A method and system for dynamic configuration of toll station lanes based on data collaboration

Abstract

The present invention relates to a method and system for dynamically configuring lanes at toll stations based on data collaboration, and belongs to the field of traffic control technology. The method comprises: in a network security data information transmission environment, a cloud computing server is used to intelligently predict the total number of outbound vehicles of different charging types in the current time segment of a target toll station based on the basic data collected by the Internet of Things sensor according to the intelligent traffic prediction model built by the edge computing network element, and then configure the number of channels of different charging types in advance; the present invention also relates to a system for dynamically configuring lanes at toll stations based on data collaboration. Through the present invention, the technical problem that the lack of basic data when the electronic toll channels and manual toll channels of toll stations are dynamically configured at any time in the prior art leads to poor traffic control effects, and the number of toll channels of different types can be configured in advance according to the total number of outbound vehicles of different charging types intelligently predicted, thereby solving the above technical problems.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic control, and in particular to a method and system for dynamically configuring lanes at a toll booth based on data collaboration. Background Art

[0002] In the traffic control of road vehicles, the traffic control management party hopes to integrate resources and improve the utilization rate of existing resources to improve the traffic efficiency as much as possible. For example, for the traffic control of each toll station on the highway, since the toll channels of the toll station are limited and have been limited in number when they are designed, general traffic management control can only roughly allocate a part of the existing toll channels as electronic toll channels, such as ETC toll channels, and the rest as manual toll channels based on manual experience. In this way, the number of electronic toll channels and manual toll channels at each toll station is also fixed, and the number will not change dynamically according to changes in time.

[0003] For example, a Chinese invention patent with publication number CN117197913A proposes a cloud computing highway ETC toll data management and analysis method, which has classification number G08G. The method includes current toll information collection, current congestion analysis of toll stations, information collection of vehicles to be dispatched, information collection of toll stations, extraction of historical toll data and vehicle dispatch analysis confirmation; the present invention calculates the current congestion of the target toll station, and when the congestion is greater than the set value, calculates the dispatch adaptation degree by combining the information of vehicles to be dispatched, toll station information and historical toll data, thereby performing vehicle dispatch, reducing the error in vehicle dispatch analysis, improving the comprehensiveness and persuasiveness of vehicle dispatch analysis, and providing a more solid support basis for the confirmation of vehicle dispatch adaptation results, thereby improving the traffic efficiency of highways, and reducing traffic congestion, thereby achieving optimization of vehicle dispatch.

[0004] A Chinese invention patent with publication number CN116403407A discloses a method, device and medium for dynamic configuration and induction of lanes at a highway toll station. The classification number is G08G, and it relates to the field of traffic management and control. The method includes: determining a road gantry at a toll station square, obtaining road data and square fleet data through the road gantry, and saving the road data and square fleet data to form historical data; obtaining the type of vehicle in front of the toll station square through the road gantry, and performing data prediction and analysis on the historical data to obtain flow parameters corresponding to the vehicle type; determining the lane type of the toll station square, and determining the lane opening index based on the flow parameters, and determining the number of lanes opened corresponding to the lane type based on the lane opening index; and inducing the vehicle according to the vehicle type and the lane opening index.

[0005] It can be seen that the above-mentioned existing technology is only limited to executing data management-based scheduling processing or on-site control of lane traffic control mechanisms at toll stations, and is unable to reliably predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment of the toll station and the various configuration information of the toll station for the future time segments of the toll station. As a result, the traffic control management party of the toll station lacks effective basic data when it hopes to dynamically configure the number of electronic toll channels and the number of manual toll channels for future time segments, and is unable to maximize the vehicle traffic rate that meets different toll requirements, resulting in the inability to further improve the traffic speed of the toll station. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a method and system for dynamic configuration of toll station lanes based on data collaboration, which can use an encrypted network to ensure the network security of the transmission of various digital information, and apply the cloud-edge data collaborative management mechanism with network security performance to the subdivided field of toll stations. Based on the data collected by the Internet of Things sensor, a customized intelligent traffic prediction model is used to synchronously predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment of the target toll station and the various configuration information of the target toll station. The total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station in the current time segment as a future time segment are predicted based on the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment of the target toll station, and the various configuration information of the target toll station, and the number of electronic toll channels and the number of manual toll channels of the target toll station in the current time segment are configured in advance based on the prediction results, so as to make full use of the limited toll channels and maximize the traffic efficiency of vehicles with various toll demand types.

[0007] According to a first aspect of the present invention, a method for dynamic configuration of lanes in a toll booth based on data collaboration is provided, the method comprising:

[0008] An IoT sensor installed at the target toll station is used to obtain the total number of vehicles leaving the station for each electronic toll collection and the total number of vehicles leaving the station for each manual toll collection corresponding to each past time segment before the current time segment;

[0009] A first edge computing network element is provided near a target toll station, and is used to output a feedforward neural network after completing multiple learnings as an intelligent traffic prediction model;

[0010] A second edge computing network element is used to be set near the target toll station, which is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station;

[0011] A cloud computing server disposed at a remote end of a target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station within the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles within the current time segment are used as a first predicted number and a second predicted number, respectively;

[0012] A cloud computing server set up at the remote end of the target toll station is used to configure in advance the number of electronic toll channels and the number of manual toll channels of the target toll station in the current time segment based on the total number of exit channels of the target toll station, the first predicted number and the second predicted number.

