A method and system for unmanned adaptive tolling for highways

Abstract

The application discloses an unmanned self-adaptive toll collection method and system for expressways, relates to the technical field of self-adaptive toll collection, and solves the technical problems that it is difficult to open toll channels in time according to the congestion condition of toll stations and it is difficult to adaptively collect tolls according to the potential risks brought by drivers to roads in the expressway toll collection system; the application acquires the number of vehicles in the toll station and target data in real time; time periods are divided, and a congestion factor of the next time period is calculated according to the number of vehicles in the current time period; the number of toll channels opened in the next time period is dynamically adjusted based on the congestion factor; after the time reaches the next time period, vehicles are distributed to corresponding toll channels through an artificial intelligence model; the road influence factor of each vehicle is determined based on the target data, and the expressway toll is calculated; the application can improve the passing efficiency of vehicles at the toll station, has a prompting effect on safe driving of drivers, and improves traffic safety.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of traffic management, and relates to adaptive tolling technology, in particular to an unmanned adaptive tolling method and system for expressways. BACKGROUND

[0002] In order to improve the efficiency of tolling, reduce labor costs, and improve vehicle passing speed. With the development of technology, especially the progress of automation, intelligent identification and information processing technology, unmanned tolling systems have emerged. This system can automatically identify vehicle information, realize fast charging and fee deduction, thereby greatly shortening the time of vehicle staying at the toll station and improving the overall traffic efficiency.

[0003] At present, most of the unmanned adaptive tolling methods and systems for expressways are difficult to adapt to the potential risks brought by drivers in the process of highway tolling, and only charge in a fixed mode according to the mileage of the car, ignoring the beneficial effect of indirectly prompting drivers to drive safely according to the adaptive tolling of the driving behavior of the driver.

[0004] Therefore, the present application discloses an unmanned adaptive tolling method and system for expressways to solve the above technical problems. SUMMARY

[0005] The present application aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present application proposes an unmanned adaptive tolling method and system for expressways to solve the technical problems that it is difficult to open the toll channel in time according to the congestion of the toll station and difficult to adapt to the potential risks brought by drivers in the process of highway tolling. The present application solves the above problems by acquiring the number of vehicles in the toll station and target data in real time; dividing the time period, and calculating the congestion factor of the next time period according to the number of vehicles in the current time period; dynamically adjusting the number of toll channels opened in the next time period based on the congestion factor; after the time arrives at the next time period, the vehicles are allocated to the corresponding toll channel through the artificial intelligence model; determining the road influence factor of each vehicle based on the target data and calculating the expressway fee.

[0006] To achieve the above purpose, the first aspect of the present application provides an unmanned adaptive tolling method for expressways, comprising:

[0007] Acquiring the number of vehicles within the management range of the toll station and the target data of each vehicle; wherein the target data includes the weight of the car, the speed, the number of brake times and the number of times of illegal driving;

[0008] Dividing the time period, determining the congestion factor of the toll station in the next time period based on the number of vehicles in the current time period;

[0009] Setting toll lanes opened in the next time period based on the congestion factor;

[0010] When the time reaches the next time period, the vehicles are guided to the toll lanes based on the set toll lanes by an artificial intelligence model;

[0011] Determining the road impact factor of each vehicle based on the target data;

[0012] Determining the highway fee of each vehicle based on the road impact factor;

[0013] Collecting the highway fee through the CPC card or the ETC card.

[0014] Preferably, the number of vehicles within the toll station management range and the target data of each vehicle are obtained, including:

[0015] When the vehicle enters the toll station management range, the number of current vehicles is obtained through video monitoring; when the vehicle drives to the area where the weighing device is located, the vehicle weight is obtained; wherein the management range is obtained by manual setting;

[0016] The number of times of braking of the vehicle on the highway this time is obtained through the vehicle-mounted sensor, the number of times of illegal driving of the vehicle on the highway this time and the speed at several time points are obtained through video monitoring technology, and the speed V of the vehicle on the highway this time is determined based on the mode, average value and maximum value of the speed;

[0017] The speed V satisfies the following formula:

[0018] V = α1 × DD + α2 × ZD + α3 × PD;

[0019] Wherein, DD is the maximum value of the speed at several time points, ZD is the mode of the speed at several time points, and PD is the average value of the speed at several time points; α1, α2 and α3 are all proportion adjustment coefficients set by manual, and α1 + α2 + α3 = 1, α1 > α2 > α3.

[0020] Preferably, the time period is divided, including:

[0021] When the time reaches 0:00 of the day, the vehicle flow at each time point of the date of the day in the history is obtained, the vehicle flow is divided into several vehicle flow groups according to the time points, and the mode of each vehicle flow group is marked as the characteristic vehicle flow of the corresponding time point; wherein the vehicle flow is the vehicle flow within the current toll station management range;

[0022] The time point with the characteristic vehicle flow greater than the vehicle flow threshold value is marked as a target time point, the continuous target time points are connected into a target time period, and the continuous non-target time points are connected into a non-target time period; wherein the vehicle flow threshold value is obtained by manual setting;

[0023] Mark the target time period or non-target time period with a time length greater than the defined time length as a management time period, and incorporate the target time period or non-target time period with a time length not greater than the defined time length into the management time period with a smaller time length among the adjacent two management time periods; divide a day into a plurality of time periods according to the time range in which each management time period is located; wherein the defined time length is obtained by manual setting.

