CN114170696A - Real-time toll calculation system and method for differential charging of expressway - Google Patents
Real-time toll calculation system and method for differential charging of expressway Download PDFInfo
- Publication number
- CN114170696A CN114170696A CN202111546535.5A CN202111546535A CN114170696A CN 114170696 A CN114170696 A CN 114170696A CN 202111546535 A CN202111546535 A CN 202111546535A CN 114170696 A CN114170696 A CN 114170696A
- Authority
- CN
- China
- Prior art keywords
- missing
- road
- charging
- road section
- toll
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07B—TICKET-ISSUING APPARATUS; FARE-REGISTERING APPARATUS; FRANKING APPARATUS
- G07B15/00—Arrangements or apparatus for collecting fares, tolls or entrance fees at one or more control points
- G07B15/06—Arrangements for road pricing or congestion charging of vehicles or vehicle users, e.g. automatic toll systems
- G07B15/063—Arrangements for road pricing or congestion charging of vehicles or vehicle users, e.g. automatic toll systems using wireless information transmission between the vehicle and a fixed station
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q20/00—Payment architectures, schemes or protocols
- G06Q20/08—Payment architectures
- G06Q20/14—Payment architectures specially adapted for billing systems
- G06Q20/145—Payments according to the detected use or quantity
Landscapes
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a real-time toll calculation system and method for highway differentiated charging, wherein the system comprises a nationwide toll road networking settlement management center, a provincial networking settlement management center and edge calculation units arranged on toll stations and ETC gantries; the national toll road networking settlement management center and the provincial networking settlement management center are used for uniformly generating, adjusting and maintaining a road network data structure table COO and transmitting the COO data to the edge computing unit in real time through a communication system; and the edge computing unit computes the minimum cost of the charging traffic medium recording the missing road section in real time by adopting a parallel heterogeneous shortest path finding method. The invention can meet the real-time requirements of differential charging and special condition charging under the increasingly complex and differential charging conditions of the highway network, and further optimizes the charging work of the highway without a provincial toll station.
Description
Technical Field
The invention belongs to the technical field of highway management, and particularly relates to a real-time toll calculation system and method for highway differentiated charging.
Background
The section such as the transportation department, 6.15.2021, further optimizes and perfects the differentiated charging mode of the branch section and steadily expands the implementation range of the differentiated charging, in the road sections with serious congestion in the trunk roads or urban roads of the common countries and small traffic flow of the parallel expressways, the road sections with large traffic flow difference among the parallel expressways and the road sections with the traffic flow obviously lower than the design capacity, flexible and various differentiated charging is implemented, and the differentiated charging in the modes of road section division, vehicle type division, time section division, direction division, toll station division and branch payment is implemented for all vehicles (including ETC vehicles and MTC vehicles) on the expressways, so that the aims of utilizing price levers to balance the traffic flow distribution of the road network, improving the integral operation efficiency of the regional road network and promoting the cost reduction and the efficiency improvement of regional logistics transportation are fulfilled.
At present, the technical scheme of 'sectional charging' implemented on the expressway is established on a technical route capable of accurately identifying a vehicle driving path, but due to the fact that malicious fee evasion, shielded wireless passing charging equipment, missed identification of ETC portal equipment and other equipment exist, a considerable part of vehicles still have the situations of partial path information loss and ambiguous paths. The current method for charging vehicles with path information missing is a tariff query method, namely, a total shortest path table (tariff) is generated by central calculation, and the tariff table related to each toll station is issued to each toll station, so that more and overlong time is consumed in the link, the application range and the charging fairness of the method are greatly limited, and the aim of charging vehicles through toll stations without stopping is difficult to achieve.
The algorithm complexity of the traditional shortest path algorithm is O (V2) or O (ElogV), the traditional shortest path algorithm is a serial processing search algorithm, the requirement of lane calculation time within 50ms-100ms for a large-scale complex road network is difficult to realize before a provincial toll station is cancelled, and the traditional shortest path algorithm is not practically applied to a toll system of the large-scale complex road network all the time. Thus, with the increasing complexity of highway networks and differentiated charging requirements, the overall shortest path table will become exceptionally complex.
In order to avoid the phenomena of toll unfairness between users with complete highway section information and users with partial missing highway section information, the real-time highway toll calculation system and method for highway differentiated toll collection aim at the characteristic that the toll changes frequently under the condition of differentiated toll collection, a COO road network data structure is changed on the conventional highway toll collection system, and an edge calculation unit is additionally arranged, so that the updating of road network data is convenient and rapid, the real-time requirement for calculating the toll under a differentiated toll collection mechanism can be met, the toll fairness is realized, the highway toll collection benefit is improved, and meanwhile, the excessive system reconstruction cost is not increased.
Disclosure of Invention
The invention mainly aims to overcome the defects and shortcomings of the prior art, and provides a real-time toll calculation system and method for highway differentiated charging, so as to meet the real-time requirement of special charging and differentiated charging under increasingly complex conditions of a highway network and further optimize the charging work of a highway without a provincial toll station.
In order to achieve the purpose, the invention adopts the following technical scheme:
a real-time toll calculation system for highway differentiated charging comprises a nationwide toll road networking settlement management center, a provincial networking settlement management center, toll stations arranged at the entrances and exits of highways and an edge calculation unit of a main line ETC portal frame;
the national toll road networking settlement management center and the provincial networking settlement management center are used for uniformly generating, adjusting and maintaining a road network data structure table COO and transmitting the COO data to the edge computing unit in real time through a communication system;
the edge calculation unit adopts SSSP algorithm based on CUDA to realize real-time calculation of the minimum cost of the toll medium recording the missing road section.