[0013] According to a second aspect of the present invention, a toll booth lane dynamic configuration system based on data collaboration is provided, the system comprising:

[0014] An IoT sensor is installed at the target toll station to obtain the total number of vehicles leaving the station for each electronic toll collection and the total number of vehicles leaving the station for each manual toll collection corresponding to each past time segment before the current time segment;

[0015] The first edge computing network element is set near the target toll station and is used to output the feedforward neural network after completing multiple learning as an intelligent traffic prediction model;

[0016] The second edge computing network element is set near the target toll station, and is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station;

[0017] A cloud computing server is arranged at a remote end of a target toll station, and is used to intelligently predict the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles of the target toll station in the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and use the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment as a first predicted number and a second predicted number, respectively;

[0018] The cloud computing server is also used to configure the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in advance within the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number.

[0019] Compared with the prior art, the present invention has at least the following six key inventive features:

[0020] First: A data collaborative management mechanism based on the cloud-edge terminal of the Internet of Things is adopted for the target toll station. The IoT sensor is used to perform local data sensing operations at the target toll station. The first edge computing network element and the second edge computing network element set near the target toll station are used to respectively execute the construction of the intelligent traffic prediction model and the analysis of various configuration information of the target toll station. The cloud computing server set at the remote end of the target toll station is used to use the intelligent traffic prediction model to intelligently predict the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles at the target toll station in the current time segment as the future time segment, and then configure the number of electronic toll collection channels and the number of manual toll collection channels in the current time segment, thereby completing the data collaborative management of the target toll station based on the cloud-edge terminal of the Internet of Things;

[0021] Second: Specifically, a cloud computing server set up at the remote end of the target toll station is used to configure the number of electronic toll channels and the number of manual toll channels of the target toll station in the future time segment in advance based on the total number of exit channels of the target toll station, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the future time segment, wherein the sum of the number of electronic toll channels and the number of manual toll exit vehicles of the target toll station configured in advance in the future time segment is equal to the total number of exit channels of the target toll station, and the ratio of the total number of electronic toll exit vehicles to the total number of manual toll exit vehicles in the future time segment is equal to the ratio of the number of electronic toll channels and the number of manual toll channels of the target toll station configured in advance in the current time segment, thereby improving the intelligent level of target toll station management;

[0022] Third: In order to ensure the network security of the transmission of digital information on the cloud edge, the cloud computing server transmits digital information with the IoT sensor, the first edge computing network element, and the second edge computing network element respectively through the encrypted network. The encrypted network is used to provide a network security environment for the transmission of digital information.

[0023] Fourthly, in order to intelligently predict the total number of outgoing vehicles with different toll forms at the target toll station in the future time segment, a customized intelligent traffic prediction model is constructed. The structure customization of the intelligent traffic prediction model is as follows: the intelligent traffic prediction model is a feedforward neural network that has completed multiple learnings. The number of learning times of the feedforward neural network is positively correlated with the distance between the target toll station and the nearest other toll stations, thereby ensuring the effectiveness and stability of the intelligent prediction results;

[0024] Fifth: In order to intelligently predict the total number of outgoing vehicles of the target toll station in different charging modes in the future time segment, a number of basic information is specifically selected, including the duration of each time segment, the total number of outgoing vehicles of each electronic toll collection and the total number of outgoing vehicles of each manual toll collection corresponding to each past time segment before the future time segment, and various configuration information of the target toll station. The various configuration information of the target toll station includes the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of outgoing channels of the target toll station, and the width of the narrowest entrance of the target toll station. The comprehensive and full selection of the above-mentioned basic information further ensures the effectiveness and stability of the intelligent prediction results;

[0025] Sixth point: In each learning execution of the feedforward neural network, the total number of electronic toll collection vehicles and the total number of manual toll collection vehicles at the known target toll station within a certain time segment are used as the output content of the feedforward neural network, and the duration of each time segment, the total number of electronic toll collection vehicles and the total number of manual toll collection vehicles corresponding to each past time segment before the certain time segment, and various configuration information of the target toll station are used as the input content of the feedforward neural network to complete this learning, thereby ensuring the learning effect of each learning of the feedforward neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0027] Figure 1 It is a technical flow chart of a method and system for dynamic configuration of toll station lanes based on data collaboration according to the present invention.

[0028] Figure 2 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 1 of the present invention.

[0029] Figure 3 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 2 of the present invention.

[0030] Figure 4 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 3 of the present invention.

[0031] Figure 5 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 4 of the present invention.

