction model for vehicle speed prediction, so as to obtain the predicted average vehicle speed of each road section on the route to be evaluated in multiple future time slices, wherein the average vehicle speed prediction model is obtained based on a preset model trained based on sample data, and the sample data at least includes road network data, historical road network speed monitoring data for N consecutive days, and historical weather data.

[0137] The second processing unit 305 is used to determine the predicted time taken by the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and to determine the reference time taken by the transport vehicle to pass through the route to be evaluated according to a preset reference speed.

[0138] The first determining unit 306 is used to determine the value of the second safety indicator by using the predicted time consumption and the reference time consumption.

[0139] The second collecting unit 307 is used to collect negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events.

[0140] The third processing unit 308 is used to predict the estimated arrival time of the transport vehicle to the location of the negative event according to the predicted average vehicle speed, and determine the value of the third safety index based on the estimated arrival time and the duration of the negative event.

[0141] In a specific implementation, the third processing unit 308 is used to determine the value of the third safety indicator based on the estimated arrival time and the duration of the negative event. Specifically, for each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable is accumulated by 1, and the value of the designated variable is used to indicate the number of times the negative event is encountered; the value of the third safety indicator is determined based on the value of the designated variable.

[0142] The second determining unit 309 is used to determine the value of the total safety index by combining the value of the first safety index, the value of the second safety index and the value of the third safety index.

[0143] In a specific implementation, the second determining unit 309 is specifically used to calculate the product of the value of the first safety index, the value of the second safety index and the value of the third safety index to obtain the value of the total safety index.

[0144] In an embodiment of the present invention, the value of the first safety index is determined by comparing the vehicle parameters of the transport vehicle and the restriction information of each road section on the route to be evaluated. The average vehicle speed prediction model is used to determine the predicted time required to pass through the route to be evaluated, and the value of the second safety index is determined based on the predicted time and the reference time. The value of the third safety index is determined using the duration of the negative event and the predicted estimated arrival time of the transport vehicle at the location of the negative event. The value of the total safety index is calculated by combining the values ​​of the first safety index, the second safety index, and the third safety index, and the route to be evaluated is safety evaluated by the value of the total safety index to ensure the reliability, timeliness, and efficiency of transportation.

[0145] Preferably, combined Figure 3The content shown, the prediction unit 304 includes: a statistical module, a construction module and a training module; the execution principle of each module is as follows:

[0146] The statistical module is used to count the sample average vehicle speed and sample weather conditions of each road section in each time slice based on the sample data, and to count the sample period information and sample date information corresponding to each time slice.

[0147] The building module is used to build a sample input vector using the sample average vehicle speed, the sample weather condition, the sample time period information and the sample date information.

[0148] The training module is used to train a preset model based on a sample input vector until the preset model converges to obtain an average vehicle speed prediction model, wherein the preset model is constructed based on a Transformer structure.

[0149] Preferably, combined Figure 3 The driving safety assessment system further includes:

[0150] The third determining unit is used to determine the preset numerical interval in which the value of the total safety index is located from multiple preset numerical intervals.

[0151] The output unit is used to output the driving safety assessment level corresponding to the preset value interval in which the value of the total safety index is located, wherein each preset value interval is set with a corresponding driving safety assessment level.

[0152] In summary, an embodiment of the present invention provides a method and system for evaluating driving safety of a dynamic route, which determines the value of a first safety index by comparing the vehicle parameters of the transport vehicle and the restriction information of each road section on the route to be evaluated. The predicted time required to pass through the route to be evaluated is determined using an average vehicle speed prediction model, and the value of a second safety index is determined based on the predicted time and the reference time. The value of a third safety index is determined using the duration of a negative event and the predicted estimated arrival time of the transport vehicle at the location of the negative event. The value of the total safety index is calculated by combining the values ​​of the first safety index, the second safety index, and the third safety index, and the route to be evaluated is safety evaluated using the value of the total safety index to ensure the reliability, timeliness, and efficiency of transportation.

[0153] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can refer to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system or system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can refer to the partial description of the method embodiment. The system and system embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without creative work.

[0154] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0155] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A driving safety assessment method for a dynamic route, characterized in that: The method comprises: Obtain restriction information for each road segment on the route to be evaluated; Comparing the restriction information with vehicle parameters of the transport vehicle, and determining a value of the first safety index according to the comparison result; Collecting the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to a plurality of consecutive time slices at the current moment to form an input vector; Inputting the input vector into a pre-built average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain the predicted average vehicle speed of each road section on the route to be evaluated in multiple future time slices, wherein the average vehicle speed prediction model is obtained based on a preset model trained with sample data, and the sample data at least includes road network data, historical road network speed monitoring data for N consecutive days, and historical weather data; Determining the predicted time taken for the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and determining the reference time taken for the transport vehicle to pass through the route to be evaluated according to a preset reference speed; Determine a value of a second safety indicator by using the predicted time consumption and the reference time consumption; Collecting negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events; Predicting an estimated arrival time of the transport vehicle at the location of the negative event according to the predicted average vehicle speed, and determining a value of a third safety indicator based on the estimated arrival time and the duration of the negative event; The value of the total safety index is determined by combining the value of the first safety index, the value of the second safety index, and the value of the third safety index.