Further, the edge computing unit comprises a CPU, a GPU, a security authentication module and a data storage module, and is used for realizing data storage, encryption authentication, data confidentiality, key dispersion, digital signature and transmission line protection.
Further, the toll passing media are specifically a CPC and an OBU;
the communication system is composed of a toll road private network and a public network based on operator communication.
Further, the COO road network data structure represents a start point code, an end point code, a start point-end point distance and a cost of each road segment, which are elements of a road network, and is represented by a four-tuple, specifically:
starting point number, end point number, distance and differentiated charging fee;
the starting point code and the end point code are unified station code numbers of toll stations and ETC gantries in a road network, the distance is the actual road section length between adjacent stations, and the cost is the differential charging cost calculated by the road section according to the vehicle type, the time period, the direction and the payment mode;
the COO table change column part can be transmitted to a toll station and an ETC portal frame edge computing terminal in real time through a public network under the condition that a toll road private network or a private network is disconnected;
the COO road network data generation is expressed as: coding toll stations and ETC gantries in a road network, forming road sections between adjacent stations, forming a quadruple by starting and ending point numbers, distances and expenses of the road sections, and integrating the quadruple of all the road sections to generate complete COO road network data;
the updating mode of the COO road network structure data table is as follows: when a certain road section is closed, the corresponding column data of the road section in the logout table is cancelled; the closing of a certain road section is removed, and the corresponding row data of the road section in the table is recovered; newly opening a certain station and a certain road section, and adding a column of data representing the corresponding road section in the table; the cost of a certain road section is changed, and the cost in the corresponding column of the road section in the table is changed.
Furthermore, a heterogeneous parallel computing method of a CPU and a GPU is used, the comparison computing load of a large number of intensive nodes and edge data in the shortest path algorithm is transferred to the GPU, the GPU processes the path data in parallel, and the CPU still runs other program codes;
the GPU of the edge calculation unit comprises p processors, each processor of the GPU is firstly assigned with n nodes to respectively calculate a local minimum value, then p processors of the GPU cooperate to calculate a global minimum value, and finally the minimum value is broadcasted;
the p processor cooperation method is as follows: when p is an even number, the back p/2 processors respectively send the local minimum values of the back p/2 processors to the corresponding front p/2 processors, and the front p/2 processors compare the smaller local minimum values of the 2 local minimum values and keep the smaller local minimum values;
when p is an odd number, setting p to be 2h +1, respectively sending values of the next h processors to the previous h processors, comparing and keeping the current minimum value; such a layer-by-layer comparison yields a unique global minimum after logp cycles.
Further, the single-source shortest path algorithm comprises the following specific steps:
firstly, defining an array path with the size of | V |, and mapping the array path to the shortest path estimation value from a source vertex s to each vertex V; defining an updating set E of vertexes, wherein the estimation of the shortest path is executed in each single program, and the estimation of the shortest path after the vertexes can be updated at any time; in order to determine whether the shortest path estimation of the vertex is changed during the program execution and avoid read-write inconsistency, the algorithm needs to define another update array bpaths;
the algorithm is realized by the following operation process: first, in the initialization process, the path [ s ] value and bpaths [ s ] value of the source vertex s are defined as 0, and the initialization path [ v ] value and bpaths [ v ] value of the remaining vertices v are defined as ∞, indicating that no other vertex v can be reached from the source point s at the beginning;
the update set E is initially the source point s, and next the algorithm is logically divided into two processes: loosening and checking;
in the relaxation process, each vertex v in the update set E continuously performs relaxation on the direct successor u, namely the comparison size between the bpaths [ u ] and the paths [ v ] + w (v, u), if the bpaths [ u ] is larger, the value of the paths [ v ] + w (v, u) is used for updating the bpaths [ u ], and then the v itself is deleted from the update set E;
in the checking process, each vertex v compares the size of bpaths [ v ] with that of the path [ v ], if the bpaths [ v ] is smaller than the path [ v ], the former is used for updating the latter, and the vertex v itself is applied and added to an updating set E;
the two operations of relaxation and checking are repeated until the update set E is empty, at which time the shortest path value from the source s to the target v is stored in the paths [ v ].
The invention also comprises a real-time calculating method of the toll for the expressway differentiated charging based on the provided system, which comprises the following steps:
s1, integrating vehicle information, reading a departure point and a passing toll collection point of a pre-toll vehicle from an OBU and a CPC card by edge computing terminals such as a highway entrance/exit toll station, a main line ETC portal frame and the like, and recording data as a vehicle driving process data matrix;
s2, judging path information, acquiring mileage charging information of the passing toll collection points of the pre-charged vehicle from the OBU and the CPC card according to the data matrix of the vehicle driving process, determining a path information missing road section and a path information missing road section, and establishing a vehicle driving information missing road section mileage matrix and a vehicle driving information missing road section mileage matrix;
s3, calculating the charging of the non-missing road sections, determining mileage matrixes of the non-missing road sections after the non-missing road sections are determined according to a vehicle driving process data matrix, acquiring corresponding road section fees according to a COO road network data table, establishing non-missing road section charging matrixes, and calculating the total charging fees of the non-missing road sections;
s4, calculating the charge of the missing road section, determining a starting point and an ending point matrix of the missing road section, calculating the shortest path of the missing road section by an edge calculation terminal through a parallel heterogeneous shortest path searching method of a CPU and a GPU according to the starting point and the ending point of the missing road section, generating the shortest path information of vehicle driving of the missing section, constructing the number of the shortest paths of the missing section, calculating the charge matrix of the missing section according to the calculated shortest path information of vehicle driving of the missing section and a COO road network data table, and calculating the total charge of the missing section;
and S5, displaying the calculated result, and displaying the total vehicle charge on the personal terminal of the staff at the road charge network.