[0032] Figure 6It is a schematic diagram of the structure of a toll booth lane dynamic configuration system based on data collaboration according to Embodiment 5 of the present invention. DETAILED DESCRIPTION

[0033] like Figure 1 As shown, a technical flow chart of a toll station lane dynamic configuration method and system based on data collaboration according to the present invention is given.

[0034] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0035] Technical process A: Establish a cloud-edge data collaborative management architecture for target toll station management, taking into account the advantages of data processing speed and saving computing resources on both the cloud and local ends. The management architecture adopts an encrypted network to ensure the network security of the transmission of various digital information;

[0036] Specifically, in the data collaborative management architecture, there are local IoT sensors, a first edge computing network element and a second edge computing network element near the local area, and a remote cloud computing server;

[0037] More specifically, the IoT sensor is used to perform local data sensing operations at the target toll station, the first edge computing network element and the second edge computing network element arranged near the target toll station are respectively used to execute the construction of the intelligent traffic prediction model and the analysis of various configuration information of the target toll station, and the cloud computing server arranged at the remote end of the target toll station is used to adopt the intelligent traffic prediction model to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station in the current time segment as the future time segment, and then configure the number of electronic toll lanes and the number of manual toll lanes in the current time segment, thereby completing the target toll station data collaborative management based on the cloud-edge end of the IoT;

[0038] Technical process B: The first edge computing network element is used to design a customized intelligent traffic prediction model to provide a model basis for the intelligent prediction of the number of electronic toll lanes and manual toll lanes in the current time segment at the subsequent target toll station;

[0039] For example, the structure customization of the intelligent traffic prediction model is mainly reflected in the following aspects:

[0040] First aspect: the intelligent traffic prediction model is a feedforward neural network that has completed multiple learnings;

[0041] Second aspect: the number of learning times of the feedforward neural network is positively correlated with the distance between the target toll station and the nearest other toll stations;

[0042] Third aspect: In each learning of the feedforward neural network, the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles at the target toll station in a certain time segment are used as the output content of the feedforward neural network, and the duration of each time segment, the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles corresponding to each past time segment before the certain time segment, and various configuration information of the target toll station are used as the input content of the feedforward neural network to complete this learning, thereby ensuring the learning effect of each learning of the feedforward neural network;

[0043] In this way, through the structural customization in the above-mentioned multiple ways, the effectiveness and stability of the intelligent prediction results of the intelligent traffic prediction model are guaranteed;

[0044] Technical process C: The IoT sensors at the target toll station are used to perform data sensing operations at the target toll station, and the second edge computing network element set near the target toll station is used to parse various configuration information of the target toll station, thereby providing sufficient and comprehensive basic information for the subsequent intelligent prediction of the number of electronic toll lanes and manual toll lanes at the target toll station in the current time segment;

[0045] Specifically, the fully comprehensive multiple basic information includes the duration of each time segment, the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each past time segment before the future time segment, and various configuration information of the target toll station, and the various configuration information of the target toll station is the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station;

[0046] In this way, the comprehensive and sufficient screening of the above-mentioned multiple basic information further ensures the effectiveness and stability of the intelligent prediction results;

[0047] Technical process D: Adopt the cloud-edge data collaborative management architecture based on encrypted network of technical process A, use the intelligent traffic prediction model with customized structure of technical process B, and adopt multiple basic information selected by technical process C to perform intelligent prediction of the number of electronic toll lanes and manual toll lanes in the current time segment of the target toll station;

[0048] Technical process E: A cloud computing server set up at a remote end of a target toll station is used to configure the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the future time segment in advance based on the total number of exit channels of the target toll station, the total number of electronic toll collection exit vehicles and the total number of manual toll collection exit vehicles in the future time segment;

[0049] Specifically, the sum of the number of electronic toll collection channels and the number of manual toll collection channels of the pre-configured target toll station in the future time segment is equal to the total number of exit channels of the target toll station, and the ratio of the total number of electronic toll collection exit vehicles to the total number of manual toll collection exit vehicles in the future time segment is equal to the ratio of the number of electronic toll collection channels to the number of manual toll collection channels of the pre-configured target toll station in the current time segment;

[0050] In summary, through the intelligent prediction results of the number of passing vehicles with different toll demand types in the target toll station in the future time segments, the number of channels with different toll types in the target toll station in the future time segments is dynamically adjusted, thereby improving the utilization rate of the limited number of channels at the target toll station while speeding up the overall passage speed of the target toll station as much as possible, thereby upgrading the intelligent level of target toll station management.

[0051] The key points of the present invention are: a cloud-edge data collaborative management architecture based on an encrypted network to ensure network security while taking into account the advantages of data processing speed and saving computing resources on both the cloud and local ends, a customized intelligent traffic prediction model, targeted screening of multiple basic information, and advance configuration of the number of electronic toll channels and manual toll channels at the target toll station in future time segments.