2. The method according to claim 1, characterized in that The process of training the preset model based on sample data to obtain the average vehicle speed prediction model includes: Based on the sample data, the sample average vehicle speed and sample weather conditions of each road section in each time slice are counted, and the sample period information and sample date information corresponding to each time slice are counted; Constructing a sample input vector using the sample average vehicle speed, the sample weather condition, the sample time period information and the sample date information; A preset model is trained based on the sample input vector until the preset model converges to obtain an average vehicle speed prediction model, wherein the preset model is constructed based on a Transformer structure.

3. The method according to claim 1, characterized in that Comparing the restriction information with the vehicle parameters of the transport vehicle, and determining the value of the first safety index according to the comparison result, including: Comparing the restriction information of each road section on the route to be evaluated with the vehicle parameters of the transport vehicle respectively; If the vehicle parameters satisfy the restriction information of each road section on the route to be evaluated, determining the value of the first safety index to be a first preset value; If the vehicle parameters do not satisfy the restriction information of any road section on the route to be evaluated, the value of the first safety index is determined to be a second preset value.

4. The method according to claim 1, characterized in that Determining a value of a third safety indicator based on the estimated arrival time and the duration of the negative event includes: For each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable is accumulated by 1, and the value of the designated variable is used to represent the number of times the negative event is encountered; A value of a third safety indicator is determined based on the value of the designated variable.

5. The method according to claim 1, characterized in that Determining a value of a total safety index by combining a value of the first safety index, a value of the second safety index, and a value of the third safety index includes: The product of the value of the first safety index, the value of the second safety index and the value of the third safety index is calculated to obtain the value of the total safety index.

6. The method according to any one of claims 1 to 5, characterized in that: Also includes: Determining, from a plurality of preset value intervals, a preset value interval in which the value of the total safety index lies; Output the driving safety assessment level corresponding to the preset value interval in which the value of the total safety index is located, wherein each of the preset value intervals is set with a corresponding driving safety assessment level.

7. A driving safety assessment system for a dynamic route, characterized in that: The system comprises: An acquisition unit, used for acquiring restriction information of each road section on the route to be evaluated; A first processing unit, configured to compare the restriction information with vehicle parameters of the transport vehicle, and determine a value of a first safety index according to the comparison result; A first collection unit is used to collect the average vehicle speed, weather conditions, time period information and date information of each road section on the route to be evaluated corresponding to a plurality of consecutive time slices at the current moment to form an input vector; A prediction unit, configured to input the input vector into a pre-built average vehicle speed prediction model to perform vehicle speed prediction, so as to obtain a predicted average vehicle speed for each road section on the route to be evaluated in multiple future time slices, wherein the average vehicle speed prediction model is obtained by training a preset model based on sample data, and the sample data at least includes road network data, historical road network speed monitoring data for N consecutive days, and historical weather data; A second processing unit, configured to determine a predicted time taken for the transport vehicle to pass through the route to be evaluated according to the predicted average vehicle speed, and to determine a reference time taken for the transport vehicle to pass through the route to be evaluated according to a preset reference speed; A first determining unit, configured to determine a value of a second safety indicator by using the predicted time consumption and the reference time consumption; A second collecting unit is used to collect negative events that affect driving safety on the route to be evaluated at the current moment and the duration of the negative events; a third processing unit, configured to predict an estimated arrival time of the transport vehicle at the location of the negative event according to the predicted average vehicle speed, and determine a value of a third safety indicator based on the estimated arrival time and the duration of the negative event; The second determining unit is used to determine a value of a total safety index by combining a value of the first safety index, a value of the second safety index, and a value of the third safety index.

8. The system according to claim 7, characterized in that The prediction unit comprises: The statistical module is used to count the sample average vehicle speed and sample weather conditions of each road section in each time slice based on the sample data, and to count the sample period information and sample date information corresponding to each time slice; A construction module, used to construct a sample input vector using the sample average vehicle speed, the sample weather condition, the sample time period information and the sample date information; A training module is used to train a preset model based on the sample input vector until the preset model converges to obtain an average vehicle speed prediction model, wherein the preset model is constructed based on a Transformer structure.

9. The system according to claim 7, characterized in that The first processing unit is specifically used to: compare the restriction information of each section on the route to be evaluated with the vehicle parameters of the transport vehicle; if the vehicle parameters meet the restriction information of each section on the route to be evaluated, determine the value of the first safety indicator to be a first preset value; if the vehicle parameters do not meet the restriction information of any section on the route to be evaluated, determine the value of the first safety indicator to be a second preset value.

10. The system according to claim 7, characterized in that The third processing unit for determining a value of a third safety indicator based on the estimated arrival time and the duration of the negative event is specifically configured to: For each negative event, if the estimated arrival time of the transport vehicle to the location of the negative event is within the duration of the negative event, the value of the designated variable is accumulated by 1, and the value of the designated variable is used to represent the number of times the negative event is encountered; A value of a third safety indicator is determined based on the value of the designated variable.

Images