[0024] Preferably, the congestion factor of the toll station in the next time period is determined based on the number of vehicles in the current time period, comprising:

[0025] When the time reaches the last time point of the current time period, the number of vehicles in the current time period in the historical days is obtained, and the number of vehicles in the next time period in the historical days is obtained;

[0026] The number of vehicles in the current time period in the historical days is integrated into a current time period number group, and the number of vehicles in the next time period in the historical days is integrated into a next time period number group;

[0027] The number YL in the next time period number group corresponding to each date in the historical days is extracted in sequence i and the number DL in the current time period number group i , the difference value C i of the number YL i minus the number DL i is obtained in sequence; the change ratio H i of the difference value C i between adjacent dates is determined based on the difference value C i ; wherein i is the number of each date in the historical days;

[0028] The change ratio H i satisfies the following formula:

[0029] H i =(C i -C i-1 ) / C i-1 ;

[0030] The average value P i of the change ratio H i is obtained, and the change ratio DH of the current day is determined based on the average value P i ; the change ratio DH satisfies the following formula:

[0031] DH=β×ln(P i +1)+1;

[0032] Wherein, β is an artificially set amplitude adjustment coefficient, and the value range of β is (0, 1];

[0033] Based on the proportion DH, the number YC of vehicles in the next time period on the current day is determined, and the number YC satisfies the following formula:

[0034] YC=(1+DH)×DC;

[0035] Wherein, DC is the number of vehicles in the current time period on the current day;

[0036] Based on the number YC, the congestion factor YZ in the next time period on the current day is determined, and the congestion factor YZ satisfies the following formula:

[0037] YZ=BZ×YC / BYC;

[0038] Wherein, BYC is an artificially set standard number, and BZ is a standard congestion factor corresponding to the standard number.

[0039] Preferably, the method further comprises:

[0040] When the congestion factor YZ is greater than the congestion threshold one, all toll channels of the current toll station are opened; wherein, the congestion threshold one and the congestion threshold two are obtained by artificial setting, and the congestion threshold one is greater than the congestion threshold two;

[0041] When the congestion factor is not greater than the congestion threshold one and not less than the congestion threshold two, the number TDL of toll channels of the current toll station is extracted, and the number KDL of open toll channels of the current toll station is determined based on the formula KDL= TDL / 2 The number KDL of open toll channels of the current toll station is determined, and the toll channels of the current toll station are opened from left to right based on the number KDL;

[0042] When the congestion factor is less than the congestion threshold two, the leftmost toll channel of the current toll station is opened.

[0043] Preferably, the method further comprises:

[0044] The open toll channels and the real-time number of vehicles in the corresponding time period in the historical reference data are extracted, and the number of each vehicle guided to pass through the toll channel is extracted; wherein, the historical reference data includes a plurality of open toll channels and the real-time number of vehicles in the corresponding time period, and the number of each vehicle guided to pass through the toll channel according to the open toll channels and the real-time number of vehicles in the corresponding time period;

[0045] The open toll channel and the real-time vehicle quantity in the corresponding time period are integrated into several groups of training data and test data, each vehicle is guided to pass through the toll channel; the artificial intelligence model is trained using the training data, the trained artificial intelligence model is tested using the test data, and the artificial intelligence model is adjusted according to the test result; and finally a channel guidance model is obtained, in which the input is the open toll channel and the real-time vehicle quantity, and the output is the number of the toll channel through which each vehicle is guided to pass.

[0046] When the time reaches the next time period, the open toll channel and the real-time vehicle quantity are input into the channel guidance model to obtain the number of the toll channel through which each vehicle is guided to pass.

[0047] Preferably, the method further comprises:

[0048] The vehicle weight CZ, speed V, brake times BC and illegal driving times WG of the current vehicle are extracted; when the speed V exceeds the speed threshold VZ or the illegal driving times WG exceeds the illegal times threshold WGZ, the road influence factor DZ is set to the maximum value of the influence factor limit range; wherein the speed threshold, the illegal times threshold and the influence factor limit range are obtained by manual setting;

[0049] When the speed V exceeds the speed threshold and the illegal driving times WG does not exceed the illegal times threshold, the road influence factor DZ of the current vehicle is determined based on the vehicle weight CZ, the speed V, the brake times BC and the illegal driving times WG; the road influence factor DZ satisfies the following formula:

[0050] ;

[0051] Wherein, BCZ is a standard vehicle weight set by manual, and BCL is a standard brake times set by manual; 、 、 and are all proportional adjustment coefficients set by manual, and , ;

[0052] When the road influence factor DZ is greater than the maximum value of the influence factor limit range, the road influence factor DZ is set to the maximum value of the influence factor limit range; when the road influence factor DZ is less than the minimum value of the influence factor limit range, the road influence factor DZ is set to the minimum value of the influence factor limit range.

[0053] Preferably, the method further comprises:

[0054] Obtain the driving mileage LC of the current vehicle in this high-speed driving, extract the road influence factor DZ, determine the high-speed fee GY of the current vehicle based on the road influence factor DZ and the driving mileage LC; the high-speed fee GY satisfies the following formula:

[0055] GY=ZY×LC×DZ / YZ;

[0056] Wherein, ZY is the reference rate of one kilometer set by artificial, YZ is the middle value of the influence factor limited range.

[0057] Preferably, the high-speed fee is collected through the CPC card or the ETC card, comprising:

[0058] When the vehicle drives to the CPC card charging place or the ETC card charging place of the toll station, the high-speed fee of this driving is paid through the CPC card or the ETC card.