[0052] Below, a method and system for dynamic configuration of lanes at a toll booth based on data collaboration of the present invention will be specifically described in the form of an embodiment.

[0053] Example 1

[0054] Figure 2 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 1 of the present invention.

[0055] like Figure 2 As shown, the toll station lane dynamic configuration method based on data collaboration includes the following specific steps:

[0056] Step S201: Using an IoT sensor installed at a target toll station to obtain the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles corresponding to each past time segment before the current time segment;

[0057] For example, obtaining the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each of the past time segments before the current time segment includes: visually distinguishing the electronic toll-collecting exit vehicles and the manual toll-collecting exit vehicles based on different appearance imaging characteristics of the electronic toll-collecting exit vehicles and the manual toll-collecting exit vehicles and / or visual recognition results of whether the electronic toll-collecting device is present on the vehicle body;

[0058] Step S202: using a first edge computing network element disposed near a target toll station to output a feedforward neural network after multiple learnings as an intelligent traffic prediction model;

[0059] For example, outputting the feedforward neural network after multiple learning as the intelligent traffic prediction model includes: using a MATLAB toolbox to complete simulation and testing of a model construction process of outputting the feedforward neural network after multiple learning as the intelligent traffic prediction model;

[0060] Step S203: A second edge computing network element is used to be arranged near the target toll station, and is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents of the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station;

[0061] Specifically, the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station are output as various configuration information of the target toll station: different data collection terminals can be used to respectively collect the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station;

[0062] Step S204: A cloud computing server disposed at a remote end of the target toll station is used to intelligently predict the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment of the target toll station according to the intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment are used as the first predicted number and the second predicted number, respectively;

[0063] For example, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment of the target toll station are intelligently predicted based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station according to the intelligent traffic prediction model, and the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment are used as the first predicted number and the second predicted number respectively, including: the duration of each time segment can be selected as 15 minutes, i.e., a quarter of an hour;

[0064] Step S205: using a cloud computing server located at a remote end of the target toll station to configure in advance the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number;

[0065] For example, configuring the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment in advance based on the total number of exit channels of the target toll station, the first predicted number, and the second predicted number includes: the current time segment is actually a future time segment, which takes the current moment as the starting moment;

[0066] The cloud computing server transmits digital information with the IoT sensor, the first edge computing network element, and the second edge computing network element through an encrypted network, and the encrypted network is used to provide a network security environment for the transmission of digital information.

[0067] Wherein, a cloud computing server disposed at a remote end of a target toll station is used to configure in advance the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number, including: the sum of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment is equal to the total number of outbound channels of the target toll station, and the ratio of the first predicted number to the second predicted number is equal to the ratio of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment;

[0068] Among them, an IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, including: the number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, and the target toll station and other toll stations near the target toll station are at different positions corresponding to different mileages of the same highway;

[0069] For example, the number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, and the different positions corresponding to different mileages of the target toll station and other toll stations near the target toll station on the same highway include: when the current time segment is from 6:00 p.m. to 6:15 p.m., the number of other toll stations near the target toll station is 3, the number of time segments of each past time segment before the current time segment is 6, the number of other toll stations near the target toll station is 4, the number of time segments of each past time segment before the current time segment is 8, and when the number of other toll stations near the target toll station is 5, the number of time segments of each past time segment before the current time segment is 10, and so on;

[0070] Among them, the IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, and also includes: each past time segment before the current time segment and the current time segment form a complete time interval and the duration of each time segment is equal;

[0071] For example, each past time segment before the current time segment and the current time segment form a complete time interval and the duration of each time segment is equal, including: the current time segment is from 6:00 p.m. to 6:15 p.m., the number of other toll stations near the target toll station is 3, and the number of time segments of each past time segment before the current time segment is 6, namely, 5:45 p.m. to 6:00 p.m., 5:30 p.m. to 5:45 p.m., 5:15 p.m. to 5:30 p.m., 5:00 p.m. to 5:15 p.m., 4:45 p.m. to 5:00 p.m., and 4:30 p.m. to 4:45 p.m.;

[0072] Among them, a first edge computing network element arranged near a target toll station is used to output a feedforward neural network after completing multiple learning as an intelligent traffic prediction model, including: taking the distance between the target toll station and the nearest other toll station as a reference distance, and the number of learning times of the feedforward neural network is positively correlated with the value of the reference distance;

[0073] For example, taking the distance between the target toll station and the nearest other toll station as the reference distance, the positive correlation between the learning times of the feedforward neural network and the value of the reference distance includes: when the distance between the target toll station and the nearest other toll station is 1 km, the learning times of the feedforward neural network is 100 times, when the distance between the target toll station and the nearest other toll station is 3 km, the learning times of the feedforward neural network is 150 times, when the distance between the target toll station and the nearest other toll station is 5 km, the learning times of the feedforward neural network is 200 times, and so on;

[0074] The number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, including: the number of other toll stations within the geographical range covered by a circle with the target toll station as the center and the set radius value as the radius is the number of other toll stations near the target toll station;

[0075] For example, the number of other toll stations within the geographical range covered by a circle with the target toll station as the center and the set radius value as the radius is the number of other toll stations near the target toll station, including: the set radius value may be 3 kilometers;

[0076] And wherein, a first edge computing network element arranged near the target toll station is used to output the feedforward neural network after completing multiple learning as an intelligent passage prediction model, which also includes: in each learning performed on the feedforward neural network, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the known target toll station in a certain time segment are used as the output content of the feedforward neural network, and the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the certain time segment, and various configuration information of the target toll station are used as the input content of the feedforward neural network to complete this learning.

