A method, device and storage medium for completing a missing path on a highway

Abstract

The application discloses a missing path completion method and device on an expressway and a storage medium, and the method comprises the following steps: in a path missing interval, calculating the posterior probability of a target vehicle driving on each candidate path according to the number of vehicles passing through in history, vehicle data and the driving time length of the target vehicle in the path missing interval; selecting the candidate path with the maximum posterior probability as the target path of the target vehicle driving in the path missing interval; and according to the path composed of gantries with vehicle data and the target path, the driving path of the target vehicle on the expressway is formed. Through the fusion of the three-dimensional information of historical passing rules, current vehicle actual data and driving time rationality in the path missing interval, the posterior probability is obtained, multi-dimensional information integration can more comprehensively and truly reflect the possibility of the target vehicle driving, the reliability of the missing path prediction is improved, and the accuracy of the path completion is improved.

Description

Technical Field

[0001] This invention relates to the field of path restoration technology, and in particular to a method, device and storage medium for completing missing paths on highways. Background Technology

[0002] In actual highway traffic scenarios, the entire process of a vehicle entering from the tollbooth entrance and exiting from the tollbooth involves passing through multiple gantry devices used to collect vehicle traffic data. This data is used to help determine the vehicle's actual travel trajectory. However, due to various objective factors such as gantry device malfunction, communication signal interference, and severe weather, some gantries often fail to collect vehicle traffic data properly, resulting in missing vehicle trajectory data. This affects the determination of the travel path and further impacts subsequent toll settlement.

[0003] To address the issue of missing data due to the lack of vehicle data collection at the gantry, the current approach uses the minimum toll assumption method. This involves enumerating all connected paths between the missing intervals, calculating the theoretical toll for each path, and selecting the path with the lowest toll as the actual travel path for the missing interval. This method completes the path completion and toll calculation. However, the minimum toll path in this method only reflects the cost characteristics of the path and often does not match the actual travel path of the vehicle. This results in low accuracy in determining the missing path, affecting the accuracy of the complete path completion and consequently impacting the subsequent revenue settlement for the road segment. Summary of the Invention

[0004] This invention provides a method, device, and storage medium for completing missing paths on highways, in order to improve the accuracy of path completion.

[0005] In a first aspect, embodiments of the present invention provide a method for completing missing paths on highways, including: Receive vehicle data collected by multiple gantries between highway entrance and exit points, which is bound to the identity of the target vehicle; When the target vehicle arrives at the highway exit, the gantries are sorted according to the timestamps of the vehicle data to obtain a gantry sequence; Based on the highway network topology map, determine whether there is a topological edge between two adjacent gantries in the gantry sequence; the highway network topology map includes nodes and topological edges; the node represents the gantry, and the topological edge represents a passable road in the highway network topology map that is directly connected to two adjacent gantries without other gantries between them; If there is no topological edge between two adjacent gantry in the gantry sequence, then the interval between the two adjacent gantry corresponding to the absence of the topological edge is determined as a path missing interval; Within the missing path interval, the posterior probability of the target vehicle traveling on each candidate path is calculated based on the historical number of vehicles passing through each candidate path, the vehicle data, and the travel time of the target vehicle within the missing path interval; the candidate paths are all passable paths within the missing path interval. The candidate path with the highest posterior probability is selected as the target path for the target vehicle to travel in the path-missing interval. The path of the target vehicle on the highway is formed by combining the path of the gantry containing the vehicle data with the target path.

[0006] Secondly, embodiments of the present invention provide a path completion device for missing routes on highways, comprising: The vehicle data receiving module is used to receive vehicle data collected by the target vehicle from multiple gantries between the highway entrance and exit points, and which is bound to the identity of the target vehicle. The gantry sequence acquisition module is used to sort the gantries according to the timestamp of the vehicle data when the target vehicle arrives at the highway exit point, so as to obtain the gantry sequence. The topology edge determination module is used to determine whether there is a topology edge between two adjacent gantries in the gantry sequence based on the highway network topology map; the highway network topology map includes nodes and topology edges; the nodes represent gantries, and the topology edges represent passable roads in the highway network topology map where there are no other gantries between two adjacent gantries and they can be directly connected; The path missing interval determination module is used to determine the interval between two adjacent gantry that do not have the topological edge in the gantry sequence as the path missing interval if there is no topological edge between two adjacent gantry. The posterior probability calculation module is used to calculate the posterior probability of the target vehicle traveling on each candidate path in the path missing interval, based on the number of vehicles that have traveled on each candidate path in history, the vehicle data, and the travel time of the target vehicle in the path missing interval; the candidate paths are all passable paths in the path missing interval. The target path determination module is used to select the candidate path with the highest posterior probability as the target path for the target vehicle to travel in the path missing interval. The path completion module is used to construct the driving path of the target vehicle on the highway based on the path formed by the gantries containing the vehicle data and the target path.

[0007] Thirdly, embodiments of the present invention also provide a computer device, the computer device comprising: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for completing missing paths on highways as provided in the first aspect of the present invention.

[0008] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method for completing missing paths on highways as provided in the first aspect of the present invention.

[0009] Fifthly, embodiments of the present invention also provide a computer program product, the computer program product comprising a computer program, which, when executed by a processor, implements the method for completing missing paths on highways provided in the first aspect of the present invention.

