Expressway automatic auditing method for inconformity between charging path and actual passing path

Abstract

The invention relates to the field of road traffic, in particular to a highway automatic auditing method aiming at the inconsistency of a charging path and an actual passing path, and aims to overcome the defects in the background technology so as to improve auditing efficiency, accuracy and real-time performance. According to the technical scheme of the invention, the highway automatic auditing method for the inconsistency between the charging path and the actual passing path comprises the following steps: S1, vehicle preliminary screening: extracting exit flow data, and preliminarily screening out problematic vehicles according to a troubleshooting rule; s2, actual toll collection path identification: establishing a mapping table of identification stations and toll collection units, and mapping the portal license plate identification data of the problem vehicle into an actual toll collection path; s3, complete charging path identification: collecting historical exit flow data, establishing a path cost mapping table, complementing identification station paths, and identifying a complete charging path in the path cost mapping table; and S4, matching the actual path and the charging path: when the complete charging path and the actual charging path do not conform to the matching rule, initiating an auditing work order.

Description

Technical Field

[0001] The invention relates to the field of road traffic, and in particular to an automatic audit method for expressways where a toll path does not match an actual travel path. Background Art

[0002] In the highway toll collection system, each vehicle will generate a corresponding toll record when it leaves the toll station. The toll record includes the vehicle's entry information, exit information, and payable fees.

[0003] Under normal circumstances, when a vehicle passes through a gantry, the ETC device will be written with gantry transaction data (gantry transaction data includes pass ID, gantry ID, license plate, and passing time). All gantry transaction data of a vehicle's one trip form a toll path, and the toll path should be consistent with the actual pass path to ensure the accuracy and rationality of the toll. However, in actual operation, due to factors such as human shielding, equipment failure, and missing gantry information, the gantry transaction data may fail to be written, resulting in a situation where the toll path does not match the actual pass path. Each time a vehicle passes through a section of the road under the jurisdiction of multiple highway operating companies, in order to protect the interests of drivers and passengers, the highway operating company will use the minimum amount for billing. At the final settlement, the superior unit will split the toll amount to each highway operating company according to the toll path. The actual toll deviation caused by the above situation not only damages the economic interests of the highway operating company, but also may cause traffic disorder and public dissatisfaction.

[0004] To address this problem, highway operating companies usually take toll audit measures, that is, review toll records to ensure their accuracy and rationality. Traditional audit methods mainly rely on manual review and limited monitoring methods, such as viewing surveillance videos and retrieving provincial and ministerial platform data to identify problem vehicles.

[0005] The defects and shortcomings are as follows:

[0006] 1. Low identification efficiency: Traditional audit methods mainly rely on manual review, which requires manual investigation by auditors, resulting in heavy workload and low efficiency. With the increase in vehicle traffic, this method is increasingly unable to meet actual needs.

[0007] 2. Accuracy is difficult to guarantee: Manual review is affected by factors such as auditors’ experience and energy, and is prone to misjudgment and omission.

[0008] 3. Insufficient data utilization: Highway toll data contains rich vehicle traffic information, but existing technologies often only focus on basic information such as toll amount, while ignoring other valuable data dimensions. This leads to insufficient utilization of data resources and inability to provide strong support for audit work.

[0009] In summary, the traditional auditing method has many defects and deficiencies in identifying vehicles whose toll path does not match the actual travel path. In order to overcome these problems, an intelligent auditing method based on highway toll data is urgently needed to improve the efficiency, accuracy and real-time nature of the audit, and bring greater economic and social benefits to highway operating companies. Summary of the invention

[0010] The purpose of the present invention is to overcome the deficiencies in the above-mentioned background technology and provide an automatic audit method for expressways where the toll path does not match the actual travel path, so as to improve the audit efficiency, accuracy and real-time performance.