Further, in step S1, the vehicle driving process data matrix is specifically:
S(A,A0,A1,S0,S1,S2...Sn)
the method comprises the following steps that A represents vehicle license plate number information, A0 represents a vehicle type, A1 represents a vehicle type grade, S0 is a vehicle starting point, S1 represents that a vehicle passes through a first charging network point, S2 represents that the vehicle passes through a second charging network point, and Sn represents that the vehicle passes through an nth charging network point;
in step S2, the road-mileage matrix in which the vehicle driving information is not missing is:
F(F1,F2,...Fn-1)
wherein, F1 represents the first road section mileage charging standard, F2 represents the second road section mileage charging standard.
The mileage matrix of the road section with the missing vehicle driving information is specifically as follows:
Q(Q1,Q2,...Qn)
wherein Q1 denotes a first missing road segment matrix, Q2 denotes a second missing road segment matrix, and Qn denotes an nth missing road segment matrix;
in step S3, the missed road section charging matrix is specifically:
FW(FW1,FW2...FWn)
wherein FW1 represents charging for the first section without missing, FW2 represents charging for the second section without missing, FWn represents charging for the nth section without missing;
in step S4, the missing link shortest path data is specifically L (L1, L2.. Ln), where L1 represents the first missing link shortest path matrix, L2 represents the second missing link shortest path matrix, and Ln represents the nth missing link shortest path matrix;
the missing road section charging matrix is specifically as follows:
QW(QW1,QW2...QWn)
where QW1 denotes the first missing link fee, QW2 denotes the second missing link fee, and QWn denotes the nth missing link fee.
Further, step S3 is specifically:
according to the vehicle driving process data matrix S, determining a mileage matrix F of a road section driven by the vehicle after determining the road section not missing, acquiring corresponding road section cost according to a COO road network data table to establish a charging matrix of the road section not missing, and calculating the total charging amount of the road section not missing by adopting the following formula:
wherein, W represents the total charging amount of the non-missing road section, and FWi represents the total charging amount of the ith non-missing road section;
after the calculation result is summed with the sum of the early-stage passing cost, writing the sum into vehicle OBU and CPC passing media through equipment such as an ETC portal frame;
step S4 specifically includes:
after determining the starting point matrix Q and the ending point matrix Q of the missing road section, according to the starting point and the ending point of the missing road section, the edge computing terminal computes the shortest path of the missing road section through a parallel heterogeneous shortest path routing method of a CPU and a GPU, generates the shortest vehicle driving path information of the missing section, constructs the shortest path data L of the missing section, computes a charging matrix QW of the missing section according to the computed shortest vehicle driving path information of the missing section and a COO road network data table, computes the total charging cost DW of the missing section according to the charging matrix QW of the missing section, and adopts the following formula to compute:
wherein DW represents the total charging fee for the missing road section, and QWi represents the total charging fee for the ith missing road section;
and writing the vehicle OBU and CPC passing media in the calculation result and the sum of the previous passing fee through equipment such as an ETC portal frame which can successfully read the vehicle information.
Further, for a vehicle with a failed passing medium and unable to read the path information, the shortest path of the vehicle path information missing segment is calculated and charged in a heterogeneous parallel calculation shortest path acceleration algorithm, which is specifically realized by adopting the following formula:
wherein SF represents the total charge of vehicles, COO Coordinate represents a COO road network data structure, private + public CNSS represents a communication system consisting of a toll road private network and an operator communication public network, nvGRAPH () represents the nvGRAPH function of the CUDA platform, namely, the shortest path of the missing road section is calculated in real time by heterogeneous parallel calculation.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. in the system and the method, the department-province-regional networking center can adjust the road network data in time according to the conditions of closing, newly opening, modifying, differentiating change and the like of the actual road network and conveniently transmit the road network data to the edge computing terminal such as an ETC portal frame, a toll station and the like.
2. The edge computing unit of the invention uses a parallel heterogeneous shortest path finding method of CPU + GPU to compute the shortest path of the missing road section, so that the computing speed is reduced to below 50ms, and each terminal has the capability of computing the shortest path between any two sites in real time.
3. The system and the method provided by the invention only need to add the edge calculation unit at the toll station and the main line ETC portal frame of the existing system, change the COO road network data structure, can conveniently realize high-speed differentiated charging, improve the fairness of charging, avoid charging loss and have extremely low modification cost for the existing system.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a flow chart of the method of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited thereto.
Examples
As shown in fig. 1, the present invention provides a real-time calculating system for tolls of highway differentiated charging, comprising: the system comprises a national toll road networking settlement management center, a provincial networking settlement management center and edge calculation units arranged at toll stations and ETC gantries;
the national toll road networking settlement management center and the provincial networking settlement management center are used for uniformly generating, adjusting and maintaining a road network data structure table COO and transmitting the COO data to the edge computing unit in real time through a communication system;
the edge calculation unit adopts an SSSP algorithm based on CUDA to realize real-time calculation of the minimum cost of the toll medium recording the missing road section; the CUDA platform is a CPU-GPU heterogeneous computing model, wherein a CPU is used as a Host (Host), a GPU is used as a Coprocessor (Coprocessor) or Device (Device), the serial logic and task scheduling of the whole program are controlled by the CPU, and the GPU is used for executing a data parallel part, namely the GPU and the CPU work cooperatively. The SSSP algorithm based on the CUDA defines an array paths, an update set E and an update array bpaths, and continuously updates and iterates through relaxation and check processes until the update set E is empty, namely, the shortest path value of the path paths is obtained.