[0077] Example 2

[0078] Figure 3 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 2 of the present invention.

[0079] like Figure 3 As shown, Figure 2 Different from the embodiment in, in the method for dynamic configuration of toll station lanes based on data collaboration, before using the Internet of Things sensor set at the target toll station to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, that is, before step S201, the method further includes:

[0080] Step S301: using a cloud computing server to remotely configure various working parameters of the IoT sensor installed at the target toll station;

[0081] Specifically, the cloud computing server is used to remotely configure the various working parameters of the Internet of Things sensors set at the target toll station, including: the cloud computing server performs remote configuration of the various working parameters of the Internet of Things sensors set at the target toll station through an encrypted network to ensure the network security of the configuration data.

[0082] Example 3

[0083] Figure 4 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 3 of the present invention.

[0084] like Figure 4 As shown, Figure 2 Different from the embodiment in, in the method for dynamic configuration of toll station lanes based on data collaboration, after adopting a first edge computing network element arranged near the target toll station to output a feedforward neural network after completing multiple learning as an intelligent traffic prediction model, that is, after step S202, the method further includes:

[0085] Step S401: using a data storage device to store various model parameters of the intelligent traffic prediction model to complete model storage of the intelligent traffic prediction model;

[0086] For example, using a data storage device to store various model parameters of the intelligent traffic prediction model to complete model storage of the intelligent traffic prediction model includes: the data storage device is a FLASH storage chip, an MMC storage chip or a TF storage chip.

[0087] Example 4

[0088] Figure 5 The present invention is a flowchart showing a method for dynamically configuring lanes in a toll booth based on data collaboration according to Embodiment 4 of the present invention.

[0089] like Figure 5 As shown, Figure 2 Different from the embodiments in the embodiment, in the toll station lane dynamic configuration method based on data collaboration, after using a cloud computing server set at the remote end of the target toll station to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles of the target toll station in the current time segment according to the intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and taking the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment as the first predicted number and the second predicted number, that is, after step S204, the method further includes:

[0090] Step S501: using an on-site display device to receive a first predicted number and a second predicted number and to realize on-site synchronous display of the first predicted number and the second predicted number;

[0091] Specifically, using an on-site display device to receive the first predicted number and the second predicted number and realize on-site synchronous display of the first predicted number and the second predicted number includes: the on-site display device can be selected as an LED display array, an LCD display array or a giant screen display.

[0092] Next, various method embodiments of the present invention are described in detail.

[0093] In the toll booth lane dynamic configuration method based on data collaboration according to various method embodiments of the present invention:

[0094] A second edge computing network element is used to be set near the target toll station, which is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station, including: each exit channel of the target toll station has two different configurations: an electronic toll channel and a manual toll channel;

[0095] Specifically, each exit channel of each toll station on the expressway has two different configurations: electronic toll channel and manual toll channel. The number of electronic toll channels and manual toll channels in each time segment of each toll station can be flexibly configured to match the number of vehicles with different toll demand types in each time segment.

[0096] Among them, the IoT sensor set at the target toll station is used to obtain the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles corresponding to each past time segment before the current time segment, including: the IoT sensor is a visual sensor based on the IoT;

[0097] Among them, an Internet of Things sensor installed at the target toll station is used to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment. It also includes: the visual sensor based on the Internet of Things has various working parameters, and the various working parameters include shooting frame rate, imaging focal length, resolution, exposure time and aperture value.

[0098] In the toll booth lane dynamic configuration method based on data collaboration according to various method embodiments of the present invention:

[0099] Taking the distance between the target toll station and the nearest other toll stations as the reference distance, the number of learning times of the feedforward neural network and the numerical value of the reference distance are positively correlated, including: using a numerical conversion function to represent the numerical conversion relationship of the number of learning times of the feedforward neural network and the numerical value of the reference distance;

[0100] Among them, the numerical conversion relationship that uses a numerical conversion function to represent the positive correlation between the number of learning times of the feedforward neural network and the numerical value of the reference distance includes: in the numerical conversion function, the number of learning times of the feedforward neural network corresponding to the numerical value of the reference distance is the output data of the numerical conversion function, and the numerical value of the reference distance is the input data of the numerical conversion function.