[0059] The second aspect of the application provides an unmanned adaptive toll system for expressway, comprising: a field scheduling module, and a data collection module and an adaptive toll module connected with the field scheduling module;

[0060] The data collection module is used for obtaining the number of vehicles in the management range of the toll station and target data of each vehicle; wherein, the target data includes the weight of the car, the speed, the number of brake times and the number of times of illegal driving;

[0061] The field scheduling module is used for dividing time periods, determining the congestion factor of the toll station in the next time period based on the number of vehicles in the current time period, setting the toll channel of the toll station opened in the next time period based on the congestion factor, and guiding the vehicles to each toll channel through the artificial intelligence model based on the set toll channel when the time reaches the next time period.

[0062] The adaptive toll module is used for determining the road influence factor of each vehicle based on the target data, determining the high-speed fee of each vehicle based on the road influence factor, and collecting the high-speed fee through the CPC card or the ETC card.

[0063] Compared with the prior art, the application has the beneficial effects that:

[0064] 1.The present application obtains the number of vehicles in the toll station and target data in real time; divides the time period, and calculates the congestion factor of the next time period according to the number of vehicles in the current time period; dynamically adjusts the number of toll channels opened in the next time period based on the congestion factor; after the time reaches the next time period, the vehicles are allocated to the corresponding toll channel through the artificial intelligence model; the road impact factor of each vehicle is determined based on the target data and the highway fee is calculated, which solves the technical problems that it is difficult to open the toll channel in time according to the congestion of the toll station in the highway toll system, and it is difficult to adaptively charge according to the potential risk of the road brought by the driver; the present application can improve the passing efficiency of vehicles in the toll station, and has a prompting effect on safe driving of the driver, and improves the traffic safety.

[0065] 2.The road impact factor calculated by the present application is a factor for calculating the damage and potential traffic hazards to the road caused by the current vehicle in this driving, which is used for subsequent settlement of the highway fee of the current vehicle, and the weight of the vehicle and the driver's operation are combined to increase the highway fee of the vehicle with great potential traffic hazards and reduce the highway fee of the vehicle without potential traffic hazards, so as to realize the principle of 'high-risk pay more' through accurate measurement and establish a differentiated charging system; this can urge the driver to comply with the driving safety rules more, so that the charging system has a certain driving prompting effect and increases the adaptability of the charging system. BRIEF DESCRIPTION OF DRAWINGS

[0066] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0067] Figure 1 The operation steps of the present application are shown in the figure;

[0068] Figure 2 The operation steps of the present application to obtain the congestion factor of the next time period are shown in the figure;

[0069] Figure 3 The system module of the present application is shown in the figure. DETAILED DESCRIPTION

[0070] The technical solutions of the present application will be described in detail below in conjunction with the embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0071] Referring to Figure 1 The first aspect of the present application provides an unmanned adaptive toll collection method for expressway, comprising:

[0072] Obtaining the number of vehicles within the management range of the toll station and target data of each vehicle; wherein the target data includes vehicle weight, speed, brake times and illegal driving times;

[0073] Dividing time periods, determining the congestion factor of the toll station in the next time period based on the number of vehicles in the current time period, and setting the toll channels of the toll station to be opened in the next time period based on the congestion factor; when the time reaches the next time period, guiding the vehicles to the toll channels through an artificial intelligence model based on the set toll channels;

[0074] Determining the road influence factor of each vehicle based on the target data, determining the expressway fee of each vehicle based on the road influence factor, and collecting the expressway fee through a CPC card or an ETC card.

[0075] In the present application, the number of vehicles within the management range of the toll station and the target data of each vehicle are obtained, comprising:

[0076] When a vehicle enters the management range of the toll station, the number of the current vehicles is obtained through video monitoring; when the vehicle drives to the area where the weight measuring device is located, the vehicle weight is obtained; wherein the management range is obtained by manual setting;

[0077] The brake times of the vehicle on the expressway this time are obtained through the vehicle-mounted sensor, the illegal driving times and the speed of the vehicle on the expressway this time are obtained through video monitoring technology, and the speed V of the vehicle on the expressway this time is determined based on the mode, average value and maximum value of the speed;

[0078] The speed V satisfies the following formula:

[0079] V = α1 × DD + α2 × ZD + α3 × PD;

[0080] Wherein, DD is the maximum value of the speed at several time points, ZD is the mode of the speed at several time points, and PD is the average value of the speed at several time points; α1, α2 and α3 are all manually set proportional adjustment coefficients, and α1 + α2 + α3 = 1, α1 > α2 > α3.

[0081] It should be noted that the illegal driving times are the sum of the overspeed times, the continuous lane changing times and the overtaking times in low speed lane.

[0082] It should be noted that the proportional adjustment coefficient α1> α2> α3 is because: α1 is multiplied by the maximum value of the speed at several times, and α2 is multiplied by the mode of the speed at several times, and α3 is multiplied by the average value of the speed at several times;

[0083] Among the hazards caused by highway vehicle driving, the higher the speed at a certain time, the greater the potential hazard at this time, and the time at the maximum speed is instantaneous, but such hazard is the greatest, therefore, the maximum proportional adjustment coefficient corresponding to the maximum value of the speed at several times is set in the application;

[0084] In addition, the mode of the speed at several times can better represent the speed of the vehicle occupying the longest time in this driving, and better represent the characteristic situation of the vehicle speed, therefore, the proportional adjustment coefficient of the mode of the speed at several times is greater than the proportional adjustment coefficient of the average value of the speed at several times.