[0010] In this embodiment of the invention, vehicle data bound to the target vehicle's identity is received from multiple gantries between the highway entrance and exit points. When the target vehicle arrives at the highway exit point, the gantries are sorted according to the timestamps of the vehicle data to obtain a gantry sequence. A topological edge is determined based on the highway network topology map. The highway network topology map includes nodes and topological edges; nodes represent gantries, and topological edges represent passable roads in the highway network topology map where there are no other gantries between adjacent gantries and they are directly connected. If two adjacent gantries in the gantry sequence... If there are no topological edges between gantries, the interval between two adjacent gantries without topological edges is defined as the path missing interval. Within the path missing interval, the posterior probability of the target vehicle traveling on each candidate path is calculated based on the historical number of vehicles passing through each candidate path, vehicle data, and the travel time of the target vehicle within the path missing interval. The candidate paths are all passable paths within the path missing interval. The candidate path with the highest posterior probability is selected as the target path for the target vehicle to travel within the path missing interval. The travel path of the target vehicle on the highway is formed by combining the paths formed by the gantries with collected vehicle data and the target path. This embodiment receives vehicle data to obtain vehicle passage records in the highway network, providing basic data support for path determination. It sorts the gantries by timestamp order to ensure the gantry sequence matches the actual vehicle travel time, avoiding data disorder and transmission delays that could lead to gantry sequence errors, thus guaranteeing the temporal rationality and authenticity of the gantry sequence. By using the physical connectivity of the highway network topology map to determine whether adjacent gantries are directly connected, if adjacent gantries are not directly connected in the topology, it can objectively and accurately identify gantries with uncollected data, precisely and objectively locating missing path intervals and improving the accuracy of missing path identification. To ensure objectivity, in the missing path interval, the posterior probability is obtained by integrating three dimensions of information: historical traffic patterns, current vehicle data, and the reasonableness of travel time. The integration of multi-dimensional information can more comprehensively and realistically reflect the probability of the target vehicle's travel, making the path judgment more in line with actual traffic conditions and improving the rationality and reliability of missing path inference. Finally, selecting the candidate path with the highest posterior probability conforms to the statistical optimal decision-making principle, ensuring that the selected candidate path is closest to the actual travel path of the vehicle in the missing path interval, improving the reliability and accuracy of missing path determination, ensuring the accuracy of subsequent complete path completion, and thus ensuring the accuracy of subsequent road segment revenue settlement. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a method for completing missing paths on a highway, as provided in Embodiment 1 of the present invention; Figure 2 This is a structural block diagram of a missing path completion device for highways provided in Embodiment 2 of the present invention; Figure 3 This is a schematic diagram of the structure of a computer device provided in Embodiment 3 of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be used interchangeably where appropriate so that the embodiments of the invention described herein can cover implementations in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] Example 1 See Figure 1 The diagram illustrates a flowchart of a method for completing missing paths on a highway according to Embodiment 1 of the present invention. This embodiment is applicable to situations where a vehicle's travel path in a missing section on a highway is inferred using a Bayesian algorithm. This method can be executed by a highway missing path completion device, which can be implemented in hardware and / or software and can be configured in a computer device. Figure 1 As shown, the method includes: Step 101: Receive vehicle data collected by the target vehicle from multiple gantries between the highway entrance and exit points, which are bound to the target vehicle's identification.

[0016] Highways are equipped with multiple gantries, typically installed at regular intervals (every few to tens of kilometers), for segmented toll collection and monitoring vehicle routes. Gantries are also placed at ramp connections where vehicles can enter or exit the highway to identify specific points of entry and exit. Highway entry and exit points are also considered gantries.

[0017] The gantry is used to collect vehicle data. This data typically includes a timestamp and comprises gantry transaction data and license plate data. If the target vehicle is not equipped with ETC (Electronic Toll Collection), the gantry transaction data is CPC card (Compound Pass Card) data. If the target vehicle is equipped with ETC, the gantry transaction data is OBU (On-Board Unit) data. The license plate data is the target vehicle's license plate number.

[0018] CPC cards are the medium for highway passage, integrating multiple communication functions. They can record information on the entry and exit of target vehicles at highway toll stations and their actual driving routes, providing a basis for highway toll calculation and settlement, and are widely used in highway passage scenarios for non-ETC vehicles.

[0019] OBU is the on-board unit of the ETC system, usually installed on the windshield of the vehicle. It can communicate wirelessly with the gantry on the side of the highway to enable functions such as non-stop toll payment and route information recording, and is a component for seamless passage.

[0020] In this embodiment, the system receives vehicle data bound to vehicle identification identifiers collected by multiple gantries between highway entry and exit points. This allows for the association and matching of vehicle data collected by each gantry with a unique target vehicle, avoiding confusion between different vehicle data. Simultaneously, it completely preserves the collection records of all valid gantries during the vehicle's journey from entering to exiting the highway. This provides a real, accurate, and clearly attributed original data foundation for the subsequent path analysis process, ensuring that all subsequent path-related processing is based on reliable data sources.

[0021] Step 102: When the target vehicle arrives at the highway exit, sort the gantries according to the timestamps of the vehicle data to obtain the gantry sequence.

[0022] In this embodiment, the arrival of the target vehicle at the highway exit indicates that the target vehicle's journey on the highway has ended. Since the vehicle data between the entry and exit points is collected in a decentralized manner, some gantries upload vehicle data to the system faster, while others upload vehicle data to the system slower due to network latency and other issues. Therefore, it is necessary to sort the gantries according to the timestamps of the vehicle data to obtain a gantry sequence. If the data is not sorted by timestamp, it is impossible to determine the order in which the vehicle passes through each gantry, making it difficult to determine the missing path intervals.

[0023] The gantry sequence obtained by sorting by timestamps can accurately reflect the time sequence of the target vehicle's driving trajectory, ensuring that the determination of adjacent gantries is consistent with the actual driving sequence of the vehicle, avoiding misjudgment of missing path intervals due to disordered gantry sequence, and laying the foundation for accurate identification of missing path parts.

[0024] Step 103: Determine whether there is a topological edge between two adjacent gantries in the gantry sequence based on the highway network topology map.

[0025] A highway network topology diagram is an abstract model of a highway network structure. It abstracts all gantries on a highway as nodes and roads that allow direct, continuous passage between any two gantries without other gantries as topological edges (i.e., a highway network topology diagram includes nodes and topological edges; nodes represent gantries, and topological edges represent passable roads that are directly connected between adjacent gantries in the highway network topology diagram without other gantries). This highway network structure model is used to express the direct connectivity and passage logic between gantries, without reflecting details such as the specific geometric length and geographical location of roads. It serves as an objective basis for determining whether direct passage between gantries is possible.

[0026] In this embodiment, after obtaining the gantry sequence sorted by time, the gantry sequence can only reflect the order in which vehicles pass through, and cannot directly reflect the actual connectivity between each gantry in the highway network. Moreover, adjacent gantries in time may not necessarily have only one connected physical road. Therefore, based on the highway network topology map, by judging whether there is a topological edge between adjacent gantries, it is clear whether the two gantries have a single direct connection relationship (rather than multiple paths or intermediate gantries). This allows for the accurate identification of the intervals where the specific driving path cannot be determined due to the lack of vehicle data collection at intermediate gantries, providing a prerequisite for the subsequent accurate definition of the missing path intervals.

[0027] Step 104: If there is no topological edge between two adjacent gantry in the gantry sequence, then the interval between the two adjacent gantry corresponding to the absence of a topological edge is determined as the path missing interval.

[0028] In this embodiment, if there is no topological edge between two adjacent gantries in the gantry sequence, it indicates that when the target vehicle arrives at the first gantry of the two adjacent gantries, the next gantry that the target vehicle passes through will not be the second gantry of the two adjacent gantries. Therefore, it can be deduced that there are other damaged gantries between the two adjacent gantries. Vehicle data cannot be collected from the damaged gantries. There may be multiple passable roads between the two adjacent gantries. At this time, it is impossible to determine which road the target vehicle traveled on in the two adjacent gantries. Therefore, the interval between the two adjacent gantries is defined as the path missing interval.