[0101] And in the toll booth lane dynamic configuration method based on data collaboration according to various method embodiments of the present invention:

[0102] A cloud computing server disposed at a remote end of a target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles of the target toll station in the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment are used as the first prediction number and the second prediction number respectively, including: the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station are input into the intelligent traffic prediction model in parallel;

[0103] And wherein, a cloud computing server set at the remote end of the target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment of the target toll station according to the intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment are used as the first predicted number and the second predicted number respectively. It also includes: running the intelligent traffic prediction model to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment of the target toll station output by the intelligent traffic prediction model.

[0104] Example 5

[0105] Figure 6 It is a schematic diagram of the structure of a toll booth lane dynamic configuration system based on data collaboration according to Embodiment 5 of the present invention.

[0106] like Figure 6 As shown, the toll station lane dynamic configuration system based on data collaboration includes the following components:

[0107] An IoT sensor is installed at the target toll station to obtain the total number of vehicles leaving the station for each electronic toll collection and the total number of vehicles leaving the station for each manual toll collection corresponding to each past time segment before the current time segment;

[0108] For example, obtaining the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each of the past time segments before the current time segment includes: visually distinguishing the electronic toll-collecting exit vehicles and the manual toll-collecting exit vehicles based on different appearance imaging characteristics of the electronic toll-collecting exit vehicles and the manual toll-collecting exit vehicles and / or visual recognition results of whether the electronic toll-collecting device is present on the vehicle body;

[0109] The first edge computing network element is set near the target toll station and is used to output the feedforward neural network after completing multiple learning as an intelligent traffic prediction model;

[0110] For example, outputting the feedforward neural network after multiple learning as the intelligent traffic prediction model includes: using a MATLAB toolbox to complete simulation and testing of a model construction process of outputting the feedforward neural network after multiple learning as the intelligent traffic prediction model;

[0111] The second edge computing network element is set near the target toll station, and is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station;

[0112] Specifically, the total mileage of the expressway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station are output as various configuration information of the target toll station: different data collection terminals can be used to respectively collect the total mileage of the expressway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station;

[0113] A cloud computing server is arranged at a remote end of a target toll station, and is used to intelligently predict the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles of the target toll station in the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and use the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment as a first predicted number and a second predicted number, respectively;

[0114] For example, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment of the target toll station are intelligently predicted based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station according to the intelligent traffic prediction model, and the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment are used as the first predicted number and the second predicted number respectively, including: the duration of each time segment can be selected as 15 minutes, i.e., a quarter of an hour;

[0115] The cloud computing server is also used to configure the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in advance within the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number;

[0116] For example, configuring the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment in advance based on the total number of exit channels of the target toll station, the first predicted number, and the second predicted number includes: the current time segment is actually a future time segment, which takes the current moment as the starting moment;

[0117] The cloud computing server transmits digital information with the IoT sensor, the first edge computing network element, and the second edge computing network element through an encrypted network, and the encrypted network is used to provide a network security environment for the transmission of digital information.

[0118] Wherein, a cloud computing server disposed at a remote end of a target toll station is used to configure in advance the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number, including: the sum of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment is equal to the total number of outbound channels of the target toll station, and the ratio of the first predicted number to the second predicted number is equal to the ratio of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment;

[0119] Among them, an IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, including: the number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, and the target toll station and other toll stations near the target toll station are at different positions corresponding to different mileages of the same highway;

[0120] For example, the number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, and the different positions corresponding to different mileages of the target toll station and other toll stations near the target toll station on the same highway include: when the current time segment is from 6:00 p.m. to 6:15 p.m., the number of other toll stations near the target toll station is 3, the number of time segments of each past time segment before the current time segment is 6, the number of other toll stations near the target toll station is 4, the number of time segments of each past time segment before the current time segment is 8, and when the number of other toll stations near the target toll station is 5, the number of time segments of each past time segment before the current time segment is 10, and so on;

[0121] Among them, the IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, and also includes: each past time segment before the current time segment and the current time segment form a complete time interval and the duration of each time segment is equal;

[0122] For example, each past time segment before the current time segment and the current time segment form a complete time interval and the duration of each time segment is equal, including: the current time segment is from 6:00 p.m. to 6:15 p.m., the number of other toll stations near the target toll station is 3, and the number of time segments of each past time segment before the current time segment is 6, namely, 5:45 p.m. to 6:00 p.m., 5:30 p.m. to 5:45 p.m., 5:15 p.m. to 5:30 p.m., 5:00 p.m. to 5:15 p.m., 4:45 p.m. to 5:00 p.m., and 4:30 p.m. to 4:45 p.m.;

[0123] Among them, a first edge computing network element arranged near a target toll station is used to output a feedforward neural network after completing multiple learning as an intelligent traffic prediction model, including: taking the distance between the target toll station and the nearest other toll station as a reference distance, and the number of learning times of the feedforward neural network is positively correlated with the value of the reference distance;