In this embodiment, the edge computing unit includes a CPU, a GPU, a security authentication module, and a data storage module, and is configured to implement functions such as data storage, encryption authentication, data confidentiality, key distribution, digital signature, and transmission line protection, and support a hardware cryptographic algorithm.
The toll passing media are CPC and OBU;
the communication system is composed of a toll road private network and a public network based on operator communication.
The COO road network data structure represents the starting point code, the end point code and the cost of different vehicle type road sections of each element of the road network and is represented by a four-tuple (starting point number, end point number, distance and differential charging cost); coding toll stations and ETC gantries in a road network, and forming road sections between adjacent stations; such a road segment is represented by a quadruple, which may represent the length, cost, time, etc. of the road segment.
The starting point code and the end point code are unified station code numbers of toll stations and ETC gantries in a road network, the distance is the actual road section length between adjacent stations, and the cost is the differential charging cost calculated by the road section according to the vehicle type, the time period, the direction and the payment mode;
the COO table change column part can be transmitted to a toll station and an ETC portal frame edge computing terminal in real time through a public network under the condition that a toll road private network or a private network is disconnected;
the updating mode of the COO road network structure data table is as follows: when a certain road section is closed, the corresponding column data of the road section in the logout table is cancelled; the closing of a certain road section is removed, and the corresponding row data of the road section in the table is recovered; newly opening a certain station and a certain road section, and adding a column of data representing the corresponding road section in the table; the cost of a certain road section is changed, and the cost in the corresponding column of the road section in the table is changed.
In this embodiment, the road network generation model is specifically as follows:
an expressway network is composed of a plurality of expressways, and the main elements are as follows:
1) point: the system comprises a toll station, an interworking junction and an ETC portal;
2) road section (side): forming a road section between adjacent stations;
3) the route is as follows: adjacent connection portions are connected in segments to form a route, i.e., a travel path.
A plurality of expressways are connected to each other by an intersection and a starting point to form a road network. In order to realize vehicle travel path identification, a vehicle path identification point (ETC portal) needs to be established on a road section where an ambiguous path may occur.
According to the above construction of the basic elements of the road network, in order to describe the whole road network information, at least the following information must be encoded to establish the relevant basic data:
1) road section: number, attribution
2) A toll station: number, road information, position information
3) ETC portal: number, position of ETC portal
4) Line segment attributes between two sites: first and last sites, total mileage, rate, and amount of charge
5) Interworking attributes between two road segments: interconnection between two sites
6) The attribute of the route between two points of the cross-road section: information of two stations, mileage, toll, route
Therefore, according to the new station information, a toll station coding table and an ETC portal coding table are compiled, and a tariff table between two adjacent stations (toll stations, interchange and ETC portals) in the whole road section is compiled, so that the foundation is laid for reducing the actual passing path of the vehicle and calculating the tariff.
By using a heterogeneous parallel computing method of a CPU and a GPU, the comparison computing load of a large amount of intensive nodes and edge data in the shortest path algorithm is transferred to the GPU, the GPU processes the path data in parallel, and the CPU still runs other program codes;
the GPU of the edge calculation unit comprises p processors, each processor of the GPU is firstly assigned with n nodes to respectively calculate a local minimum value, then p processors of the GPU cooperate to calculate a global minimum value, and finally the minimum value is broadcasted;
the p processor cooperation method is as follows: when p is an even number, the back p/2 processors respectively send the local minimum values of the back p/2 processors to the corresponding front p/2 processors, and the front p/2 processors compare the smaller local minimum values of the 2 local minimum values and keep the smaller local minimum values;
when p is an odd number, setting p to be 2h +1, respectively sending values of the next h processors to the previous h processors, comparing and keeping the current minimum value; such a layer-by-layer comparison yields a unique global minimum after logp cycles.
The single-source shortest path algorithm comprises the following specific steps:
firstly, defining an array path with the size of | V |, and mapping the array path to the shortest path estimation value from a source vertex s to each vertex V; defining an updating set E of vertexes, wherein the estimation of the shortest path is executed in each single program, and the estimation of the shortest path after the vertexes can be updated at any time; in order to determine whether the shortest path estimation of the vertex is changed during the program execution and avoid read-write inconsistency, the algorithm needs to define another update array bpaths;
the algorithm is realized by the following operation process: first, in the initialization process, the path [ s ] value and bpaths [ s ] value of the source vertex s are defined as 0, and the initialization path [ v ] value and bpaths [ v ] value of the remaining vertices v are defined as ∞, indicating that no other vertex v can be reached from the source point s at the beginning;
the update set E is initially the source point s, and next the algorithm is logically divided into two processes: loosening and checking;
in the relaxation process, each vertex v in the update set E continuously performs relaxation on the direct successor u, namely the comparison size between the bpaths [ u ] and the paths [ v ] + w (v, u), if the bpaths [ u ] is larger, the value of the paths [ v ] + w (v, u) is used for updating the bpaths [ u ], and then the v itself is deleted from the update set E;
in the checking process, each vertex v compares the size of bpaths [ v ] with that of the path [ v ], if the bpaths [ v ] is smaller than the path [ v ], the former is used for updating the latter, and the vertex v itself is applied and added to an updating set E;
the above two operations are repeated until the update set E is empty, and the shortest path value from the source point s to the target v is stored in the paths [ v ].
For a vehicle which cannot read the path information due to failure of a passing medium, calculating the shortest path of a vehicle path information missing section and charging the shortest path by a heterogeneous parallel calculation shortest path acceleration algorithm, and specifically adopting a formula (1):
wherein, SF represents the total charge of vehicles, W represents the total charge of the undisrupted road sections, DW represents the total charge of the lost road sections, COO Coordinate represents a COO road network data structure, private + public CNSS represents a communication system consisting of a toll road private network and an operator communication public network, nvGRAPH () represents the nvGRAPH function of the CUDA platform, namely, the shortest path of the lost road sections is calculated in real time by heterogeneous parallel calculation.