[0029] For example, each gantry has a gantry identifier (ID). The gantry identifiers of the gantry that have collected vehicle data are sorted in ascending order of timestamp to obtain the gantry sequence of the target vehicle's current journey on the highway. Verifying the gantry sequence using the highway network topology map can confirm whether there are any gantry instances where vehicle data was not collected. For example, if the collected gantry sequence is (1, 3, 4, 5), and if the highway network topology map confirms that there is no connecting topological edge between gantry identifier 1 and gantry identifier 3, it means that after the target vehicle passes gantry identifier 1, the next gantry identifier will not be 3. This indicates that there is a gantry between 1 and 3 in the collected gantry sequence (1, 3, 4, 5) where vehicle data was not collected. In this case, the interval between gantry identifier 1 and gantry identifier 3 can be defined as a path missing interval. The above example is only for illustrative purposes and does not limit the embodiments of the present invention.

[0030] Step 105: In the missing path interval, calculate the posterior probability of the target vehicle traveling on each candidate path based on the number of vehicles passing through each candidate path in history, vehicle data, and the travel time of the target vehicle in the missing path interval.

[0031] In this embodiment, multiple candidate paths exist within the path-missing interval. Existing gantry-collected vehicle data is typically used to determine the start and end gantries of the missing interval, but cannot directly determine the specific path the target vehicle actually travels within that interval. Therefore, it is necessary to calculate the posterior probability of the target vehicle traveling on each candidate path using the historical number of vehicles traveling on each path, vehicle data, and the target vehicle's travel time within the path-missing interval. This quantitatively measures the likelihood of each candidate path becoming the actual travel path, providing an objective numerical basis for selecting a unique and reasonable candidate path. The candidate paths are all passable paths within the path-missing interval.

[0032] By incorporating historical vehicle traffic data for candidate routes, we can leverage long-term accumulated road network traffic patterns to provide a foundational reference for the probability of each candidate route, avoiding the blindness of probability calculations due to a lack of historical data and ensuring that probability assessments are reasonably supported. Secondly, by combining vehicle data associated with the target vehicle, we can focus on the target vehicle's own traffic characteristics, eliminating interfering factors unrelated to the target vehicle, making probability calculations more closely aligned with the actual traffic conditions of the target vehicle, and improving the relevance of the probability assessment. Thirdly, by including the target vehicle's travel time in the missing sections, we can correlate the actual travel time of the vehicle with the traffic characteristics of the candidate routes, further... The system first filters out candidate paths that match vehicle driving behavior, narrowing down the range of reasonable candidate paths and improving the accuracy of probability assessment. The synergistic effect of these three aspects enables the calculated posterior probability to comprehensively and objectively quantify the likelihood of each candidate path becoming the actual driving path of the target vehicle. This provides a reliable numerical basis for selecting the candidate path with the highest posterior probability, effectively avoiding completion bias caused by subjective judgment, ensuring that the path completion results conform to the real road network traffic patterns and actual vehicle driving behavior. It eliminates the need for redundant data in the calculation, balancing the accuracy and computational efficiency of path completion, and improving the practicality and reliability of missing path completion.

[0033] In one embodiment of the present invention, step 105 includes the following steps: Step 1051: In the missing path interval, calculate the prior probability of the target vehicle traveling on each candidate path based on the number of vehicles that have traveled on each candidate path in history.

[0034] In this embodiment, there are multiple passable candidate paths within the missing path interval. However, due to the lack of gantry data, the actual driving path of the vehicle cannot be directly determined within this missing path interval, making it difficult to directly distinguish the probability of each candidate path. Therefore, the prior probability is calculated based on the historical number of vehicles passing through each candidate path. An initial probability (i.e., the prior probability of the target vehicle driving on each candidate path) that conforms to the historical traffic patterns is formed for this missing path interval. This provides an objective preliminary judgment basis for further accurate deduction of the actual driving path within the missing interval by combining real-time data.

[0035] By calculating prior probabilities based on historical vehicle traffic within the missing path interval, the system can accurately reflect the regular traffic preferences and flow distribution of each candidate path within that interval. This ensures that the prior probability judgment aligns with the actual operation of highways, avoiding errors caused by subjectively set probabilities. It guarantees the rationality and relevance of the missing path judgment and provides stable and reliable preliminary support for subsequent calculation of posterior probabilities based on comprehensive real-time information to determine the final travel path, ensuring that the missing path completion process is objective and orderly.

[0036] For example, the time interval of the missing path interval is determined based on the timestamp of vehicle data collection at the starting gantry and the timestamp of vehicle data collection at the ending gantry of the missing path interval; in the historical period that coincides with the start and end times of the time interval, for each candidate path, the number of vehicles that travel on the candidate path and whose vehicle type matches the target vehicle's vehicle type is counted; the number of vehicles traveling on each candidate path is counted.

[0037] The total number of vehicles passing through each candidate path is obtained by summing the historical vehicle counts. For each candidate path, the number of vehicles passing through is divided by the total historical vehicle count to obtain the prior probability of the target vehicle traveling on the candidate path. By combining the current time interval of the missing path with the historical vehicle counts based on the target vehicle model, the interference from differences in traffic behavior caused by different time periods and vehicle models can be eliminated. This ensures that the historical vehicle counts used are highly consistent with the current traffic conditions, allowing the prior probability to accurately reflect the actual traffic probability of each candidate path under the current environment, and significantly improving the scientific nature and reliability of the prior probability setting.

[0038] Step 1052: In the path missing interval, calculate the likelihood of the target vehicle traveling on each candidate path based on the vehicle data and the travel time of the target vehicle in the path missing interval.

[0039] In this embodiment, based on the collected real-time vehicle data and the travel time of the target vehicle in the missing path section, the matching degree between each candidate path and the actual travel trajectory of the target vehicle is quantitatively evaluated. By converting multi-dimensional observation information into a unified likelihood value, the probability of each candidate path becoming the actual travel path under the given observation data conditions is objectively reflected. This provides sufficient and reliable observation basis for subsequent calculation of posterior probability by combining prior probability, so that the travel path reasoning process can make full use of the real-time collected vehicle data and actual travel time information, and improve the fit of the travel path determination result to the actual traffic situation.

[0040] On the one hand, by combining vehicle data with actual travel time to calculate likelihood, we can fully explore and utilize the collected effective traffic information, maximize the supporting role of real-time vehicle data in route matching, provide scientific and quantitative evaluation results for each candidate route, and avoid the judgment bias of candidate routes due to insufficient dimensions of single vehicle data. On the other hand, by completing the fusion and normalization of multi-source information in the form of likelihood, we can make the matching degree of different candidate routes have a unified comparison standard, providing a clear and intuitive basis for subsequent posterior probability calculation and optimal route selection.