[0085] The time period is divided in the application, including:

[0086] When the time reaches 0:00 of a day, the vehicle flow of each time point of the date of the day in the history is obtained, the vehicle flow is divided into several vehicle flow groups according to the time points, and the mode of each vehicle flow group is marked as the characteristic vehicle flow of the corresponding time point; wherein the vehicle flow is the vehicle flow in the current toll station management range;

[0087] The time point with the characteristic vehicle flow greater than the vehicle flow threshold is marked as the target time point, the continuous target time points are connected into a target time period, and the continuous non-target time points are connected into a non-target time period; wherein the vehicle flow threshold is obtained by manual setting;

[0088] The target time period or non-target time period with a time length greater than the defined time length is marked as a management time period, and the target time period or non-target time period with a time length not greater than the defined time length is integrated into the management time period with smaller time length among the adjacent two management time periods; a day is divided into several time periods according to the time range of each management time period; wherein the defined time length is obtained by manual setting.

[0089] It should be noted that the unit of time point in the application can be 1 minute, 5 minutes or 10 minutes.

[0090] It should be noted that the management range is obtained by manual setting, for example, the management range can be within 500 meters before and after the toll station.

[0091] It should be noted that the vehicle flow threshold is a threshold for judging whether the vehicle flow in the management range of the toll station within a unit time is saturated, and the vehicle flow threshold can be 100 vehicles.

[0092] It should be noted that the non-target time point is a time point at which the feature vehicle flow is less than or equal to the vehicle flow threshold.

[0093] It should be noted that the defined duration can be 2 hours, 4 hours, 6 hours or 8 hours.

[0094] Please refer to Figure 2 In the present application, the congestion factor of the toll station in the next time period is determined based on the number of vehicles in the current time period, including:

[0095] When the time reaches the last time point of the current time period, the number of vehicles in the current time period in the historical several days is obtained, and the number of vehicles in the next time period in the historical several days is obtained;

[0096] The number of vehicles in the current time period in the historical several days is integrated into a current time period number group; the number of vehicles in the next time period in the historical several days is integrated into a next time period number group;

[0097] The number YL i in the next time period number group corresponding to each date in the historical several days is extracted in turn i , and the number DL i in the current time period number group is extracted in turn i , the difference C i between the number YL i and the number DL i is obtained in turn; the change ratio H i of the difference C i between adjacent dates is determined based on the difference C ; wherein i is the number of each date in the historical several days;

[0098] The change ratio H i satisfies the following formula:

[0099] H i =(C i-1 -C i-1 ) / C i ;

[0100] The average value P i of the change ratio H i is obtained, and the change ratio DH of the day is determined based on the average value P ; the change ratio DH satisfies the following formula:

[0101] DH=β×ln(P i +1)+1;

[0102] Wherein, β is an artificially set amplitude adjustment coefficient, and the value range of β is (0, 1];

[0103] The number YC of vehicles in the next time period on the current day is determined based on the proportion DH, and the number YC satisfies the following formula:

[0104] YC=(1+DH)×DC;

[0105] Wherein, DC is the number of vehicles in the current time period on the current day;

[0106] The congestion factor YZ in the next time period on the current day is determined based on the number YC, and the congestion factor YZ satisfies the following formula:

[0107] YZ=BZ×YC / BYC;

[0108] Wherein, BYC is a standard number set artificially, and BZ is a standard congestion factor corresponding to the standard number set artificially.

[0109] It is worth noting that in modern traffic management, the passing efficiency of the high-speed toll station directly affects the regional traffic fluency, user satisfaction and overall logistics efficiency; by dynamically predicting the congestion of the toll station in the next time period, and adjusting the number of open lanes according to the traffic scale in the peak period or sudden situation, the number of lanes and the traffic demand can be accurately matched; for example: increasing the number of lanes in the peak period can effectively reduce the queuing time of vehicles; reducing the number of lanes in the low peak period can avoid resource waste. This flexibility can not only reduce the waste of lane resources, but also reduce the traffic delay caused by congestion, and improve the overall traffic efficiency.

[0110] It should be noted that the historical several days in the application are several days close to the current date.

[0111] It should be noted that the number of vehicles in the current time period in the historical several days is integrated into the current time period number group, for example: if the current time period is 15:00-19:00 and the historical days are 100 days, the number of vehicles in 15:00-19:00 of each day in the 100 days is integrated into the current time period number group as a data, so there are 100 numbers of vehicles in 15:00-19:00 in the current time period number group.

[0112] It should be noted that the amplitude adjustment coefficient β of the ln() function is used to adjust the influence degree of the average value P i of the change proportion H i on the change proportion DH; when other conditions are unchanged, the greater β, the greater the influence of the change proportion DH, and the smaller β, the smaller the influence of the change proportion DH.