In this embodiment, taking the Guangdong province toll road as an example, the Guangdong province toll road stations are coded, and after the specific stations are converted into numbers, differentiated charging policies of different time periods, different vehicle types and different payment forms are considered, so that a final COO road network data table is formed as shown in the following table 1.
TABLE 1
In another embodiment, a method for calculating the toll for highway differentiated charging based on the above embodiment in real time is also provided, as shown in fig. 2, including the following steps:
s1, integrating vehicle information, reading a departure point and a passing toll collection point of a pre-toll vehicle from an OBU and a CPC card by an edge computing terminal such as an ETC portal frame and the like, and recording data as a vehicle driving process data matrix S (A, A0, A1, S0, S1, S2.. Sn);
the method comprises the following steps that A represents vehicle license plate number information, A0 represents a vehicle type, A1 represents a vehicle type grade, S0 is a vehicle starting point, S1 represents that a vehicle passes through a first charging network point, S2 represents that the vehicle passes through a second charging network point, and Sn represents that the vehicle passes through an nth charging network point;
s2, judging path information, acquiring mileage charging information of the passing toll collection points of the pre-charged vehicle from the OBU and the CPC card according to the vehicle driving process data matrix S, determining a path information missing road section and a non-missing road section, and establishing a road section mileage matrix F (F1, F2,. Fn-1) without vehicle driving information missing and a road section mileage matrix Q (Q1, Q2,. Qn) with vehicle driving information missing;
wherein, F1 represents the first road section mileage charging standard, F2 represents the second road section mileage charging standard. Q1 denotes a first missing link matrix, Q2 denotes a second missing link matrix, and Qn denotes an nth missing link matrix.
S3, calculating the charge of the un-missing road section, determining the mileage matrix F of the un-missing road section after the un-missing road section is determined according to the vehicle driving process data matrix S, obtaining the corresponding road section charge according to the COO road network data table to establish the charge matrix FW (FW1, FW2.. FWn) of the un-missing road section, calculating the total charge of the un-missing road section, and adopting a formula (2) to calculate the total charge of the un-missing road section:
wherein FW1 represents charging for the first section without missing, FW2 represents charging for the second section without missing, FWn represents charging for the nth section without missing; w represents the total charge amount of the non-missing road section, and FWi represents the total charge amount of the ith non-missing road section;
and writing the vehicle OBU and CPC passing media into the ETC portal frame through equipment such as the ETC portal frame after the calculation result is summed with the sum of the previous passing fees.
S4, calculating the charge of the missing road section, determining a starting point and an end point matrix Q of the missing road section, calculating the shortest path of the missing road section by an edge calculation terminal through a parallel heterogeneous shortest path routing method of a CPU and a GPU according to the starting point and the end point of the missing road section, generating the shortest path information of vehicle driving of the missing section, constructing shortest path data L (L1, L2.. Ln) of the missing section, calculating a charge matrix QW (QW1, QW2.. QWn) of the missing section according to the calculated shortest path information of vehicle driving of the missing section and a COO road network data table, calculating the total charge DW of the missing section, and calculating by adopting a formula (3):
wherein DW represents the total charging fee for the missing road section, and QWi represents the total charging fee for the ith missing road section;
and writing the vehicle OBU and CPC passing media in the calculation result and the sum of the previous passing fee through equipment such as an ETC portal frame which can successfully read the vehicle information.
And S5, displaying the total vehicle charge DW + W on the personal terminal of the staff at the road toll collection point according to the calculation result.
The present embodiment considers the expansion of the road network scale based on the current road network node scale. Based on the virtual road network NY (264346 nodes in total, 733846 sides), 52 road network data are constructed in the number of nodes from 5000 nodes to 260000 nodes at intervals of 5000 nodes. Meanwhile, charging standards of various types of passenger cars and trucks published by the official are imported into the virtual road network, the passing fee of the path information missing section is calculated, and the simulation calculation of the actual passing rate is carried out.
And a heterogeneous parallel computing method of the CPU and the GPU is used for realizing the nvGRAPH function which is newly introduced by a CUDA platform. This function takes advantage of the linear algebra function of the GPU to handle the largest graphics analysis and big data analysis problems. The specific operation flow is as follows: first, nvgraph create () is called to initialize the library. Next, the uploading of graphics data into the library may continue through the nvGRAPH API; graphics data can be appended to the vertices and/or edges of the graphics using nvgraphSetVertexData () and nvgraphSetEdgeData (), respectively. Multiple values of data may exist simultaneously on each edge or vertex, each value accessing the dataset array by an index. A graph algorithm may then be run on the data, extracting subgraphs from the data, and reformatting the data using the nvGRAPH API. The results may be downloaded back to the host or copied to other locations on the device. After all operations are run, nvgraphdestory () is called to release resources. Extract subgraphs from the data, or reformat the data using the nvGRAPH API.
The method carries out simulation calculation and analysis on the traffic cost of the road section with missing path information in a virtual road network experiment, and the calculation time is less than 50ms in the simulation calculation. Meanwhile, the specific path of the calculation result can be fed back for query, and the technical feasibility of the real-time calculation scheme is fully demonstrated.
It should also be noted that in this specification, terms such as "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A real-time toll calculation system for highway differentiated charging is characterized by comprising a national toll road networking settlement management center, a provincial networking settlement management center, toll stations arranged at the entrances and exits of a highway and an edge calculation unit of a main line ETC portal frame;
the national toll road networking settlement management center and the provincial networking settlement management center are used for uniformly generating, adjusting and maintaining a road network data structure table COO and transmitting the COO data to the edge computing unit in real time through a communication system;
the edge calculation unit adopts SSSP algorithm based on CUDA to realize real-time calculation of the minimum cost of the toll medium recording the missing road section.