[0041] In one embodiment of the present invention, step 1052 may include the following steps: Step 10521: In the missing path interval, for each candidate path, determine the target gantry passage confidence of the candidate path by traversing the gantry flow data in the vehicle data of each target gantry.

[0042] In this embodiment, gantry flow data is the basic data reflecting vehicle travel trajectories, while candidate paths are all connected paths between missing path intervals. By traversing the gantry flow data of each target gantry in the missing interval, the theoretical passage gantry nodes of each candidate path can be matched one by one with the actual collected gantry flow data. This transforms the discrete gantry collection status into target gantry passage confidence that can be used for path evaluation, thereby quantifying the degree of matching between each candidate path and the actual gantry collection data. The target gantry is the gantry arranged in the candidate path.

[0043] For example, vehicle data includes gantry flow data; for each candidate path, each target gantry is traversed; each candidate path is analyzed separately, and each candidate path requires traversing each target gantry.

[0044] During the traversal of each target gantry, for each target gantry, if the candidate path has a target gantry and the target gantry has collected gantry flow data, then the initial gantry passage confidence of the candidate path to the target gantry is determined as the historical gantry flow data read rate of the target gantry. Whether a candidate path has a target gantry can be determined based on the highway network topology map. If a target gantry exists on a candidate path in the highway network topology map, then the candidate path has that target gantry. The historical gantry flow data read rate represents the probability that the target gantry has collected gantry flow data. For example, if 10 vehicles pass through a target gantry and the target gantry can collect gantry flow data from 9 vehicles, then the historical gantry flow data read rate of that gantry is 90%.

[0045] During the process of traversing each target gantry, for each target gantry, if the candidate path is equipped with a target gantry and no gantry flow data has been collected for the target gantry, the initial gantry passage confidence of the candidate path to the target gantry is determined to be the difference between 1 and the historical gantry flow data reading rate of the target gantry.

[0046] During the traversal of each target gantry, for each target gantry, if the candidate path does not have a target gantry and no gantry flow data has been collected for the target gantry, the initial gantry passage confidence of the candidate path to the gantry is set to 1.

[0047] During the traversal of each target gantry, for each target gantry, if the candidate path does not have a target gantry and the target gantry has collected gantry flow data, then the initial gantry passage confidence of the candidate path to the gantry is determined to be 0.

[0048] For each candidate path, the access confidence scores of each initial gantry for the candidate path are multiplied together to obtain the access confidence score of the target gantry for the candidate path. If there are n target gantry structures, then each candidate path corresponds to n initial gantry access confidence scores. In this case, the n initial gantry access confidence scores corresponding to the candidate path are multiplied together to obtain the access confidence score of the target gantry for that candidate path.

[0049] The target gantry access confidence of the candidate path is calculated using the following formula: in, Candidate paths Target gantry passage confidence; The historical gantry flow data read rate for target gantry k; The indicator variable is the passage of the target gantry k; Indicate candidate path The target gantry k is deployed; Read the indicator variable for the gantry flow data of the target gantry k; This indicates that the target gantry k collects gantry flow data for the target vehicle; Indicate candidate path No target gantry k was deployed; This indicates that no gantry flow data for the target vehicle was collected at the target gantry k. The target gantry k represents one of the gantry options among all candidate paths, and the gantry identifier for this target gantry is k.

[0050] Step 10522: In the missing path interval, for each candidate path, determine the target license plate reading confidence of the candidate path by traversing the license plate data in the vehicle data of each target gantry.

[0051] In this embodiment, the license plate data collected by the target gantry is used as an auxiliary verification basis for the rationality of the driving path. By traversing the license plate data of all target gantries, for each candidate road, the matching verification between all target gantries and the actual license plate data is performed. The discretely distributed license plate data is transformed into the target license plate reading confidence that can quantify the rationality of the candidate path. This provides an auxiliary judgment basis for the candidate path that is independent of the gantry flow data. The path reasoning process in the missing path interval can make full use of the license plate data actually collected by the target gantries to supplement and verify the passability of the candidate path, further enrich the data dimension of the path evaluation, and improve the completeness and rigor of the candidate path judgment logic.

[0052] For example, vehicle data also includes license plate data; for each candidate path, each target gantry is traversed.

[0053] During the traversal of each target gantry, for each target gantry, if a target gantry is deployed on the candidate path and the target gantry collects license plate data, the initial license plate reading confidence of the candidate path to the target gantry is determined as the historical license plate data reading rate of the target gantry. The historical license plate data reading rate is the probability that the target gantry can read license plate data. For example, if 10 vehicles pass through the target gantry and the target gantry can collect license plate data of 9 vehicles, it means that the historical license plate data reading rate of the target gantry is 90%.

[0054] During the process of traversing each target gantry, for each target gantry, if the candidate path is equipped with a target gantry and the target gantry has not collected license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined to be the difference between 1 and the historical license plate data reading rate of the target gantry.

[0055] During the process of traversing each target gantry, for each target gantry, if the candidate path does not have a target gantry and the target gantry has not collected license plate data, the initial license plate reading confidence of the candidate path to the target gantry is determined to be 1.

[0056] During the traversal of each target gantry, for each target gantry, if the candidate path does not have a target gantry and the target gantry has collected license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined to be 0.

[0057] For each candidate path, the initial license plate reading confidence scores of each candidate path are multiplied together to obtain the target license plate reading confidence score of the candidate path.

[0058] Step 10523: In the missing path interval, for each candidate path, calculate the travel time confidence of the candidate path based on the travel time of the target vehicle in the missing path interval.

[0059] In this embodiment, a spatiotemporal constraint is introduced: the actual travel time of the target vehicle in the path-missing section. A time matching evaluation mechanism is established between candidate paths and actual travel trajectories. Based on the actual travel time of the target vehicle in the path-missing section, the time dimension feature is transformed into a quantifiable travel time confidence level. Through time rationality verification, preliminary screening and credibility assessment of candidate paths are achieved. This ensures that the path reasoning process not only relies on gantry-collected data but also incorporates strong constraints based on the vehicle's actual travel time. This effectively eliminates candidate paths that clearly do not conform to actual traffic patterns in terms of time, improving the accuracy and logic of travel path judgment.

[0060] For example, in a missing path interval, the difference between the timestamp of the vehicle data collected by the gantry and the timestamp of the vehicle data collected by the starting gantry is calculated as the travel time of the target vehicle in the missing interval; for each candidate path, the square of the difference between the travel time and the theoretical travel time of the candidate path is calculated as the time deviation; for each candidate path, the time deviation is calculated by dividing by the square of the time standard deviation and then multiplying by -1 / 2 to obtain the time ratio; for each candidate path, an exponential operation with the natural constant as the base and the time ratio as the exponent is calculated as the exponent value; for each candidate path, the exponent value is calculated by dividing by the square root of the product to obtain the travel time confidence of the candidate path; the square root of the product is the product of twice pi (under the square root) and the time standard deviation. The natural constant is usually e.

[0061] The confidence level for travel time is calculated using the following formula: in, Candidate paths The confidence level of the travel time This is the time standard deviation (i.e., the range of fluctuation in allowable travel time). For travel time, Candidate paths The theoretical driving time.

[0062] Step 10524: For each candidate path, calculate the product of the target gantry passage confidence, the target license plate reading confidence, and the travel time confidence as the likelihood that the target vehicle will travel on the candidate path.

[0063] In this embodiment, the three independent confidence scores from the gantry flow data dimension, license plate data dimension, and driving time dimension are multiplied and fused. This transforms the scattered, heterogeneous, and different-scale individual verification results into a single, standardized likelihood value that can be directly used for probability calculation. This fully integrates the advantages of the three types of evidence and achieves mutual reinforcement, enabling the rationality of each candidate path to be comprehensively, objectively, and quantitatively characterized. This provides standardized and highly reliable observational input for subsequent calculation of posterior probability based on prior probability, ensuring that the final driving path determination result is based on complete information fusion rather than a single-dimensional judgment.

[0064] Step 1053: In the missing path interval, calculate the posterior probability of the target vehicle traveling on each candidate path based on the prior probability and likelihood.

[0065] In this embodiment, the prior probability obtained based on the number of vehicles passing through in history is organically integrated with the likelihood obtained based on real-time collected vehicle data and travel time. By constructing a scientific and reasonable probability calculation logic, a quantitative index (i.e., the posterior probability of travel on each candidate path) is formed to comprehensively reflect the degree of matching between the candidate path and the actual travel situation of the target vehicle. This allows the path reasoning process to not only rely on historical big data to reflect the objective laws of road network traffic, but also combine the current vehicle data to ensure the real-time nature and accuracy of the judgment, thereby providing a scientific, rigorous and statistically significant decision-making basis for finally selecting the most reasonable travel path.

[0066] On the one hand, by combining prior probability with likelihood to calculate posterior probability, the system can effectively balance historical traffic patterns with actual current traffic characteristics. This avoids the disconnect between historical data and actual current driving conditions, while also preventing the impact of local errors or missing information on judgments caused by relying solely on real-time data. This achieves complementary advantages from multiple sources of information and significantly improves the reliability of path reasoning results. On the other hand, as a comprehensive evaluation indicator, posterior probability can uniformly and quantitatively rank and compare the rationality of each candidate path, providing a clear, intuitive, and logically rigorous judgment standard for selecting the optimal candidate path. This effectively improves the standardization and operability of the candidate path determination process. At the same time, this calculation logic can stably adapt to various actual highway traffic scenarios such as incomplete gantry data collection, equipment malfunctions, and communication interference. Even under conditions of incomplete data, it maintains strong judgment ability and robustness, ensuring the continuous and stable operation of related businesses such as highway driving path determination, accurate toll fee breakdown, and road network operation management.

[0067] For example, for each candidate path, the product of the likelihood and the prior probability is calculated as the joint probability of the candidate path; the joint probabilities of each candidate path are added together to obtain the total probability of path observation; for each candidate path, the joint probability is calculated and divided by the total probability of path observation to obtain the posterior probability of the candidate path.

[0068] The posterior probability is calculated using the following formula: in, Candidate paths The posterior probability, Candidate paths Likelihood (i.e., candidate path) The probability of observing D), where D is the observed data, including vehicle data and driving time. Candidate paths The prior probability is given by m, where m is the number of candidate paths. Candidate paths likelihood Candidate paths The prior probability.

[0069] Step 106: Select the candidate path with the highest posterior probability as the target path for the target vehicle to travel in the path-deficient area.

[0070] In this embodiment, from all connected candidate paths between missing path intervals, the candidate path with the highest degree of matching with the current vehicle data, historical number of passing vehicles, and actual travel time is selected as the true path (i.e., the target path) of the target vehicle in the missing path interval. By using the maximum posterior probability as the decision basis, the quantitative evaluation results obtained by multi-dimensional data fusion calculation are transformed into a final path conclusion that can be directly applied. This provides clear, reliable, and statistically optimal path determination results for subsequent toll fee splitting, path tracing, road network management, and other businesses. It ensures that even in complex scenarios where gantry data is missing or the collected information is incomplete, the path determination result that is closest to the actual traffic situation can still be output, thus ensuring the integrity and practicality of the entire driving path determination method.

[0071] Step 107: Based on the path formed by the gantry containing the vehicle data and the target path, construct the driving path of the target vehicle on the highway.

[0072] In this embodiment, the gantries that have collected vehicle data can only form discontinuous path segments and cannot reflect the complete driving trajectory of the target vehicle. By combining the path formed by the gantries that have collected vehicle data with the target path in the missing path interval, the path interruption caused by the failure of gantry collection and the lack of vehicle data can be filled, and the discrete and discontinuous path segments can be connected into a continuous and complete driving path.

[0073] Highway networks are typically constructed and operated in phases by different construction and operation entities. The complete travel path of the same target vehicle from the entrance to the exit of the highway toll station will pass through multiple independent road segments managed by different operation entities. Each road segment has independent operating rights. Therefore, the complete travel path of the target vehicle includes multiple road segments in terms of physical composition and ownership division. Different road segments correspond to different operation and management entities. It is necessary to match the ownership and accurately calculate the toll fees of each road segment according to the actual traffic conditions of the target vehicle in order to meet the actual business needs of segmented operation, separate management, and separate accounting of highways.

[0074] In another embodiment of the present invention, for each segment of the driving route, the product of the mileage of the segment and the vehicle type rate of the target vehicle is calculated as the theoretical driving cost of the segment; the driving costs of each segment are added together to obtain the theoretical total cost; for each segment, the theoretical driving cost is calculated and divided by the theoretical total cost to obtain the cost ratio of the segment; for each segment, the product of the cost ratio and the cost incurred by the target vehicle on the highway is calculated to obtain the actual driving cost of the segment.

[0075] The theoretical driving cost for a road segment is calculated using the following formula: in, Let be the theoretical travel cost for road segment s. Let be the mileage of road segment s. Rates are based on vehicle type.

[0076] The actual driving cost for this route is calculated using the following formula: in, The actual driving cost for road segment s. The cost incurred by the target vehicle while traveling on the highway. The theoretical driving cost for road segment s. This represents the theoretical total cost.

[0077] In this embodiment of the invention, vehicle data bound to the target vehicle's identity is received from multiple gantries between the highway entrance and exit points. When the target vehicle arrives at the highway exit point, the gantries are sorted according to the timestamps of the vehicle data to obtain a gantry sequence. A topological edge is determined based on the highway network topology map. The highway network topology map includes nodes and topological edges; nodes represent gantries, and topological edges represent passable roads in the highway network topology map where there are no other gantries between adjacent gantries and they are directly connected. If two adjacent gantries in the gantry sequence... If there are no topological edges between gantries, the interval between two adjacent gantries without topological edges is defined as the path missing interval. Within the path missing interval, the posterior probability of the target vehicle traveling on each candidate path is calculated based on the historical number of vehicles passing through each candidate path, vehicle data, and the travel time of the target vehicle within the path missing interval. The candidate paths are all passable paths within the path missing interval. The candidate path with the highest posterior probability is selected as the target path for the target vehicle to travel within the path missing interval. The travel path of the target vehicle on the highway is formed by combining the paths formed by the gantries with collected vehicle data and the target path. This embodiment receives vehicle data to obtain vehicle passage records in the highway network, providing basic data support for path determination. It sorts the gantries by timestamp order to ensure the gantry sequence matches the actual vehicle travel time, avoiding data disorder and transmission delays that could lead to gantry sequence errors, thus guaranteeing the temporal rationality and authenticity of the gantry sequence. By using the physical connectivity of the highway network topology map to determine whether adjacent gantries are directly connected, if adjacent gantries are not directly connected in the topology, it can objectively and accurately identify gantries with uncollected data, precisely and objectively locating missing path intervals and improving the accuracy of missing path identification. To ensure objectivity, in the missing path interval, the posterior probability is obtained by integrating three dimensions of information: historical traffic patterns, current vehicle data, and the reasonableness of travel time. The integration of multi-dimensional information can more comprehensively and realistically reflect the probability of the target vehicle's travel, making the path judgment more in line with actual traffic conditions and improving the rationality and reliability of missing path inference. Finally, selecting the candidate path with the highest posterior probability conforms to the statistical optimal decision-making principle, ensuring that the selected candidate path is closest to the actual travel path of the vehicle in the missing path interval, improving the reliability and accuracy of missing path determination, ensuring the accuracy of subsequent complete path completion, and thus ensuring the accuracy of subsequent road segment revenue settlement.

[0078] Example 2 Figure 2 This is a schematic diagram of a missing path completion device for highways provided in Embodiment 2 of the present invention, as shown below. Figure 2 As shown, the device includes: The vehicle data receiving module 201 is used to receive vehicle data collected by the target vehicle from multiple gantries between the highway entrance point and the highway exit point, which is bound to the identity of the target vehicle. The gantry sequence acquisition module 202 is used to sort the gantries according to the timestamp of the vehicle data when the target vehicle arrives at the highway exit point, so as to obtain a gantry sequence. The topology edge determination module 203 is used to determine whether there is a topology edge between two adjacent gantries in the gantry sequence based on the highway network topology map; the highway network topology map includes nodes and topology edges; the nodes represent gantries, and the topology edges represent passable roads in the highway network topology map where there are no other gantries between two adjacent gantries and they can be directly connected; The path missing interval determination module 204 is used to determine the interval between two adjacent gantry that do not have the topological edge as a path missing interval if there is no topological edge between two adjacent gantry in the gantry sequence. The posterior probability calculation module 205 is used to calculate the posterior probability of the target vehicle traveling on each candidate path in the path missing interval, based on the number of vehicles that have traveled on each candidate path in history, the vehicle data, and the travel time of the target vehicle in the path missing interval; the candidate paths are all passable paths in the path missing interval. The target path determination module 206 is used to select the candidate path with the highest posterior probability as the target path for the target vehicle to travel in the path missing interval. The path completion module 207 is used to form the driving path of the target vehicle on the highway based on the path formed by the gantry containing the vehicle data and the target path.

[0079] In one embodiment of the present invention, the posterior probability calculation module 205 includes: The prior probability calculation module is used to calculate the prior probability of the target vehicle traveling on each candidate path based on the historical number of vehicles passing through each candidate path in the path missing interval. The likelihood calculation module is used to calculate the likelihood of the target vehicle traveling on each candidate path in the path missing interval, based on the vehicle data and the travel time of the target vehicle in the path missing interval. The posterior probability fusion module is used to calculate the posterior probability of the target vehicle traveling on each of the candidate paths based on the prior probability and the likelihood in the path missing interval.

[0080] In one embodiment of the present invention, the prior probability calculation module includes: The time interval determination module is used to determine the time interval of the missing path interval based on the timestamp of the vehicle data collected at the starting gantry and the timestamp of the vehicle data collected at the ending gantry of the missing path interval. The vehicle passing number statistics module is used to count the number of vehicles that are the same type as the target vehicle and pass through each candidate path during a period of time that is consistent with the start and end times of the time interval in history. The historical total vehicle count calculation module is used to add up the number of vehicles passing through each of the candidate paths to obtain the historical total vehicle count; The prior probability determination module is used to divide the number of passing vehicles by the total number of historical vehicles for each candidate path as the prior probability that the target vehicle travels on the candidate path.

[0081] In one embodiment of the present invention, the likelihood calculation module includes: The target gantry passage confidence determination module is used to determine the target gantry passage confidence of each candidate path in the missing path interval by traversing the gantry flow data in the vehicle data of each target gantry; the target gantry is the gantry arranged in the candidate path; The target license plate reading confidence determination module is used to determine the target license plate reading confidence of each candidate path in the missing path interval by traversing the license plate data in the vehicle data of each target gantry. The driving time confidence determination module is used to calculate the driving time confidence of each candidate path in the missing path interval based on the driving time of the target vehicle in the missing path interval. The likelihood determination module is used to calculate, for each candidate path, the product of the target gantry passage confidence, the target license plate reading confidence, and the travel time confidence as the likelihood of the target vehicle traveling on the candidate path.

[0082] In one embodiment of the present invention, the vehicle data includes gantry flow data; the target gantry passage confidence determination module includes: The first target gantry traversal module is used to traverse each target gantry for each candidate path; The first initial gantry passage confidence determination module is used to determine the initial gantry passage confidence of the candidate path to the target gantry as the historical gantry flow data reading rate of the target gantry during the traversal of each target gantry. If the candidate path is arranged with the target gantry and the target gantry has collected gantry flow data, the module is used to determine the initial gantry passage confidence of the candidate path to the target gantry as the historical gantry flow data reading rate of the target gantry. The second initial gantry passage confidence determination module is used to determine the initial gantry passage confidence of the candidate path to the target gantry as the difference between 1 and the historical gantry passage data reading rate of the target gantry during the traversal of each target gantry. If the candidate path is arranged with the target gantry and no gantry flow data is collected for the target gantry, then the initial gantry passage confidence of the candidate path to the target gantry is determined as the difference between 1 and the historical gantry flow data reading rate of the target gantry. The third initial gantry passage confidence determination module is used to determine the initial gantry passage confidence of the candidate path to the gantry as 1 if, during the traversal of each target gantry, the candidate path does not have the target gantry and the target gantry has not collected gantry flow data. The fourth initial gantry passage confidence determination module is used to determine the initial gantry passage confidence of the candidate path to the gantry as 0 if, during the traversal of each target gantry, the candidate path does not have the target gantry and the target gantry has collected gantry flow data. The target gantry access confidence calculation module is used to multiply the initial gantry access confidence of each candidate path for each candidate path to obtain the target gantry access confidence of the candidate path. In one embodiment of the present invention, the vehicle data further includes license plate data; the target license plate reading confidence determination module includes: The second target gantry traversal module is used to traverse each target gantry for each candidate path; The first initial license plate reading confidence determination module is used to determine the initial license plate reading confidence of the candidate path to the target gantry as the historical license plate data reading rate of the target gantry during the process of traversing each target gantry. If the candidate path is arranged with the target gantry and the target gantry has collected license plate data, the module is used to determine the initial license plate reading confidence of the candidate path to the target gantry as the historical license plate data reading rate of the target gantry. The second initial license plate reading confidence determination module is used to determine the initial license plate reading confidence of the candidate path to the target gantry as the difference between 1 and the historical license plate data reading rate of the target gantry when the candidate path is arranged with the target gantry and the target gantry has not collected license plate data during the traversal of each target gantry. The third initial license plate reading confidence determination module is used to determine the initial license plate reading confidence of the candidate path to the target gantry as 1 if, during the process of traversing each target gantry, the candidate path does not have the target gantry and the target gantry has not collected license plate data. The fourth initial license plate reading confidence determination module is used to determine the initial license plate reading confidence of the candidate path to the target gantry as 0 if, during the process of traversing each target gantry, the candidate path does not have the target gantry and the target gantry has collected license plate data. The target license plate reading confidence calculation module is used to multiply the initial license plate reading confidence scores of each candidate path for each candidate path to obtain the target license plate reading confidence score of the candidate path.

[0083] In one embodiment of the present invention, the driving time confidence determination module includes: The driving time calculation module is used to calculate the difference between the timestamp of the vehicle data collected by the gantry and the timestamp of the vehicle data collected by the starting gantry in the missing path interval as the driving time of the target vehicle in the missing interval. The time deviation calculation module is used to calculate the square of the difference between the travel time and the theoretical travel time of the candidate path for each candidate path, and take it as the time deviation. The time ratio calculation module is used to calculate the time ratio for each candidate path by dividing the time deviation by the square of the time standard deviation and then multiplying it by negative half. The exponent value calculation module is used to calculate an exponent value for each candidate path, with the natural constant as the base and the time ratio as the exponent. The travel time confidence calculation module is used to calculate the travel time confidence of each candidate path by dividing the index value by the square root multiplier; the square root multiplier is the product of twice the square root of pi and the time standard deviation.

[0084] In one embodiment of the present invention, the posterior probability fusion module includes: The joint probability calculation module is used to calculate the product between the likelihood and the prior probability for each candidate path as the joint probability of the candidate path. The path observation total probability calculation module is used to add up the joint probabilities of each candidate path to obtain the path observation total probability; The posterior probability determination module is used to calculate the joint probability divided by the total probability of path observation for each candidate path as the posterior probability of the target vehicle traveling on the candidate path.

[0085] In one embodiment of the present invention, it further includes: The driving path composition module is used to compose the driving path of the target vehicle on the highway based on the path composed of the gantry containing the collected vehicle data and the target path. The theoretical driving cost calculation module is used to calculate the product of the mileage of the road segment and the vehicle type rate of the target vehicle for each road segment of the driving route, as the theoretical driving cost of the road segment. The theoretical total cost calculation module is used to add up the theoretical driving costs of each road segment to obtain the theoretical total cost; The cost ratio calculation module is used to calculate the theoretical driving cost divided by the theoretical total cost for each road segment to obtain the cost ratio of that road segment. The actual driving cost determination module is used to calculate the product of the cost ratio and the cost incurred by the target vehicle on the highway for each road segment, so as to obtain the actual driving cost of the road segment.

[0086] The missing path completion device for highways provided in this embodiment of the invention can execute the missing path completion method for highways provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the missing path completion method for highways.

[0087] Example 3 See Figure 3 This diagram illustrates a structural schematic of a computer device according to an embodiment of the present invention. The term "computer device" is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The computer device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0088] like Figure 3As shown, the computer device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer programs stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the computer device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0089] Multiple components in computer device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows computer device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the method for completing missing paths on a highway.

[0091] In some embodiments, the method for completing missing paths on highways can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on computer device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for completing missing paths on highways described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the method for completing missing paths on highways by any other suitable means (e.g., by means of firmware).

[0092] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0093] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0094] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0095] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer device having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0096] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0097] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0098] Example 4 This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the method for completing missing paths on highways as provided in any embodiment of this invention.

[0099] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0100] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0101] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for completing missing paths on highways, characterized in that, include: Receive vehicle data collected by multiple gantries between highway entrance and exit points, which is bound to the identity of the target vehicle; When the target vehicle arrives at the highway exit, the gantries are sorted according to the timestamps of the vehicle data to obtain a gantry sequence; Based on the highway network topology map, determine whether there is a topological edge between two adjacent gantries in the gantry sequence; the highway network topology map includes nodes and topological edges; the node represents the gantry, and the topological edge represents a passable road in the highway network topology map that is directly connected to two adjacent gantries without other gantries between them; If there is no topological edge between two adjacent gantry in the gantry sequence, then the interval between the two adjacent gantry corresponding to the absence of the topological edge is determined as a path missing interval; Within the missing path interval, the posterior probability of the target vehicle traveling on each candidate path is calculated based on the historical number of vehicles passing through each candidate path, the vehicle data, and the travel time of the target vehicle within the missing path interval; the candidate paths are all passable paths within the missing path interval. The candidate path with the highest posterior probability is selected as the target path for the target vehicle to travel in the path-missing interval. The path of the target vehicle on the highway is formed by combining the path of the gantry containing the vehicle data with the target path.

2. The method according to claim 1, characterized in that, The calculation of the posterior probability of the target vehicle traveling on each candidate path based on the historical number of vehicles passing through each candidate path, the vehicle data, and the travel time of the target vehicle in the missing path interval includes: In the missing path interval, the prior probability of the target vehicle traveling on each candidate path is calculated based on the number of vehicles that have traveled on each candidate path in history. In the missing path interval, the likelihood of the target vehicle traveling on each candidate path is calculated based on the vehicle data and the travel time of the target vehicle in the missing path interval; Within the missing path interval, the posterior probability of the target vehicle traveling on each candidate path is calculated based on the prior probability and the likelihood.

3. The method according to claim 2, characterized in that, The step of calculating the prior probability of the target vehicle traveling on each candidate path within the missing path interval, based on the historical number of vehicles traveling on each candidate path, includes: The time interval of the missing path interval is determined based on the timestamp of the vehicle data collected at the starting gantry and the timestamp of the vehicle data collected at the ending gantry of the missing path interval. In a historical period that coincides with the start and end times of the time interval, for each candidate path, the number of vehicles whose vehicle type matches that of the target vehicle that traveled on the candidate path is counted. The total number of vehicles passing through each of the candidate paths is summed to obtain the total number of vehicles in history. For each candidate path, the number of passing vehicles divided by the total number of historical vehicles is used as the prior probability that the target vehicle travels on the candidate path.

4. The method according to claim 2, characterized in that, The step of calculating the likelihood of the target vehicle traveling on each candidate path within the missing path interval, based on the vehicle data and the target vehicle's travel time within the missing path interval, includes: In the missing path interval, for each candidate path, the target gantry passage confidence of the candidate path is determined by traversing the gantry flow data in the vehicle data of each target gantry; the target gantry is the gantry arranged in the candidate path; In the missing path interval, for each candidate path, the target license plate reading confidence of the candidate path is determined by traversing the license plate data in the vehicle data of each target gantry; In the missing path interval, for each candidate path, the travel time confidence of the candidate path is calculated based on the travel time of the target vehicle in the missing path interval; For each candidate path, the product of the target gantry passage confidence, the target license plate reading confidence, and the travel time confidence is calculated as the likelihood that the target vehicle travels on the candidate path.

5. The method according to claim 4, characterized in that, The vehicle data includes gantry flow data; the step of determining the target gantry passage confidence of each candidate path within the missing path interval by traversing the gantry flow data in the vehicle data of each target gantry includes: For each candidate path, traverse each target gantry; During the process of traversing each of the target gantries, for each target gantry, if the candidate path is arranged with the target gantry and the target gantry has collected gantry flow data, then the initial gantry passage confidence of the candidate path to the target gantry is determined as the historical gantry flow data reading rate of the target gantry. During the process of traversing each of the target gantries, for each target gantry, if the candidate path is arranged with the target gantry and no gantry flow data is collected for the target gantry, then the initial gantry passage confidence of the candidate path to the target gantry is determined to be the difference between 1 and the historical gantry flow data reading rate of the target gantry. During the process of traversing each of the target gantries, for each target gantry, if the candidate path does not have the target gantry and the target gantry has not collected gantry flow data, then the initial gantry passage confidence of the candidate path to the gantry is determined to be 1. During the process of traversing each of the target gantries, for each target gantry, if the candidate path does not have the target gantry and the target gantry has collected gantry flow data, then the initial gantry passage confidence of the candidate path to the gantry is determined to be 0. For each candidate path, the initial gantry access confidence scores of each candidate path are multiplied together to obtain the target gantry access confidence score of the candidate path. The vehicle data also includes license plate data; the step of determining the target license plate reading confidence level of each candidate path within the missing path interval by traversing the license plate data in the vehicle data of each target gantry includes: For each candidate path, traverse each target gantry; During the process of traversing each of the target gantries, for each target gantry, if the candidate path is arranged with the target gantry and the target gantry collects license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined as the historical license plate data reading rate of the target gantry. During the process of traversing each of the target gantries, for each target gantry, if the candidate path is arranged with the target gantry and the target gantry has not collected license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined to be the difference between 1 and the historical license plate data reading rate of the target gantry. During the process of traversing each of the target gantries, for each target gantry, if the candidate path does not have the target gantry and the target gantry has not collected license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined to be 1. During the process of traversing each of the target gantries, for each target gantry, if the candidate path does not have the target gantry and the target gantry has collected license plate data, then the initial license plate reading confidence of the candidate path to the target gantry is determined to be 0. For each candidate path, the initial license plate reading confidence scores of each candidate path are multiplied together to obtain the target license plate reading confidence score of the candidate path.

6. The method according to claim 4, characterized in that, In the missing path interval, for each candidate path, the travel time confidence score of the candidate path is calculated based on the travel time of the target vehicle in the missing path interval, including: In the missing path interval, the difference between the timestamp of the vehicle data collected by the gantry and the timestamp of the vehicle data collected by the starting gantry is calculated as the travel time of the target vehicle in the missing interval; For each candidate path, the square of the difference between the travel time and the theoretical travel time of the candidate path is calculated as the time deviation. For each candidate path, the time deviation is calculated by dividing the square of the time standard deviation and then multiplying it by negative half to obtain the time ratio. For each candidate path, an exponential operation with the natural constant as the base and the time ratio as the exponent is calculated as the exponent value; For each candidate path, the index value is calculated by dividing it by the square root multiplier to obtain the travel time confidence of the candidate path; the square root multiplier is the product of twice the square root of pi and the time standard deviation.

7. The method according to claim 2, characterized in that, In the missing path interval, calculating the posterior probability of the target vehicle traveling on each candidate path based on the prior probability and the likelihood includes: For each candidate path, the product of the likelihood and the prior probability is calculated as the joint probability of the candidate path; The joint probabilities of each candidate path are summed to obtain the total probability of path observation; For each candidate path, the joint probability is calculated and divided by the total probability of path observations to obtain the posterior probability of the target vehicle traveling on the candidate path.

8. The method according to any one of claims 1-7, characterized in that, Also includes: For each segment of the driving route, the product of the mileage of the segment and the vehicle type rate of the target vehicle is calculated as the theoretical driving cost of the segment. The theoretical driving costs for each road segment are added together to obtain the theoretical total cost; For each road segment, the theoretical driving cost is calculated and divided by the theoretical total cost to obtain the cost ratio for that road segment; For each road segment, the actual travel cost of that road segment is obtained by multiplying the cost ratio by the cost incurred by the target vehicle while traveling on the highway.

9. A computer device, characterized in that, The computer device includes: One or more processors; Storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method for completing missing paths on highways as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for completing missing paths on highways as described in any one of claims 1-8.

Images