[0011] The technical solution of the present invention is:

[0012] An automatic audit method for a highway where a toll path does not match an actual travel path comprises the following steps:

[0013] S1: Vehicle initial screening

[0014] Extract export flow data and preliminarily screen out problematic vehicles based on the troubleshooting rules;

[0015] S2: Identification of actual toll path

[0016] Establish a mapping table between identification stations and toll collection units to map the door frame plate recognition data of the problem vehicle to the actual toll collection route;

[0017] S3: Complete toll path identification

[0018] Collect historical export flow data, establish a path cost mapping table, complete the identification station path, and identify the complete toll path in the path cost mapping table;

[0019] S4: Matching the actual path with the toll path

[0020] When the complete charging path and the actual charging path do not meet the matching rules, an audit work order is initiated.

[0021] In step S1, the exit flow data includes license plate, vehicle type, entrance toll station, entrance time, exit toll station, exit time, toll path, identification station path, toll amount, and preferential policies.

[0022] In step S1, the troubleshooting rule is that the charging of the export flow data is abnormal or the identification station is missing.

[0023] The step S2 comprises:

[0024] 1) According to the longitude and latitude of the identification station and the longitude and latitude of the gantry, find the nearest charging unit to the identification station and establish a mapping table between the identification station and the charging unit;

[0025] 2) First, retrieve the gantry license plate recognition data of the problem vehicle through the license plate and time interval, then find the corresponding toll unit through the gantry license plate recognition data, and finally obtain the corresponding identification point through the identification station and toll unit mapping table to obtain the actual toll path.

[0026] In step S2, the mapping table between the identification station and the charging unit includes the identification point ID, the identification point longitude, the identification point latitude, the charging unit ID, the charging unit longitude, the charging unit latitude, and the shortest distance between the identification station and the charging unit.

[0027] The step S3 comprises:

[0028] 1) Collect historical export flow data, establish a route cost mapping table based on vehicle type, entrance toll station, exit toll station, preferential policies, and toll amount, and complete the missing identification points in the identification station path through big data analysis;

[0029] 2) According to the exit flow data of the problem vehicle, the records with the same entrance toll station, exit toll station, vehicle type, preferential policy and toll amount are retrieved in the path cost mapping table, and the identification station path of the retrieval result is used as the complete toll path.

[0030] In step S3, the path fee mapping table includes vehicle type, entrance toll station, exit toll station, preferential policies, identification station path, and toll amount.

[0031] In step S4, the matching rule is: all identification points of the actual toll path appear in the complete toll path in the same order.

[0032] The beneficial effects of the present invention are:

[0033] 1. High efficiency: The present invention realizes automatic identification of vehicles whose toll routes do not match the actual travel routes through big data analysis, which significantly improves the audit efficiency and reduces the workload of manual review.

[0034] 2. Accurate identification: The present invention is designed in combination with business scenarios, which can deeply mine the potential information in vehicle data, accurately determine the consistency between the toll path and the actual travel path, and reduce the probability of misjudgment and missed judgment.

[0035] 3. Intelligence and data utilization: The present invention makes full use of the multi-dimensional information in the highway toll data (such as entrance information, exit information, travel time, vehicle type, etc.), and improves the utilization rate of data resources and the accuracy of auditing work through data integration and analysis.

[0036] 4. Scalability: The technical solution of the present invention is based on actual high-speed toll collection business and is scalable to meet the auditing needs of different highway operating companies.

[0037] 5. Economic and social benefits: By accurately identifying vehicles that do not conform to the toll path, the present invention helps highway operating companies to recover toll evasion losses in a timely manner, protect their economic interests, and improve traffic order. It also helps to combat illegal behaviors such as toll evasion and detours, and improve highway traffic order and public satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0040] like Figure 1 As shown, a method for automatically checking a highway where a toll path does not match an actual travel path comprises the following steps:

[0041] S1: Vehicle initial screening

[0042] Due to the huge daily vehicle volume (reaching millions), and the fact that the toll routes of most vehicles are consistent with their actual routes, it would be a huge waste of resources to check all vehicles. Therefore, we first preliminarily screen out vehicles that may have problems, and then conduct further checks on these vehicles to save computing resources. Specifically, we extract exit flow data from each exit toll station and preliminarily screen out problem vehicles based on the checking rules.

[0043] The export flow data includes license plate, vehicle model, entrance toll station, entry time, exit toll station, exit time, toll route, identification station route, toll amount, and preferential policies.

[0044] The troubleshooting rules include:

[0045] ① Abnormal charging: the charging amount of the export flow data is 0, which means that all gantries have failed to trade, and there is a possibility of malicious blocking of ETC equipment;

[0046] ② Missing identification station: that is, the identification station path of the export flow data does not match the identification station completion path.

[0047] As long as one of the rules is met, it can be considered a problem vehicle.

[0048] The definition of the identification station is: in order to assist in determining the actual path, the identification station is deployed on any unambiguous section of the highway network (that is, there is only one unique path from the start point to the end point of this section). When the vehicle passes the identification station, an identifier will be written to the "identification station path vc_marks" field of the vehicle's ETC card, such as A, *, 3 and other unique identifiers. The identification station path can be read at the exit toll station to determine the vehicle's trajectory. If the identification station path is "A*3", it means that the vehicle passes through the three corresponding identification stations A, *, and 3 in sequence. When the vehicle model is the same, the entrance toll station is the same, and the exit toll station is the same, the toll path and the identification station path should be consistent.

[0049] However, considering the possibility of identifier writing failure, the exit toll station will complete the missing identifiers through the provincial and ministerial platforms and generate a new field "identification station completion path vc_fix_marks". However, the exit completion function still has limitations. If the missing is serious, the identification station completion path may still have missing identifiers. Therefore, it is necessary to use the path cost mapping table generated by historical data in the following steps to determine the actual path.

[0050] S2: Identification of actual toll path

[0051] 1) Establish a mapping table between identification stations and charging units

[0052] According to the longitude and latitude of the identification station and the gantry, find the toll unit closest to the identification station and establish a mapping table between the identification station and the toll unit so as to facilitate the actual path mapping in the next step.

[0053] The mapping table between identification stations and charging units includes identification point ID, identification point longitude, identification point latitude, charging unit ID, charging unit longitude, charging unit latitude, and the shortest distance between the identification station and the charging unit.

[0054] The toll unit is one-way. For example, the Beijing-Xiong'an direction of the Beijing-Xiongan Expressway has an indefinite division interval. Two consecutive gantries can be regarded as a toll unit, or three consecutive gantries can be regarded as a toll unit. The purpose of this division is to ensure the billing success rate. If there is no malicious shielding, it is impossible for two or three gantries to not be charged (at least one gantry transaction data will be generated in the same toll unit).

[0055] The identification station is bidirectional. For example, on the Beijing-Xiong'an direction and the Xiong'an-Beijing direction of the Beijing-Xiong'an Expressway, one identification point covers 4-6 gantries in both directions (2-3 consecutive gantries in one direction). The gantries in this area share one identification point to ensure that the identification point can be successfully written into the card.

[0056] 2) Map the gantry plate recognition data into the actual toll path through the identification station and toll unit mapping table

[0057] First, according to the problem vehicle detected in step S1, the gantry license plate recognition data is retrieved through the license plate and time interval (between the entry time and the exit time), and then the corresponding charging unit ID is found through the gantry ID of the gantry license plate recognition data. Finally, the corresponding identification point ID is obtained through the identification station and charging unit mapping table, and all the identification point IDs are connected in series by time to obtain the actual charging path.

[0058] The gantry license plate data is the data captured by the gantry bayonet camera when the vehicle passes through the gantry. The gantry license plate data includes the gantry ID, license plate, and vehicle passing time. Due to factors such as human shielding, equipment failure, and missing gantry information, the gantry transaction data may fail to be written, but the gantry camera can accurately obtain this part of the data, thereby obtaining the true passage trajectory.

[0059] S3: Complete toll path identification

[0060] After completion at the exit toll station, the path vc_fix_marks completed at the identification station may still be incomplete. For this reason, historical data is used to construct a path cost mapping table to achieve accurate identification of the toll path.

[0061] For vehicle traffic records with the same entry toll station, exit toll station, vehicle model, and preferential policies (focusing on factors such as ETC issuance location and whether it is a container), there is a one-to-one correspondence between the toll amount and the actual toll path. This means that even if the vc_fix_marks information of the identification station to complete the path generated by a specific trip may be incomplete, there must be a complete record based on the historical full data.

[0062] 1) Collect historical export flow data, establish a path cost mapping table, and complete the identification station path

[0063] Collect historical export flow data, merge the export flow data of the same vehicle type, the same entrance toll station, the same exit toll station, the same preferential policy, and the same toll amount into one record, and obtain the path cost mapping table. Then, use big data analysis to complete the missing identification points in the identification station path of each record. Big data analysis is an existing method, which usually uses the full amount of export flow data in the past one to three years to generate the path cost mapping table

[0064] For example:

[0065] There are 3 export flow data with the same vehicle type, entrance toll station, exit toll station, preferential policy and toll amount. The identification station paths are "L#(I", "IJ*E" and "L(IE" respectively. The completed identification station path is "L#(IJ*E".

[0066] The path cost mapping table includes vehicle type, entrance toll station, exit toll station, preferential policies, identification station path, and toll amount.

[0067] Preferential policies include within / outside the province and whether it is a container.

[0068] The path cost mapping table is shown in Table 1:

[0069] Model Entrance toll booth Exit toll booth In the province / outside the province Whether container Identification station path Amount of Charge 14 Shanxi Qiao Family Courtyard Scenic Area Station Zhejiang Sunrui Station Out of province Non-Containerized X! @t+ 2693.44 16 Shanghai Zhuanghu Highway Station Zhejiang Sunrui Station Province Non-Containerized L#(IJ*E 277.51 15 Shanghai Zhuanghu Highway Station Zhejiang Sunrui Station Out of province Non-Containerized L@t+ 306.38 11 Shanghai Zhuanghu Highway Station Zhejiang Sunrui Station Out of province Non-Containerized L@t+ 78.74

[0070] Table 1

[0071] 2) Identify the complete toll path in the path cost map

[0072] The exit flow data of the problem vehicle identified in step S1 is searched in the path fee mapping table, and the search conditions are the same entrance toll station, the same exit toll station, the same vehicle model, the same preferential policy, and the same toll amount. The identification station path of the search result is used as the complete toll path. This not only makes full use of the integrity of historical data, but also ensures the accuracy and reliability of toll path identification.

[0073] S4: Matching the actual toll path with the complete toll path

[0074] When the complete toll path and the actual toll path meet the matching rules, it means that the toll path is consistent with the actual route. When the actual toll path and the complete toll path do not meet the matching rules, it means that the toll path is inconsistent with the actual route. An audit work order needs to be initiated based on the actual toll path.

[0075] The matching rule is: all identification points of the actual toll path appear in the complete toll path in the same order.

[0076] The details are shown in Table 2:

[0077]

[0078] Table 2

[0079] Example

[0080] The present invention is mainly used to solve the problem of under-collection or missed collection of fees by shielding ETC, running long and buying short, etc. when vehicles pass through a certain ambiguous road section. Assuming that there are two paths between points AB, one of which is short but has a high fee, and the other is long but has a low fee, the vehicle passes along the shorter path by shielding ETC. Since the highway operating company cannot know the specific path, it can only choose the minimum amount for charging.

[0081] The present invention obtains the real passage trajectory through the gantry plate recognition data, and matches it with the complete toll path generated by historical data, so as to determine whether it is a problem vehicle.

[0082] The toll collection data of a highway operating company for one day was selected for testing, with a total of 7,000 truck export flow data. The route cost mapping table uses the export flow data of the last three years. Within 10 hours, the present invention found 53 work orders suspected of mismatching the actual toll path and the complete toll path by means of blocking ETC, running long and buying short, etc., and the accuracy rate verified by the auditors was 98.1%.

[0083] Taking one of them as an example, the initial toll gantry in the province is the Zhejiang-Shanghai Main Line-Jiashan Fu 1, the final toll gantry in the province is Keqiao, the vehicle type is a truck (category 5), and the actual toll path is A@R@f@e@s). Due to evasion or other reasons, the actual toll split to the highway operating company is only 3.98 yuan. The present invention restores the complete toll path of the vehicle to A@fBGI*, and the actual toll path does not match the complete toll path. Therefore, an audit work order is initiated based on the actual toll path, and the toll split to the highway operating company should be 26.11 yuan after re-billing through the provincial audit system.

[0084] The present invention can greatly recover the loss of tolls due to under-collection or missed collection, and avoids the defects of the existing method that is time-consuming and labor-intensive, difficult to fully cover all vehicles, has low audit efficiency and is prone to missing problematic vehicles.

[0085] The preferred embodiments of the present invention are shown in the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described in this specification. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

Claims

1. A method for automatically checking a highway whose toll path does not match the actual travel path, comprising the following steps: S1: Vehicle initial screening Extract export flow data and preliminarily screen out problematic vehicles based on the troubleshooting rules; S2: Identification of actual toll path Establish a mapping table between identification stations and toll collection units to map the door frame plate recognition data of the problem vehicle to the actual toll collection route; S3: Complete toll path identification Collect historical export flow data, establish a path cost mapping table, complete the identification station path, and identify the complete toll path in the path cost mapping table; S4: Matching the actual path with the toll path When the complete charging path and the actual charging path do not meet the matching rules, an audit work order is initiated.

2. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 1, characterized in that: In step S1, the exit flow data includes license plate, vehicle type, entrance toll station, entrance time, exit toll station, exit time, toll path, identification station path, toll amount, and preferential policies.

3. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 2, characterized in that: In step S1, the troubleshooting rule is that the charging of the export flow data is abnormal or the identification station is missing.

4. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 1, characterized in that: The step S2 comprises: 1) According to the longitude and latitude of the identification station and the longitude and latitude of the gantry, find the nearest charging unit to the identification station and establish a mapping table between the identification station and the charging unit; 2) First, retrieve the gantry license plate recognition data of the problem vehicle through the license plate and time interval, then find the corresponding toll unit through the gantry license plate recognition data, and finally obtain the corresponding identification point through the identification station and toll unit mapping table to obtain the actual toll path.

5. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 4, characterized in that: In step S2, the mapping table between the identification station and the charging unit includes the identification point ID, the identification point longitude, the identification point latitude, the charging unit ID, the charging unit longitude, the charging unit latitude, and the shortest distance between the identification station and the charging unit.

6. The method for automatically checking a highway for discrepancies between a toll path and an actual travel path according to claim 1, characterized in that: The step S3 comprises: 1) Collect historical export flow data, establish a route cost mapping table based on vehicle type, entrance toll station, exit toll station, preferential policies, and toll amount, and complete the missing identification points in the identification station path through big data analysis; 2) According to the exit flow data of the problem vehicle, the records with the same entrance toll station, exit toll station, vehicle type, preferential policy and toll amount are retrieved in the path cost mapping table, and the identification station path of the retrieval result is used as the complete toll path.

7. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 6, characterized in that: In step S3, the path fee mapping table includes vehicle type, entrance toll station, exit toll station, preferential policies, identification station path, and toll amount.

8. The method for automatically checking a highway whose toll path does not match the actual travel path according to claim 1, characterized in that: In step S4, the matching rule is: all identification points of the actual toll path appear in the complete toll path in the same order.

Images