[0124] For example, taking the distance between the target toll station and the nearest other toll station as the reference distance, the positive correlation between the learning times of the feedforward neural network and the value of the reference distance includes: when the distance between the target toll station and the nearest other toll station is 1 km, the learning times of the feedforward neural network is 100 times, when the distance between the target toll station and the nearest other toll station is 3 km, the learning times of the feedforward neural network is 150 times, when the distance between the target toll station and the nearest other toll station is 5 km, the learning times of the feedforward neural network is 200 times, and so on;

[0125] The number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, including: the number of other toll stations within the geographical range covered by a circle with the target toll station as the center and the set radius value as the radius is the number of other toll stations near the target toll station;

[0126] For example, the number of other toll stations within the geographical range covered by a circle with the target toll station as the center and the set radius value as the radius is the number of other toll stations near the target toll station, including: the set radius value may be 3 kilometers;

[0127] And wherein, a first edge computing network element arranged near the target toll station is used to output the feedforward neural network after completing multiple learning as an intelligent passage prediction model, which also includes: in each learning performed on the feedforward neural network, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the known target toll station in a certain time segment are used as the output content of the feedforward neural network, and the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the certain time segment, and various configuration information of the target toll station are used as the input content of the feedforward neural network to complete this learning.

[0128] In addition, the present invention may also cite the following technical contents to highlight the significant technical progress of the present invention:

[0129] The method of inputting the duration of each time segment, the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station into the intelligent traffic prediction model in parallel includes: using a parallel driving interface to input the duration of each time segment, the total number of electronic toll-collecting vehicles and the total number of manual toll-collecting vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station into the intelligent traffic prediction model in parallel;

[0130] For example, the parallel drive interface is used to input the duration of each time segment, the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station into the intelligent traffic prediction model in parallel, including: the parallel drive interface is a programmable logic device;

[0131] Specifically, the duration of each time segment input into the intelligent traffic prediction model in parallel, the total number of vehicles leaving the station through electronic toll collection and the total number of vehicles leaving the station through manual toll collection corresponding to each past time segment before the current time segment, and various configuration information of the target toll station are respectively represented by octal numerical values;

[0132] And wherein, running the intelligent traffic prediction model to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station output by the intelligent traffic prediction model within the current time segment includes: the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station output by the intelligent traffic prediction model within the current time segment are both octal numerical data.

[0133] Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limitations on the present application. Ordinary technicians in this field can change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for dynamic configuration of toll station lanes based on data collaboration, characterized in that: The method comprises: An IoT sensor installed at the target toll station is used to obtain the total number of vehicles leaving the station for each electronic toll collection and the total number of vehicles leaving the station for each manual toll collection corresponding to each past time segment before the current time segment; A first edge computing network element is provided near a target toll station, and is used to output a feedforward neural network after completing multiple learnings as an intelligent traffic prediction model; A second edge computing network element is used to be set near the target toll station, which is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station; A cloud computing server disposed at a remote end of a target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the target toll station within the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles within the current time segment are used as a first predicted number and a second predicted number, respectively; A cloud computing server set up at the remote end of the target toll station is used to configure in advance the number of electronic toll channels and the number of manual toll channels of the target toll station in the current time segment based on the total number of exit channels of the target toll station, the first predicted number and the second predicted number.

2. The method for dynamic configuration of toll booth lanes based on data collaboration according to claim 1, characterized in that: The cloud computing server disposed at the remote end of the target toll station is used to configure the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in the current time segment in advance based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number, including: the sum of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment is equal to the total number of outbound channels of the target toll station, and the ratio of the first predicted number to the second predicted number is equal to the ratio of the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station configured in advance in the current time segment; The IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, including: the number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, and the target toll station and other toll stations near the target toll station are located at different positions corresponding to different mileages of the same highway; The IoT sensor set at the target toll station is used to obtain the total number of outbound vehicles corresponding to each past time segment before the current time segment, and further includes: each past time segment before the current time segment and the current time segment form a complete time interval and the duration of each time segment is equal; Among them, the cloud computing server executes the transmission of digital information with the Internet of Things sensor, the first edge computing network element and the second edge computing network element respectively through the encrypted network, and the encrypted network is used to provide a network security environment for the transmission of digital information.

3. The method for dynamic configuration of toll booth lanes based on data collaboration as claimed in claim 2, characterized in that: The first edge computing network element arranged near the target toll station is used to output the feedforward neural network after completing multiple learning as the intelligent traffic prediction model, including: taking the distance between the target toll station and the nearest other toll station as the reference distance, and the number of learning times of the feedforward neural network is positively correlated with the value of the reference distance; The number of time segments of each past time segment before the current time segment is proportional to the number of other toll stations near the target toll station, including: the number of other toll stations within the geographical range covered by a circle with the target toll station as the center and the set radius value as the radius is the number of other toll stations near the target toll station; Among them, the use of a first edge computing network element set near the target toll station to output the feedforward neural network after multiple learning as an intelligent traffic prediction model also includes: in each learning performed on the feedforward neural network, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles at the known target toll station in a certain time segment are used as the output content of the feedforward neural network, the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the certain time segment, and various configuration information of the target toll station are used as the input content of the feedforward neural network to complete this learning.

4. The method for dynamic configuration of toll booth lanes based on data collaboration as claimed in claim 3, characterized in that: Before using an Internet of Things sensor disposed at a target toll station to obtain the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each past time segment before the current time segment, the method further includes: A cloud computing server is used to remotely configure various working parameters of the IoT sensors installed at the target toll station.

5. The method for dynamic configuration of toll station lanes based on data collaboration as claimed in claim 3, characterized in that: After adopting a first edge computing network element disposed near a target toll station to output a feedforward neural network after completing multiple learnings as an intelligent traffic prediction model, the method further includes: The data storage device is used to store various model parameters of the intelligent traffic prediction model to complete the model storage of the intelligent traffic prediction model.

6. The method for dynamic configuration of toll station lanes based on data collaboration as claimed in claim 3, characterized in that: After using a cloud computing server disposed at a remote end of the target toll station to intelligently predict the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles of the target toll station in the current time segment according to the intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and taking the total number of electronic toll-collecting exit vehicles and the total number of manual toll-collecting exit vehicles in the current time segment as the first predicted number and the second predicted number, the method further includes: An on-site display device is used to receive the first predicted number and the second predicted number and realize on-site synchronous display of the first predicted number and the second predicted number.

7. The method for dynamic configuration of toll station lanes based on data collaboration according to any one of claims 3 to 6, characterized in that: A second edge computing network element is used to be set near the target toll station, which is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station, including: each exit channel of the target toll station has two different configurations: an electronic toll channel and a manual toll channel; The IoT sensor set at the target toll station is used to obtain the total number of electronic toll-exit vehicles and the total number of manual toll-exit vehicles corresponding to each past time segment before the current time segment, including: the IoT sensor is a visual sensor based on the IoT; Among them, the Internet of Things sensor set at the target toll station is used to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and also includes: the visual sensor based on the Internet of Things has various working parameters, and the various working parameters include shooting frame rate, imaging focal length, resolution, exposure time and aperture value.

8. The method for dynamic configuration of toll station lanes based on data collaboration according to any one of claims 3 to 6, characterized in that: Taking the distance between the target toll station and the nearest other toll stations as the reference distance, the number of learning times of the feedforward neural network and the numerical value of the reference distance are positively correlated, including: using a numerical conversion function to represent the numerical conversion relationship of the number of learning times of the feedforward neural network and the numerical value of the reference distance; Among them, the numerical conversion relationship that uses a numerical conversion function to represent the positive correlation between the number of learning times of the feedforward neural network and the numerical value of the reference distance includes: in the numerical conversion function, the number of learning times of the feedforward neural network corresponding to the numerical value of the reference distance is the output data of the numerical conversion function, and the numerical value of the reference distance is the input data of the numerical conversion function.

9. The method for dynamic configuration of toll station lanes based on data collaboration according to any one of claims 3 to 6, characterized in that: A cloud computing server disposed at a remote end of a target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles of the target toll station in the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment are used as the first prediction number and the second prediction number respectively, including: the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station are input into the intelligent traffic prediction model in parallel; Among them, a cloud computing server set up at the remote end of the target toll station is used to intelligently predict the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment of the target toll station according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment are used as the first predicted number and the second predicted number respectively. It also includes: running the intelligent traffic prediction model to obtain the total number of electronic toll exit vehicles and the total number of manual toll exit vehicles in the current time segment of the target toll station output by the intelligent traffic prediction model.

10. A toll station lane dynamic configuration system based on data collaboration, characterized in that: The system comprises: The IoT sensor is installed at the target toll station to obtain the total number of vehicles leaving the station for each electronic toll collection and the total number of vehicles leaving the station for each manual toll collection corresponding to each past time segment before the current time segment; The first edge computing network element is set near the target toll station and is used to output the feedforward neural network after completing multiple learning as an intelligent traffic prediction model; The second edge computing network element is set near the target toll station, and is used to output the total mileage of the highway where the target toll station is located, the number of permanent residents in the city closest to the target toll station, the total number of exit channels of the target toll station, and the width of the narrowest part of the entrance of the target toll station as various configuration information of the target toll station; A cloud computing server is arranged at a remote end of a target toll station, and is used to intelligently predict the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles of the target toll station in the current time segment according to an intelligent traffic prediction model based on the duration of each time segment, the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles corresponding to each past time segment before the current time segment, and various configuration information of the target toll station, and use the total number of electronic toll-collecting outbound vehicles and the total number of manual toll-collecting outbound vehicles in the current time segment as a first predicted number and a second predicted number, respectively; The cloud computing server is also used to configure the number of electronic toll collection channels and the number of manual toll collection channels of the target toll station in advance within the current time segment based on the total number of outbound channels of the target toll station, the first predicted number, and the second predicted number.

Images