[0113] It should be noted that in another embodiment, the congestion factor YZ can also be obtained by the following steps:

[0114] when the time reaches the last time point of the current time period, obtaining the number of vehicles in the current time period in the historical several days, and obtaining the number of vehicles in the next time period in the historical several days;

[0115] integrating the number of vehicles in the current time period in the historical several days into a current time period number group, and integrating the number of vehicles in the next time period in the historical several days into a next time period number group;

[0116] extracting the current time period number group and the next time period number group in turn, marking the extracted current time period number group or next time period number group as an analysis number group, obtaining the variance of the data in the analysis number group, and determining whether the variance is less than the corresponding variance limit value; if yes, calculating the characteristic number by averaging the data in the analysis number group; if no, removing the number that is most different from the number mode in the analysis number group, and re-determining the variance until the variance of the analysis number group is less than the variance limit value, and then calculating the characteristic number by averaging the remaining data in the analysis number group;

[0117] when the analysis number group is the current time period number group, marking the corresponding characteristic number as a characteristic number DTL; when the analysis number group is the next time period number group, marking the corresponding characteristic number as a characteristic number YTL; and determining a congestion factor YZ of the current day in the next time period based on the characteristic number DTL and the characteristic number YTL; the congestion factor YZ satisfies the following formula:

[0118] YZ=BZ×(DC×YTL / DTL) / BYC;

[0119] wherein, BYC is a standard number set by an artificial, BZ is a standard congestion factor corresponding to the standard number set by an artificial, and DC is the number of vehicles in the current time period of the current day.

[0120] In the present application, the toll lanes of the toll station in the next time period are opened based on the congestion factor, comprising:

[0121] extracting the congestion factor YZ; when the congestion factor YZ is greater than a congestion threshold value one, opening all the toll lanes of the current toll station; wherein, the congestion threshold value one and the congestion threshold value two are obtained by artificial setting, and the congestion threshold value one is greater than the congestion threshold value two;

[0122] when the congestion factor is not greater than the congestion threshold value one and not less than the congestion threshold value two, extracting the number TDL of the toll lanes of the current toll station, and determining the number KDL of the open toll lanes of the current toll station based on the formula KDL= TDL / 2 determining the number KDL of the open toll lanes of the current toll station, and opening the toll lanes of the current toll station from left to right based on the number KDL;

[0123] When the congestion factor is less than the congestion threshold two, the leftmost toll lane in the toll lanes of the current toll station is opened.

[0124] It should be noted that the number of opened toll lanes obtained by analysis in the present application is the number of one-way lanes.

[0125] It should be noted that in the present application, the toll lanes of the current toll station are opened from left to right, and the leftmost toll lane in the toll lanes of the current toll station is opened, which is mainly based on the left lane. This setting is because the leftmost lane generally overlaps less with the path of vehicles traveling straight on the main road, avoiding accidents caused by frequent lane changes.

[0126] In the present application, the vehicle is guided to each toll lane based on the set toll lane by an artificial intelligence model, including:

[0127] The opened toll lanes and the number of real-time vehicles in the corresponding time period in the historical reference data, and the number of each vehicle guided to pass through the toll lane are extracted; wherein the historical reference data includes a plurality of opened toll lanes and the number of real-time vehicles in the corresponding time period, and the number of each vehicle guided to pass through the toll lane by experts according to the number of real-time vehicles in the corresponding time period in the toll lane;

[0128] The opened toll lanes and the number of real-time vehicles in the corresponding time period, and the number of each vehicle guided to pass through the toll lane are integrated into a plurality of groups of training data and test data; the training data is used to train the artificial intelligence model, and the test data is used to test the trained artificial intelligence model, and the artificial intelligence model is adjusted according to the test result; Finally, the channel guidance model is obtained, which inputs the opened toll lanes and the number of real-time vehicles, and outputs the number of each vehicle guided to pass through the toll lane;

[0129] When the time reaches the next time period, the opened toll lanes and the number of real-time vehicles are input into the channel guidance model to obtain the number of each vehicle guided to pass through the toll lane.

[0130] It should be noted that the number of toll channels through which the vehicle passes used in data training is set by the expert according to the open toll channels and the real-time number of vehicles in the corresponding time period; when the channel guidance model is built, the number of the toll channel through which each vehicle is guided to pass can be: when the real-time number of vehicles is less than the vehicle threshold, the toll channel through which each vehicle is guided to pass is set according to the distance of the vehicle from each open channel; for example: vehicle 1 is close to toll channel 1, so vehicle 1 is guided to pass through toll channel 1; when the real-time number of vehicles is not less than the vehicle threshold, the toll channel through which each vehicle is guided to pass is set according to the order in which the vehicle enters the toll station management range; for example: 6 toll channels are opened, and 6 vehicles are taken as a group, and the toll channel through which each vehicle is guided to pass is selected from left to right according to the order in which each group of vehicles enters the toll station; wherein the vehicle threshold is obtained by manual setting.

[0131] Specifically, the trained artificial intelligence model is tested using test data, and the specific steps of adjusting the artificial intelligence model according to the test result are as follows:

[0132] The open toll channels and the real-time number of vehicles in the corresponding time period in the test data are input into the trained artificial intelligence model to obtain the corresponding number of the toll channel through which each vehicle is guided to pass. The number of the toll channel through which each vehicle is guided to pass is compared with the corresponding number of the toll channel through which each vehicle is guided to pass in the test data. When the two numbers are the same, no parameter adjustment is needed, and the next group of test data is tested; if the two numbers are different, the corresponding parameters are adjusted until the two numbers are the same, and the next group of test data is tested. When the number of test data with the same number obtained by all test data accounts for 90% or more of the total amount of test data, the channel guidance model with the input of open toll channels and real-time number of vehicles and the output of the number of the toll channel through which each vehicle is guided to pass is obtained.