2. The real-time toll calculation system for highway differentiated charging according to claim 1, wherein the edge calculation unit comprises a CPU, a GPU, a security authentication module and a data storage module, and is used for realizing data storage, encryption authentication, data confidentiality, key distribution, digital signature and transmission line protection.
3. The real-time highway toll collection computing system according to claim 1, wherein toll collection passing media are specifically CPC and OBU;
the communication system is composed of a toll road private network and a public network based on operator communication.
4. The system of claim 1, wherein the COO road network data structure represents the start point code, the end point code, the distance between the start point and the end point and the cost of the road section for each element-road section of the road network, and is represented by a four-tuple, which comprises:
starting point number, end point number, distance and differentiated charging fee;
the starting point code and the end point code are unified station code numbers of toll stations and ETC gantries in a road network, the distance is the actual road section length between adjacent stations, and the cost is the differential charging cost calculated by the road section according to the vehicle type, the time period, the direction and the payment mode;
the COO table change column part can be transmitted to a toll station and an ETC portal frame edge computing terminal in real time through a public network under the condition that a toll road private network or a private network is disconnected;
the COO road network data generation is expressed as: coding toll stations and ETC gantries in a road network, forming road sections between adjacent stations, forming a quadruple by starting and ending point numbers, distances and expenses of the road sections, and integrating the quadruple of all the road sections to generate complete COO road network data;
the updating mode of the COO road network structure data table is as follows: when a certain road section is closed, the corresponding column data of the road section in the logout table is cancelled; the closing of a certain road section is removed, and the corresponding row data of the road section in the table is recovered; newly opening a certain station and a certain road section, and adding a column of data representing the corresponding road section in the table; the cost of a certain road section is changed, and the cost in the corresponding column of the road section in the table is changed.
5. The real-time toll calculation system for highway differentiated charging according to claim 1, wherein the workload of comparison calculation of a large amount of intensive node and edge data in the shortest path algorithm is transferred to the GPU by using a heterogeneous parallel calculation method of CPU plus GPU, the GPU processes the path data in parallel, and the CPU runs the rest program codes;
the GPU of the edge calculation unit comprises p processors, each processor of the GPU is firstly assigned with n nodes to respectively calculate a local minimum value, then p processors of the GPU cooperate to calculate a global minimum value, and finally the minimum value is broadcasted;
the p processor cooperation method is as follows: when p is an even number, the back p/2 processors respectively send the local minimum values of the back p/2 processors to the corresponding front p/2 processors, and the front p/2 processors compare the smaller local minimum values of the 2 local minimum values and keep the smaller local minimum values;
when p is an odd number, setting p to be 2h +1, respectively sending values of the next h processors to the previous h processors, comparing and keeping the current minimum value; such a layer-by-layer comparison yields a unique global minimum after logp cycles.
6. The real-time toll calculation system for highway differentiated charging according to claim 5, wherein the single-source shortest path algorithm comprises the following specific steps:
firstly, defining an array path with the size of | V |, and mapping the array path to the shortest path estimation value from a source vertex s to each vertex V; defining an updating set E of vertexes, wherein the estimation of the shortest path is executed in each single program, and the estimation of the shortest path after the vertexes can be updated at any time; in order to determine whether the shortest path estimation of the vertex is changed during the program execution and avoid read-write inconsistency, the algorithm needs to define another update array bpaths;
the algorithm is realized by the following operation process: first, in the initialization process, the path [ s ] value and bpaths [ s ] value of the source vertex s are defined as 0, and the initialization path [ v ] value and bpaths [ v ] value of the remaining vertices v are defined as ∞, indicating that no other vertex v can be reached from the source point s at the beginning;
the update set E is initially the source point s, and next the algorithm is logically divided into two processes: loosening and checking;
in the relaxation process, each vertex v in the update set E continuously performs relaxation on the direct successor u, namely the comparison size between the bpaths [ u ] and the paths [ v ] + w (v, u), if the bpaths [ u ] is larger, the value of the paths [ v ] + w (v, u) is used for updating the bpaths [ u ], and then the v itself is deleted from the update set E;
in the checking process, each vertex v compares the size of bpaths [ v ] with that of the path [ v ], if the bpaths [ v ] is smaller than the path [ v ], the former is used for updating the latter, and the vertex v itself is applied and added to an updating set E;
the two operations of relaxation and checking are repeated until the update set E is empty, at which time the shortest path value from the source s to the target v is stored in the paths [ v ].
7. The real-time calculation method for toll charge of highway differential charging based on the system of any one of claims 1-6 is characterized by comprising the following steps:
s1, integrating vehicle information, reading a departure point and a passing toll collection point of a pre-toll vehicle from an OBU and a CPC card by edge computing terminals such as a highway entrance/exit toll station, a main line ETC portal frame and the like, and recording data as a vehicle driving process data matrix;
s2, judging path information, acquiring mileage charging information of the passing toll collection points of the pre-charged vehicle from the OBU and the CPC card according to the data matrix of the vehicle driving process, determining a path information missing road section and a path information missing road section, and establishing a vehicle driving information missing road section mileage matrix and a vehicle driving information missing road section mileage matrix;
s3, calculating the charging of the non-missing road sections, determining mileage matrixes of the non-missing road sections after the non-missing road sections are determined according to a vehicle driving process data matrix, acquiring corresponding road section fees according to a COO road network data table, establishing non-missing road section charging matrixes, and calculating the total charging fees of the non-missing road sections;
s4, calculating the charge of the missing road section, determining a starting point and an ending point matrix of the missing road section, calculating the shortest path of the missing road section by an edge calculation terminal through a parallel heterogeneous shortest path searching method of a CPU and a GPU according to the starting point and the ending point of the missing road section, generating the shortest path information of vehicle driving of the missing section, constructing the number of the shortest paths of the missing section, calculating the charge matrix of the missing section according to the calculated shortest path information of vehicle driving of the missing section and a COO road network data table, and calculating the total charge of the missing section;
and S5, displaying the calculated result, and displaying the total vehicle charge on the personal terminal of the staff at the road charge network.