[0133] In this application, the road influence factor of each vehicle is determined based on target data, which includes:

[0134] The current vehicle's car weight CZ, speed V, brake times BC, and illegal driving times WG are extracted; when the speed V exceeds the speed threshold VZ or the illegal driving times WG exceeds the illegal times threshold WGZ, the road influence factor DZ is set to the maximum value of the influence factor limit range; wherein the speed threshold, the illegal times threshold, and the influence factor limit range are obtained by manual setting;

[0135] When the speed V exceeds the speed threshold and the number of times of illegal driving WG does not exceed the number of times of illegal driving threshold, the road influence factor DZ of the current vehicle is determined based on the vehicle weight CZ, the speed V, the number of times of braking BC and the number of times of illegal driving WG; the road influence factor DZ satisfies the following formula:

[0136] ;

[0137] Wherein, BCZ is the standard vehicle weight artificially set, and BCL is the standard number of times of braking artificially set; 、 、 and are all proportional adjustment coefficients artificially set, and , ;

[0138] When the road influence factor DZ is greater than the maximum value of the influence factor limiting range, the road influence factor DZ is taken as the maximum value of the influence factor limiting range; when the road influence factor DZ is less than the minimum value of the influence factor limiting range, the road influence factor DZ is taken as the minimum value of the influence factor limiting range.

[0139] It is worth noting that the road influence factor calculated by the present application is a factor for calculating the damage and potential traffic hazards caused by the current vehicle to the road in this driving, which is used for subsequent settlement of high-speed charges of the current vehicle, and the weight of the vehicle and the operation of the driver are combined to increase the high-speed charges of the vehicle with greater potential traffic hazards and reduce the high-speed charges of the vehicle without potential traffic hazards, so as to realize the principle of 'high-risk pay more' through accurate measurement and establish a differentiated charging system. In this way, the driver can be urged to comply with the driving safety rules more, so that the charging system has a certain driving urging effect and increases the adaptability of the charging system.

[0140] It is worth noting that the present application considers the vehicle weight, speed, number of times of braking and number of times of illegal driving when calculating the influence of the vehicle on other vehicles and the road itself;

[0141] The vehicle weight is considered because the vehicle weight directly affects the degree of wear on the road surface, and heavy vehicles can accelerate the damage of road collapse, cracks and other damage, and can have a long-term impact on infrastructure;

[0142] The speed is considered because the higher the speed, the greater the kinetic energy of the vehicle, the higher the load that the braking system needs to bear, and the braking distance is significantly increased, and the potential hazards to other vehicles are also greater;

[0143] The number of times of braking is considered because braking can increase the friction between the tire and the road surface, causing damage to the road surface;

[0144] The number of times of illegal driving is considered because the number of times of illegal driving will destroy the order of road traffic, cause congestion or "broken window effect", and increase the risk of accidents.

[0145] It should be noted that in the present application, the speed threshold can be set to 120km / h, and the threshold of the number of times of illegal driving can be set to 3 times.

[0146] It should be noted that the range of the influence factor is used to limit the range of the rationality of the road influence factor DZ to avoid the high or low of the subsequent calculation of the high-speed fee.

[0147] It should be noted that the proportional adjustment coefficient Because: The related data of the number of times of illegal driving WG is affected, The related data of the weight of the car CZ is multiplied, The related data of the speed V is multiplied, The related data of the number of times of braking BC is multiplied; because in high-speed driving, the number of times of illegal driving WG records the number of times of directly violating traffic regulations, which is the most direct data reflecting the potential risk of the current vehicle, therefore, the proportional adjustment coefficient Corresponding to the number of times of illegal driving WG is set to be the largest; the weight of the car CZ is the main data that can affect the highway, therefore, the proportional adjustment coefficient Corresponding to the weight of the car CZ is set to be the second largest in descending order; the speed V is the core index of driving behavior, which directly affects safety and accident probability, but its influence can be offset by other factors, such as driving experience, therefore, the proportional adjustment coefficient Corresponding to the speed V is set to be the third largest in descending order; the brake will cause the friction between the tire and the road surface to increase, causing the road surface to be damaged, but if the proportional adjustment coefficient Corresponding to the number of times of braking BC is too large, it may cause the safety hidden danger that the driver is unwilling to step on the brake, therefore, the proportional adjustment coefficient Corresponding to the number of times of braking BC is set to be the smallest.

[0148] It should be noted that when the road impact factor DZ is taken as the maximum value of the impact factor limit range, the car weight CZ and the brake times BC are not limited, which is because: the car weight CZ is generally screened at the toll station of the highway entrance, and the car weight CZ generally does not exceed the bearing capacity of the highway itself, so the car weight CZ does not need to be further limited; the brake can reduce the speed to ensure the safety of the driver, and if the brake times BC are limited, it may cause the driver to be unwilling to brake and cause safety hazards; at the same time, the application analyzes the brake times BC in the subsequent, and the value of the proportion adjustment coefficient set for the brake times BC is very small, for example, the value range is [0, 0.1]; this is to avoid the safety hazards caused by the driver's unwillingness to brake due to the statistics of the brake times BC.

[0149] It should be noted that in the present application, the standard car weight BCZ can be set to 49 tons, and the standard brake times BCL are obtained by analyzing the length of the route, and the standard brake times are proportional to the length of the route, for example, when the length of the route is 100 kilometers, the standard brake times are 20 times, and when the length of the route is 200 kilometers, the standard brake times are 40 times.

[0150] In the present application, the highway fee of each vehicle is determined based on the road impact factor, including:

[0151] The driving mileage LC of the current vehicle in the current highway driving is obtained, the road impact factor DZ is extracted, and the highway fee GY of the current vehicle is determined based on the road impact factor DZ and the driving mileage LC; the highway fee GY satisfies the following formula:

[0152] GY=ZY×LC×DZ / YZ;

[0153] Wherein, ZY is a manually set reference rate of one kilometer, and YZ is the middle value of the impact factor limit range.

[0154] In the present application, the highway fee is collected through the CPC card or the ETC card, including:

[0155] When the vehicle drives to the CPC card charging place or the ETC card charging place of the toll station, the highway fee of the current driving is paid through the CPC card or the ETC card.

[0156] Please refer to Figure 3 The second aspect of the present application provides an unmanned adaptive toll system for a highway, including: a field dispatching module, and a data collection module and an adaptive toll module connected with the field dispatching module;

[0157] The data collection module is configured to obtain the number of vehicles within the management range of the toll station and target data of each vehicle, wherein the target data comprises vehicle weight, speed, brake times and illegal driving times.

[0158] The field scheduling module is configured to divide time periods, determine a congestion factor of the toll station in the next time period based on the number of vehicles in the current time period, set toll channels of the toll station to be opened in the next time period based on the congestion factor, and guide vehicles to the toll channels based on the set toll channels by using an artificial intelligence model when the time reaches the next time period.

[0159] The adaptive toll module is configured to determine a road influence factor of each vehicle based on the target data, determine a highway fee of each vehicle based on the road influence factor, and collect the highway fee through a CPC card or an ETC card.

[0160] Some data in the above formula is calculated by removing the dimension and taking the numerical value, the formula is obtained by simulating a large amount of collected data to obtain a formula closest to the real situation, and the preset parameters and the preset threshold in the formula are set by a person skilled in the art according to the actual situation or obtained by simulating a large amount of data.

[0161] Working principle of the present application:

[0162] The number of vehicles within the management range of the toll station and the target data of each vehicle are obtained, which provides data support for subsequent analysis; time periods are divided, which provides personalized time periods for subsequent analysis; a congestion factor of the toll station in the next time period is determined based on the number of vehicles in the current time period, which calculates the congestion of the toll station in the next time period and provides a theoretical basis for the opening of the toll channels in the subsequent step; the toll channels of the toll station to be opened in the next time period are set based on the congestion factor; when the time reaches the next time period, vehicles are guided to the toll channels based on the set toll channels by using an artificial intelligence model.

[0163] A road influence factor of each vehicle is determined based on the target data, which provides a dynamic road influence factor by analyzing the driving behavior of the driver, and the highway fee in the subsequent step can be reasonably changed by using the road influence factor; a highway fee of each vehicle is determined based on the road influence factor; and the highway fee is collected through a CPC card or an ETC card.

[0164] The above embodiments are only used to illustrate the technical method of the present application and not to limit it, although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical method of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical method of the present application.

Claims

1. A method for unmanned adaptive toll collection on highways, characterized in that, include: Obtain the number of vehicles within the toll station's management area and target data for each vehicle; the target data includes vehicle weight, speed, number of braking attempts, and number of traffic violations. Divide the time period and determine the congestion factor of the toll station in the next time period based on the number of vehicles in the current time period; The toll lanes that will be open at the toll station in the next time period will be determined based on congestion factors; When the next time period arrives, vehicles will be guided to the designated toll lanes using an artificial intelligence model. Determine the road impact factors for each vehicle based on the target data; The highway toll for each vehicle is determined based on road impact factors. Tolls are collected via CPC card or ETC card. The method of determining the congestion factor of the toll station for the next time period based on the number of vehicles in the current time period includes: When the time reaches the last point of the current time period, obtain the number of vehicles in the current time period over a certain number of historical days, and obtain the number of vehicles in the next time period over a certain number of historical days. The number of vehicles in the current time period is consolidated into a current time period group based on the number of vehicles in the previous several days; the number of vehicles in the next time period is consolidated into a next time period group based on the number of vehicles in the previous several days. Extract the quantity YL from the next time period corresponding to each date in the historical data over several days. i and the quantity DL in the current time period group i Quantify the number of times the same number i is obtained in sequence. i Subtract quantity DL i The difference C i Based on the difference C i Determine the difference C between adjacent dates i The change ratio H i Where i is the number of each date in a historical period; The change ratio H i Satisfy the following formula: H i =(C i -C i-1 ) / |C i-1 |; Obtain several change ratios H i The average value P i Based on the average value P i Determine the daily change ratio DH; the change ratio DH satisfies the following formula: DH=β×ln(P i +1)+1; Where β is the amplitude adjustment coefficient, and the value range of β is (0,1]; The number of vehicles YC in the next time period is determined based on the ratio DH, and the number YC satisfies the following formula: YC = (1 + DH) × DC; Where DC represents the number of vehicles on that day within the current time period; The congestion factor YZ for the next time period is determined based on the quantity YC, and the congestion factor YZ satisfies the following formula: YZ = BZ × YC / BYC; Where BYC is the standard quantity and BZ is the standard congestion factor corresponding to the standard quantity.

2. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The acquisition of the number of vehicles within the toll station's management area and the target data for each vehicle includes: When a vehicle enters the toll station's management area, the number of vehicles is obtained through video monitoring; when a vehicle travels to the area where the weighing equipment is located, the vehicle's weight is obtained. The number of times the vehicle braked on the highway was obtained by the vehicle-mounted sensors, and the number of times the vehicle violated traffic rules and its speed at several moments were obtained by video monitoring technology. The speed V of the vehicle on the highway was determined based on the mode, average and maximum of the speed. The speed V satisfies the following formula: V = α1 × DD + α2 × ZD + α3 × PD; Where DD is the maximum speed at several moments, ZD is the mode of speed at several moments, and PD is the average speed at several moments; α1, α2 and α3 are all proportional adjustment coefficients, and α1+α2+α3=1, α1>α2>α3.

3. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The time periods are divided into: When the time reaches 0:00 of the day, the traffic flow at each time point of the current date in the historical data is obtained. The traffic flow is divided into several traffic flow groups according to the time points, and the mode of each traffic flow group is marked as the characteristic traffic flow of the corresponding time point; wherein, the traffic flow is the traffic flow within the current toll station management area. The time points when the characteristic traffic flow is greater than the traffic flow threshold are marked as target time points, and consecutive target time points are connected into target time periods, and consecutive non-target time points are connected into non-target time periods. Target time periods or non-target time periods with a duration longer than the defined duration are marked as management time periods. Target time periods or non-target time periods with a duration shorter than the defined duration are merged into the management time period with the shorter duration among two adjacent management time periods. A day is divided into several time periods according to the time range in which each management time period is located.

4. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The method of setting the toll lanes to be opened at the toll station in the next time period based on congestion factors includes: Extract the congestion factor YZ. When the congestion factor YZ is greater than the congestion threshold one, open all toll lanes at the current toll station. Wherein, the congestion threshold one is greater than the congestion threshold two. When the congestion factor is no greater than congestion threshold one and no less than congestion threshold two, extract the number of toll lanes TDL at the current toll station, based on the formula KDL= TDL / 2 Determine the number of open toll lanes KDL at the current toll station, and open the toll lanes at the current toll station from left to right based on the number KDL; When the congestion factor is less than the congestion threshold of 2, the leftmost toll lane of the current toll station is opened.

5. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The toll lanes, based on the established settings, guide vehicles to each toll lane using an artificial intelligence model, including: Extract the open toll lanes and the real-time number of vehicles within the corresponding time period from the historical reference data, as well as the toll lane number to which each vehicle is guided; wherein, the historical reference data includes several open toll lanes and the real-time number of vehicles within the corresponding time period, as well as the toll lane number to which experts guide each vehicle based on the toll lanes and the real-time number of vehicles within the corresponding time period. The open toll lanes and the real-time number of vehicles within the corresponding time period, as well as the toll lanes through which each vehicle is guided, are integrated into several sets of training data and test data. The training data is used to train the artificial intelligence model, and the test data is used to test the trained artificial intelligence model. The artificial intelligence model is then adjusted based on the test results. The final result is a lane guidance model with the input of the open toll lanes and the real-time number of vehicles, and the output of the toll lane number through which each vehicle is guided. When the next time period arrives, the number of open toll lanes and the real-time number of vehicles are input into the lane guidance model to obtain the number of the toll lane through which each vehicle is guided.

6. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The determination of road impact factors for each vehicle based on target data includes: Extract the current vehicle's weight CZ, speed V, number of braking attempts BC, and number of traffic violations WG; when the speed V exceeds the speed threshold VZ or the number of traffic violations WG exceeds the number of violations threshold WGZ, set the road impact factor DZ to the maximum value within the influence factor's defined range; When the speed V exceeds a speed threshold and the number of violations WG does not exceed a violation count threshold, the road impact factor DZ of the current vehicle is determined based on the vehicle weight CZ, speed V, number of braking attempts BC, and number of violations WG; the road impact factor DZ satisfies the following formula: ； Where BCZ is the standard vehicle weight and BCL is the standard number of braking cycles; , , and All of these are adjusted using proportional coefficients, and , ; When the road impact factor DZ is greater than the maximum value of the impact factor limit range, the road impact factor DZ will be set to the maximum value of the impact factor limit range; when the road impact factor DZ is less than the minimum value of the impact factor limit range, the road impact factor DZ will be set to the minimum value of the impact factor limit range.

7. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The method for determining highway tolls for each vehicle based on road impact factors includes: Obtain the current vehicle's mileage LC during this highway trip, extract the road impact factor DZ, and determine the current vehicle's highway toll GY based on the road impact factor DZ and the mileage LC; the highway toll GY satisfies the following formula: GY = ZY × LC × DZ / YZ; Where ZY is the base rate per kilometer, and YZ is the median value of the range of influence factors.

8. The unmanned adaptive toll collection method for highways according to claim 1, characterized in that, The collection of highway tolls via CPC card or ETC card includes: When a vehicle arrives at the CPC card payment counter or ETC card payment counter at a toll station, the highway toll for this trip can be paid using the CPC card or ETC card.

9. An unmanned adaptive toll collection system for highways, operating based on the unmanned adaptive toll collection method for highways according to any one of claims 1 to 8, characterized in that, include: The on-site dispatch module, as well as the data collection module and adaptive charging module connected to the on-site dispatch module; The data collection module is used to acquire the number of vehicles within the toll station's management area and target data for each vehicle; the target data includes vehicle weight, speed, number of braking attempts, and number of traffic violations. The on-site dispatch module is used to divide time periods and determine the congestion factor of the toll station in the next time period based on the number of vehicles in the current time period. Based on congestion factors, the toll lanes that will be open at the toll station in the next time period are set; when the next time period arrives, vehicles are guided to the respective toll lanes using an artificial intelligence model based on the set toll lanes. The adaptive toll collection module is used to determine the road impact factor of each vehicle based on target data; determine the highway toll of each vehicle based on the road impact factor; and collect the highway toll through CPC card or ETC card.

Images