8. The method for calculating the toll of differential charging on the highway according to claim 7, wherein in step S1, the vehicle driving process data matrix is specifically:
S(A,A0,A1,S0,S1,S2...Sn)
the method comprises the following steps that A represents vehicle license plate number information, A0 represents a vehicle type, A1 represents a vehicle type grade, S0 is a vehicle starting point, S1 represents that a vehicle passes through a first charging network point, S2 represents that the vehicle passes through a second charging network point, and Sn represents that the vehicle passes through an nth charging network point;
in step S2, the road-mileage matrix in which the vehicle driving information is not missing is:
F(F1,F2,...Fn-1)
wherein, F1 represents the first road section mileage charging standard, F2 represents the second road section mileage charging standard.
The mileage matrix of the road section with the missing vehicle driving information is specifically as follows:
Q(Q1,Q2,...Qn)
wherein Q1 denotes a first missing road segment matrix, Q2 denotes a second missing road segment matrix, and Qn denotes an nth missing road segment matrix;
in step S3, the missed road section charging matrix is specifically:
FW(FW1,FW2...FWn)
wherein FW1 represents charging for the first section without missing, FW2 represents charging for the second section without missing, FWn represents charging for the nth section without missing;
in step S4, the missing link shortest path data is specifically L (L1, L2.. Ln), where L1 represents the first missing link shortest path matrix, L2 represents the second missing link shortest path matrix, and Ln represents the nth missing link shortest path matrix;
the missing road section charging matrix is specifically as follows:
QW(QW1,QW2...QWn)
where QW1 denotes the first missing link fee, QW2 denotes the second missing link fee, and QWn denotes the nth missing link fee.
9. The method for calculating the toll of differential charging on the highway according to claim 8, wherein the step S3 is specifically as follows:
according to the vehicle driving process data matrix S, determining a mileage matrix F of a road section driven by the vehicle after determining the road section not missing, acquiring corresponding road section cost according to a COO road network data table to establish a charging matrix of the road section not missing, and calculating the total charging amount of the road section not missing by adopting the following formula:
wherein, W represents the total charging amount of the non-missing road section, and FWi represents the total charging amount of the ith non-missing road section;
after the calculation result is summed with the sum of the early-stage passing cost, writing the sum into vehicle OBU and CPC passing media through equipment such as an ETC portal frame;
step S4 specifically includes:
after determining the starting point matrix Q and the ending point matrix Q of the missing road section, according to the starting point and the ending point of the missing road section, the edge computing terminal computes the shortest path of the missing road section through a parallel heterogeneous shortest path routing method of a CPU and a GPU, generates the shortest vehicle driving path information of the missing section, constructs the shortest path data L of the missing section, computes a charging matrix QW of the missing section according to the computed shortest vehicle driving path information of the missing section and a COO road network data table, computes the total charging cost DW of the missing section according to the charging matrix QW of the missing section, and adopts the following formula to compute:
wherein DW represents the total charging fee for the missing road section, and QWi represents the total charging fee for the ith missing road section;
and writing the vehicle OBU and CPC passing media in the calculation result and the sum of the previous passing fee through equipment such as an ETC portal frame which can successfully read the vehicle information.
10. The method for calculating the traffic fee for the differentiated charging on the expressway according to claim 9, wherein for the vehicles which cannot read the path information due to the failure of the traffic medium, the shortest path of the missing section of the vehicle path information is calculated and charged by a heterogeneous parallel calculation shortest path acceleration algorithm, which is specifically realized by adopting the following formula:
wherein SF represents the total charge of vehicles, COO Coordinate represents a COO road network data structure, private + public CNSS represents a communication system consisting of a toll road private network and an operator communication public network, nvGRAPH () represents the nvGRAPH function of the CUDA platform, namely, the shortest path of the missing road section is calculated in real time by heterogeneous parallel calculation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111546535.5A CN114170696A (en) | 2021-12-16 | 2021-12-16 | Real-time toll calculation system and method for differential charging of expressway |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111546535.5A CN114170696A (en) | 2021-12-16 | 2021-12-16 | Real-time toll calculation system and method for differential charging of expressway |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114170696A true CN114170696A (en) | 2022-03-11 |
Family
ID=80487008
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111546535.5A Pending CN114170696A (en) | 2021-12-16 | 2021-12-16 | Real-time toll calculation system and method for differential charging of expressway |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114170696A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973675A (en) * | 2022-05-26 | 2022-08-30 | 华南理工大学 | Highway turn-around organization system utilizing service area communication channel |
| CN115240414A (en) * | 2022-07-19 | 2022-10-25 | 陕西蓝德智慧交通科技有限公司 | Traffic condition investigation method based on ETC portal system |
| CN115631544A (en) * | 2022-12-14 | 2023-01-20 | 高德软件有限公司 | Method and device for determining differential charging of expressway and computer equipment |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101894051A (en) * | 2010-07-29 | 2010-11-24 | 中国科学技术大学 | Primary and secondary data structure-based CPU-GPU cooperative computing method |
| CN102194259A (en) * | 2010-03-16 | 2011-09-21 | 上海交技发展股份有限公司 | Method for computing toll by modeling networked tolling lane of highway based on graph theory |
| CN102932003A (en) * | 2012-09-07 | 2013-02-13 | 上海交通大学 | Accelerated QC-LDPC (Quasi-Cyclic Low-Density Parity-Check Code) decoding method based on GPU (Graphics Processing Unit) framework |
| CN110853163A (en) * | 2019-10-28 | 2020-02-28 | 北京中交国通智能交通系统技术有限公司 | AI and vehicle road cooperation based vehicle path identification charging system and method |
| CN110992499A (en) * | 2019-11-20 | 2020-04-10 | 北京握奇智能科技有限公司 | High-speed charging method and system based on block chain |
| CN111784857A (en) * | 2020-06-22 | 2020-10-16 | 浙江大华技术股份有限公司 | Parking space management method and device and computer storage medium |
-
2021
- 2021-12-16 CN CN202111546535.5A patent/CN114170696A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102194259A (en) * | 2010-03-16 | 2011-09-21 | 上海交技发展股份有限公司 | Method for computing toll by modeling networked tolling lane of highway based on graph theory |
| CN101894051A (en) * | 2010-07-29 | 2010-11-24 | 中国科学技术大学 | Primary and secondary data structure-based CPU-GPU cooperative computing method |
| CN102932003A (en) * | 2012-09-07 | 2013-02-13 | 上海交通大学 | Accelerated QC-LDPC (Quasi-Cyclic Low-Density Parity-Check Code) decoding method based on GPU (Graphics Processing Unit) framework |
| CN110853163A (en) * | 2019-10-28 | 2020-02-28 | 北京中交国通智能交通系统技术有限公司 | AI and vehicle road cooperation based vehicle path identification charging system and method |
| CN110992499A (en) * | 2019-11-20 | 2020-04-10 | 北京握奇智能科技有限公司 | High-speed charging method and system based on block chain |
| CN111784857A (en) * | 2020-06-22 | 2020-10-16 | 浙江大华技术股份有限公司 | Parking space management method and device and computer storage medium |
Non-Patent Citations (4)
| Title |
|---|
| 卢香清: ""图最短路径并行化及其应用研究"", 《计算机工程与应用》 * |
| 周灿: ""基于MPI的矩阵运算并行算法研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李承远: ""取消省界收费站后的高速公路通行费用实时计算方案优化研究"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
| 武亮亮: ""基于MPI和Linux机群环境的矩阵运算并行算法应用研究"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973675A (en) * | 2022-05-26 | 2022-08-30 | 华南理工大学 | Highway turn-around organization system utilizing service area communication channel |
| CN114973675B (en) * | 2022-05-26 | 2023-09-26 | 华南理工大学 | Expressway turning organization system utilizing service area contact channel |
| CN115240414A (en) * | 2022-07-19 | 2022-10-25 | 陕西蓝德智慧交通科技有限公司 | Traffic condition investigation method based on ETC portal system |
| CN115240414B (en) * | 2022-07-19 | 2024-04-02 | 陕西蓝德智慧交通科技有限公司 | Traffic condition investigation method based on ETC portal system |
| CN115631544A (en) * | 2022-12-14 | 2023-01-20 | 高德软件有限公司 | Method and device for determining differential charging of expressway and computer equipment |
| CN115631544B (en) * | 2022-12-14 | 2023-02-28 | 高德软件有限公司 | Method and device for determining differential charging of expressway and computer equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114170696A (en) | Real-time toll calculation system and method for differential charging of expressway | |
| CN112700072B (en) | Traffic condition prediction method, electronic device, and storage medium | |
| US20190279247A1 (en) | Crowd-sourced mapping | |
| CN101676689B (en) | Driving evaluation system and driving evaluation method | |
| CN112069573B (en) | City group space simulation method, system and equipment based on cellular automaton | |
| Zhang et al. | Smart transport system based on “The Internet of Things” | |
| CN112734956B (en) | ETC portal determination method and device and storage medium | |
| CN112863172A (en) | Highway traffic running state judgment method, early warning method, device and terminal | |
| CN111914940B (en) | Shared vehicle station clustering method, system, device and storage medium | |
| Yue et al. | Low-cost urban test scenario generation using microscopic traffic simulation | |
| CN112249009A (en) | Vehicle speed control method, device and system and electronic equipment | |
| Rieck et al. | Efficient traffic simulator coupling in a distributed v2x simulation environment | |
| CN112885112A (en) | Vehicle driving detection method, vehicle driving early warning method and device | |
| Hou et al. | The effect of the dataset on evaluating urban traffic prediction | |
| Zou et al. | The method of dynamic identification of the maximum speed limit of expressway based on electronic toll collection data | |
| CN114509081B (en) | Method and device for calculating vehicle passing cost based on differential charging | |
| CN109979198B (en) | Urban expressway vehicle speed discrete identification method based on large-scale floating vehicle data | |
| CN114913492A (en) | Intelligent vehicle road system scene generation method and system, storage medium and computing equipment | |
| CN113177046B (en) | Road network topological graph generation method, device, equipment and storage medium | |
| Ali et al. | Future connected vehicles: challenges and opportunities for spatio-temporal computing | |
| Yu et al. | GPS data in urban bicycle-sharing: Dynamic electric fence planning with assessment of resource-saving and potential energy consumption increasement | |
| Liu et al. | New generation of smart highway: Framework and insights | |
| Li et al. | Game dynamics of route choice in heterogenous population | |
| CN111739288A (en) | Vehicle driving risk analysis method based on artificial intelligence and related device | |
| Fan et al. | Multimodal traffic travel time prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |