A real-time recognition system and method for abnormal operation behavior of large-scale transport vehicles

Abstract

The present invention relates to a real-time identification system and method for abnormal operation behavior of large-scale transport vehicles, including a processing module, a pass acquisition module, a weighing data acquisition module, a license plate information acquisition module and a positioning data acquisition module; the processing module determines the transport vehicle that needs to be supervised according to the expiration date of the pass, determines the load state of the corresponding transport vehicle according to the weight information, license plate information and weighing data specified in the pass, and determines the transport information of the corresponding transport vehicle according to the route information, license plate information and positioning data specified in the pass. The present invention is helpful to improve the supervision ability of large-scale transport behavior, crack down on various related illegal behaviors, protect highway infrastructure, and improve the safety level of highway transportation.

Description

Technical Field

[0001] The present invention relates to the technical field of large-scale transport management, and in particular to a real-time recognition system and method for abnormal operation behavior of large-scale transport vehicles. Background Art

[0002] The objects of large-scale transportation are generally large and heavy equipment for construction projects in industries such as electricity, chemical, petroleum, machinery, and metallurgy. With the increasing number of infrastructure investment projects, the number of large-scale transportation is also showing a rapid growth trend. At the same time, traffic safety accidents caused by irregular large-scale transportation behavior are also emerging in an endless stream, exposing the major defect of insufficient supervision capacity in the current industry.

[0003] Compared with general cargo transportation, large-scale transportation has the characteristics of high risk, long cycle, and slow transportation speed. Once a safety accident occurs, it often causes serious damage to the road and is extremely destructive. At present, the supervision capacity of large-scale transportation of vehicles is relatively weak. It is mainly supervised by setting up checkpoints at the entrances and exits of highways and manual random inspections, and can only supervise a few points in the large-scale transportation process. At the same time, in order to evade inspections, some large-scale transportation drivers often choose to bypass the inspection points, making the entire large-scale transportation process completely out of the control of the regulatory authorities, and posing great safety hazards. Therefore, it is urgent to strengthen the supervision of large-scale transportation, ensure that large-scale transportation is carried out in compliance with regulations, improve the safety level of large-scale transportation, protect the lives of the people, and reduce property losses in road infrastructure. Summary of the invention

[0004] In order to solve the problems existing in the above-mentioned prior art, the present invention provides a real-time identification system for abnormal operating behavior of large-scale transport vehicles, including a processing module, a pass acquisition module, a weighing data acquisition module, a license plate information acquisition module and a positioning data acquisition module.

[0005] The pass acquisition module is used to obtain the pass information for large-scale transportation, and the relevant pass information can be queried from other systems. The weighing data acquisition module is used to obtain one or any combination of weighing station weighing data, highway weighing data and off-site weighing data. The weighing information can be obtained through a specially set device or by querying other systems. The positioning data acquisition module is used to obtain the positioning information of the transport vehicle, and the license plate information acquisition module is used to obtain the license plate information of the transport vehicle.

[0006] The processing module determines the transport vehicles that need to be supervised according to the expiration date of the pass, determines the load status of the corresponding transport vehicles according to the weight information, license plate information and weighing data specified in the pass, and determines the transport information of the corresponding transport vehicles according to the route information, license plate information and positioning data specified in the pass.

[0007] Furthermore, it also includes a weighing station weighing data uploading module, a highway weighing data uploading module and an off-site weighing data uploading module, which are adapted to the weighing data acquisition module and are used to upload the weighing station weighing data, highway weighing data and off-site weighing data corresponding to the license plate information.

[0008] Furthermore, it also includes a license plate information and positioning data uploading module for uploading the license plate information and positioning information of the corresponding transport vehicle.

[0009] The present invention also provides a method for real-time identification of abnormal operation behavior of large-scale transport vehicles, which is used in the above-mentioned system and comprises the following steps:

[0010] S1. According to the expiration date ExpireTime of the pass, the pass with t≤ExpireTime+A is extended by A days and included in the supervision scope. t is the current time, and the pass entry time Ts is recorded as the current time t. Based on the expiration date ExpireTime of the pass, the pass data PassData with an expiration time not exceeding A days is uniformly entered, and no unified entry of passes is performed subsequently.

[0011] S2. Determine the interval between the current time t and the pass entry time Ts. If t-Ts ≥ 1 day, access the pass data PassData whose pass verification time VerifyTime is between the pass entry time Ts and the current time t, and update the pass entry time Ts to the current time. Increase the number of supervised vehicles according to a fixed time period, determine and dynamically update the key supervised vehicles, and supervise all vehicles that have newly applied for passes to ensure sufficient supervision scope.

[0012] S3. Determine in turn whether the expiration time of the pass of the supervised vehicle Ci (i=1, 2…n) exceeds A days. If t>ExpireTime+A, the vehicle Ci will no longer be supervised. Vehicles whose passes have been expired for too long will no longer be supervised to reduce the supervision pressure and ultimately ensure the dynamic determination and update of key supervised vehicles.

[0013] S4. Determine whether there is still a supervised vehicle. If not, go to S2.

[0014] S5. Determine the upper limit WTMax and the lower limit WTMin of the weighing time WeightTime of the weighing data to be accessed, and access the weighing data WeightData whose weighing time is greater than or equal to WTMin and less than or equal to WTMax.

[0015] S6. Delete the weighing data WeightData whose weight is not greater than B tons. For supervised vehicles with multiple weighing records, keep the most recent weighing record to improve the subsequent matching speed.

[0016] S7. Traverse the supervised vehicles Ci, and match the weighing data WeightData with the weight Ws specified in the pass based on the license plate information; if the supervised vehicle has weighing data, update the vehicle's large-scale transportation status TransferStatus.

[0017] S8. For supervised vehicles that exceed the designated weight, the supervised vehicle is deemed to be transporting large items in excess of the designated weight; for supervised vehicles that are currently transporting large items (including those that meet the designated weight and those that exceed the designated weight), determine whether their passes are valid. If the pass is invalid (including not in effect or has expired), the supervised vehicle is deemed not to be transporting large items at the designated time.

[0018] S9. Determine the upper limit RTMax and the lower limit RTMin of the recording time RecordTime of the real-time positioning data to be accessed, and access the positioning data LocationData of the supervised vehicle whose recording time is greater than or equal to RTMin and less than or equal to RTMax.

[0019] S10, arrange the positioning data of the supervised vehicle Ci within D minutes in positive order according to the recording time RecordTime, calculate the Manhattan distances of the two positioning data recorded adjacently in sequence, sum them up and approximately convert them into plane distances, and obtain the vehicle driving distance M1 within D minutes; calculate the difference between the recording time RecordTime of the first positioning data and the recording time RecordTime of the last positioning data in the positioning data sequence of the supervised vehicle Ci within D minutes, and obtain the running time T1.

[0020] S11. Calculate the average driving speed V1 of the supervised vehicle Ci within D minutes, and if V1 is less than F, determine that the vehicle is in a stopped state; otherwise, determine that the vehicle is in a driving state.

[0021] S12. If the supervised vehicle Ci in the stopped state exceeds G minutes, the current trip of the supervised vehicle is deemed to have ended; the text information of the transportation destination is extracted from the designated driving route, and the destination text information is converted into longitude and latitude information having the same coordinate system as the supervision vehicle positioning data; the distance M2 between the longitude and latitude recorded at the last time in the supervision vehicle positioning data and the longitude and latitude of the destination is calculated; if M2 is not greater than H, it is deemed that the supervised vehicle has arrived at the destination, and the heavy-duty transportation of the supervised vehicle is recorded to have been completed; otherwise, the transportation state of the supervised vehicle is changed to no longer continuing heavy-duty transportation, and the stop state duration StopTime is changed to 0.

[0022] S13. For supervised vehicles in driving state, the latest positioning data is selected, and the road R where the vehicle is located is determined based on the current positioning data of the supervised vehicle; all road name text information is extracted from the designated driving route, and the current road R is selected to match with the j-th road Rj in turn to determine whether the current driving road of the vehicle meets the designated route; if the current driving road of the vehicle does not match the designated route, it is determined that the vehicle is not transporting large items according to the designated route.

[0023] S14. In order to maintain continuous and uninterrupted supervision of large-scale transport vehicles, the next step is to go to S2 for the next round of identification.

[0024] Furthermore, in S13, the matching method includes:

[0025] S131. Compare the character lengths of the current road R with those of the j-th road Rj. The longer one is marked as L, the shorter one is marked as S, and the character length of S is recorded as X. Determine in turn whether each character s of S is in L. If so, store the character s in the empty set O. After the determination is completed, if the set O is empty, the match rate MatchRate of L and S is 0, and proceed to step S134.

[0026] S132, extract each character s in the set O in turn, determine its position in L, and starting from this position, continuously extract X characters from L to form a character string Lx, and store it in the set U; if the number of characters extracted from this position is less than X, only the last character of L is extracted; after completion, the number of character strings in the set U is calculated to be u.

[0027] S133, traverse the kth Lx in the set U, extract the first 1, 2, ..., X consecutive characters in Lx respectively, form X strings with lengths of 1, 2, ..., X respectively, determine whether they exist in S in turn, and record the longest length of the string existing in S as the matching length Xp of the kth Lx and S; after traversing Lx, accumulate all matching lengths of each Lx and S, divide by the number of characters in S to obtain the matching rate MatchRate of L and S.

[0028] S134. Calculate the maximum matching rate MatchRate between the current road R and the j-th road Rj. If the maximum MatchRate is greater than Y, it is determined that the current road of the supervised vehicle matches the road specified in the pass; otherwise, it is determined that the current road of the supervised vehicle does not meet the specified route.

[0029] Furthermore, S8 also includes: if the ratio of the weighing data W of the supervised vehicle to the weight Ws specified in the pass is less than Z, it is determined that the vehicle is not currently transporting large items.

[0030] Furthermore, S3 also includes: deleting the pass data with t>ExpireTime+A to reduce the pressure of system supervision.

[0031] Furthermore, in S5, for the weighing data WeightData accessed for the first time, WTMin is t-30, and WTMax is t; for the weighing data WeightData accessed not for the first time, WTMin is Tw, and WTMax is t; Tw is the vehicle weighing data access time.

[0032] Furthermore, in S9, for the positioning data LocationData accessed for the first time, RTMin is t-5, and RTMax is t; for the positioning data LocationData not accessed for the first time, RTMin is Tq, and RTMax is t; Tq is the access time of the real-time positioning data.

[0033] Furthermore, A is preset to 20-30, B is preset to 20-50, D is preset to 2-10, F is preset to 1-3, G is preset to 120-240, H is preset to 1-3km, Y is preset to 0.9-0.99, and Z is preset to 0.7-0.99.

[0034] The present invention solves the current problems of weak "single-point" supervision of large-scale transportation and drivers' deliberate avoidance of inspection. Based on the supervision idea of ​​"supervisory vehicle screening-large-scale transportation status judgment-abnormal operation behavior identification", it makes full use of the pass data, weighing data and real-time positioning data of large-scale transportation vehicles to supervise the large-scale transportation vehicles throughout the whole process, and can provide real-time warnings for the four abnormal operation behaviors of large-scale transportation vehicles: "large-scale transportation not according to the prescribed time", "large-scale transportation not according to the prescribed route", "large-scale transportation exceeding the prescribed weight", and "large-scale transportation not according to the prescribed route and exceeding the prescribed weight".

[0035] The present invention helps to improve the supervision capability of large-scale transportation, combat various related illegal activities, protect highway infrastructure, and improve the safety level of highway transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0037] Figure 1 It is a simplified flow chart of the method for real-time identification of abnormal operating behavior of large-scale transport vehicles in the present invention. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0039] The present embodiment is a real-time identification system for abnormal operation behavior of large-scale transport vehicles, including a processing module, a pass acquisition module, a weighing data acquisition module, a license plate information acquisition module and a positioning data acquisition module.

[0040] The pass acquisition module is used to obtain the pass information for large-scale transportation, and the relevant pass information can be queried from other systems. The weighing data acquisition module is used to obtain one or any combination of weighing station weighing data, highway weighing data and off-site weighing data. The weighing information can be obtained through a specially set device or by querying other systems. The positioning data acquisition module is used to obtain the positioning information of the transport vehicle, and the license plate information acquisition module is used to obtain the license plate information of the transport vehicle.

[0041] The processing module determines the transport vehicles that need to be supervised according to the expiration date of the pass, determines the load status of the corresponding transport vehicles according to the weight information, license plate information and weighing data specified in the pass, and determines the transport information of the corresponding transport vehicles according to the route information, license plate information and positioning data specified in the pass.

[0042] Preferably, it also includes a weighing station weighing data upload module, a highway weighing data upload module and an off-site weighing data upload module, which are adapted to the weighing data acquisition module and are arranged at the corresponding weighing stations, highway entrances and off-site weighing locations, and are used to upload the weighing station weighing data, highway weighing data and off-site weighing data corresponding to the license plate information.

[0043] Preferably, it also includes a license plate information and positioning data uploading module, which is arranged on the transport vehicle and is used to upload the license plate information and positioning information of the corresponding transport vehicle. Example

[0044] See also Figure 1 A real-time identification method for abnormal operation behavior of a large-scale transport vehicle in this embodiment is used in the system in the above embodiment, and includes the following steps:

[0045] First, determine the supervised vehicles Ci (i=1, 2…n) for large-scale transportation, where n is the number of vehicles that currently need to be supervised, and add and delete supervised vehicles according to a fixed time period. Determining the supervised vehicles is a prerequisite for identifying abnormal operating behaviors, and the supervised vehicles are generally determined using pass data. Since some vehicles do not carry out large-scale transportation at the specified time, supervising the vehicles only during the validity period of the pass may lead to the omission of abnormal behaviors in large-scale transportation. Therefore, it is necessary to supervise all vehicles with passes. However, the large number of passes makes full supervision difficult and the benefits are not high. In order to achieve the determination and dynamic update of key supervised vehicles, the present invention supervises all vehicles that have newly applied for passes to ensure a sufficient supervision scope, and at the same time no longer supervises vehicles whose passes have been invalid for too long, so as to reduce the supervision pressure and ultimately ensure the dynamic determination and update of key supervised vehicles. Specifically includes the following steps:

[0046] S1. Determine whether it is the first time to determine the supervised vehicle. If so, based on the expiration date ExpireTime of the pass, uniformly enter the pass data PassData with an expiration time not exceeding A days. A is preset to be 20-30, and this embodiment takes 30 as an example for explanation. No unified entry of passes is performed subsequently, and the pass entry time Ts is recorded as the current time t.

[0047] The uniformly entered pass data PassData is selected according to the following formula:

[0048] t≤ExpireTime+30

[0049] S2. Determine whether the interval between the current time t and the pass entry time Ts is greater than or equal to 1 day, that is, t-Ts ≥ 1 day. If so, access the pass data PassData whose pass verification time VerifyTime is between the pass entry time Ts and the current time t, and update the pass entry time Ts to the current time t.

[0050] The method for selecting the pass data PassData according to the pass verification time VerifyTime is as follows:

[0051]

[0052] Where Ts is the entry time of the pass, VerifyTime is the time when the pass is reviewed and approved, and t is the current time.

[0053] S3. No longer supervise vehicles whose pass expires more than 30 days ago. Determine in sequence whether the pass expiration time of supervised vehicles Ci (i=1, 2…n) exceeds 30 days. If yes, no longer supervise the vehicle Ci.

[0054] The formula for determining whether to delete a supervised vehicle Ci (i=1, 2…n) based on the expiration time of the pass is as follows:

[0055]

[0056] Where t is the current time, and ExpireTime is the expiration date of the supervised vehicle pass.

[0057] S4. Determine whether there is still a supervised vehicle. If yes, execute S5; if no, execute S2.

[0058] Update the large-scale transport status of supervised vehicles Ci (i=1, 2…n) at a fixed time period to determine whether there is any abnormal operation behavior. Judging the large-scale transport status of the vehicle is a prerequisite for identifying abnormal vehicle operation behavior, which can avoid misidentification of vehicle behavior during non-large-scale transportation. The non-large-scale transportation process of supervised vehicles is not supervised by this method. Only when the supervised vehicle is carrying out large-scale transportation does it need to perform real-time identification of the abnormal behavior mentioned above. However, the existing data does not record the start and end time of the vehicle's actual large-scale transportation, resulting in the inability to accurately determine the specific start and end times. In order to determine whether the vehicle is carrying out large-scale transportation, the present invention compares the vehicle's weighing weight and the large-scale transportation weight recorded in the vehicle pass according to a fixed period, infers and updates the vehicle's large-scale transportation status, and identifies abnormal operation behavior during vehicle large-scale transportation. Specifically includes the following steps:

[0059] S5. Determine the upper limit WTMax (WeightTimeMax) and the lower limit WTMin (WeightTimeMin) of the weighing time WeightTime of the weighing data to be connected, and connect the weighing data WeightData whose weighing time is greater than or equal to WTMin and less than or equal to WTMax.

[0060] The method for determining the lower limit WTMin and upper limit WTMax of the weighing time of the accessed weighing data WeightData is as follows:

[0061]

[0062] The method of selecting the weighing data WeightData according to the weighing time WeightTime is as follows:

[0063]

[0064] Where t is the current time, Tw is the vehicle weighing data access time, and WeightTime is the weighing time.

[0065] S6. Delete the weighing data WeightData whose weight is not greater than B tons, and for the supervised vehicles with multiple weighing records, only keep the latest weighing record to improve the subsequent matching speed. B is preset to 20-50, and is selected according to needs. This embodiment takes 40 as an example for explanation.

[0066] S7, traverse the supervised vehicles Ci, and match the weighing data WeightData with the weight Ws specified in the pass based on the license plate number. If the supervised vehicle has weighing data, update the vehicle's large-scale transport status TransferStatus; otherwise, the large-scale transport status remains unchanged.

[0067] S8. For vehicles whose TransferStatus indicates that they are transporting large items with a weight exceeding the specified weight, the vehicle is deemed to be transporting large items with a weight exceeding the specified weight. If the ratio of the vehicle weighing data W to the weight Ws specified in the pass is less than Z, the vehicle is deemed not to be transporting large items. Z is preset to 0.7-0.99. This embodiment uses 0.7 as an example to explain the large item transport status of the vehicle into three types:

[0068] 1. Overweight transportation in excess of the designated weight. If the vehicle weighs more than the weight specified in the pass, the vehicle is deemed to have overweighted and is under overweight transportation.

[0069] 2. Not transporting large items. If the vehicle weighs less than 70% of the weight specified in the pass, it is deemed that the vehicle is not currently transporting large items;

[0070] 3. Large-item transportation in compliance with the specified weight. If the vehicle weighs no less than 70% of the weight specified in the pass and is less than the weight specified in the pass, the vehicle is deemed to be transporting large-items and the weight meets the weight requirements specified in the pass.

[0071] The formula for judging the status of large-scale transportation of vehicles is as follows:

[0072]

[0073] Where W is the vehicle weight and Ws is the weight specified in the pass.

[0074] For vehicles that are transporting large items (including those that meet the specified weight and those that exceed the specified weight), determine whether their passes are valid. If the pass is invalid (including those that have not been effective or have expired), it is deemed that the vehicle is not transporting large items at the specified time.

[0075] The formula for whether a pass is valid is as follows:

[0076]

[0077] Where t is the current time, EffectTime is the effective date of the pass, and ExpireTime is the expiration date of the pass.

[0078] Update the driving or stopping status of the supervised vehicle at a fixed time period. The vehicle may be in a driving or stopping state during the transportation of large items. The possible abnormal transportation behaviors of the vehicle in different states are different and need to be identified separately. At the same time, since the existing data does not record the driving or stopping status information of the vehicle and the state may change dynamically, the present invention uses the real-time positioning data of the vehicle to calculate the vehicle speed, and infers and updates the driving or stopping status of the supervised vehicle at a fixed time period. Specifically, the following steps are included:

[0079] S9. Determine the upper limit RTMax (RecordTimeMax) and the lower limit RTMin (RecordTimeMin) of the recording time RecordTime of the real-time positioning data to be accessed, and access the positioning data LocationData of the supervised vehicle whose recording time is greater than or equal to RTMin and less than or equal to RTMax.

[0080] The method for determining the lower limit RTMin and upper limit RTMax of the recording time of the access positioning data LocationData is as follows:

[0081]

[0082] The method for selecting the vehicle real-time positioning data LocationData according to the real-time positioning data recording time RecordTime is as follows:

[0083]

[0084] Where T l is the access time of real-time positioning data, RecordTime is the recording time of positioning data, and t is the current time.

[0085] S10, arrange the longitude and latitude sequence of the supervised vehicle Ci within D minutes and the positioning data of the vehicle in positive order according to the recording time, and obtain , D is preset to 2-10 and is selected according to actual needs. This embodiment takes 5 as an example for explanation, where N is the number of longitude and latitude data recorded within 5 minutes. The Manhattan distances of the two adjacent longitude and latitude positions are calculated in sequence, summed up and approximately converted into plane distances in km to obtain the vehicle travel distance M1 within 5 minutes.

[0086] The specific calculation formula for calculating the Manhattan distance of the two longitude and latitude positions of the adjacent records in sequence, summing them up and approximately converting them into plane distance is as follows:

[0087]

[0088] In the formula and is the longitude information of two adjacent longitude and latitude data, and The latitude information of two adjacent longitude and latitude data, N is the number of longitude and latitude data recorded within 5 minutes, and 100 is the coefficient value for approximately converting the longitude and latitude distance into the plane distance.

[0089] Calculate the difference between the recording time of the first longitude and latitude data and the recording time of the Nth longitude and latitude data in the longitude and latitude sequence of the supervised vehicle Ci within 5 minutes, in hours, to obtain the running time T1. The specific calculation formula is as follows:

[0090]

[0091] In the formula and They are the recording time of the first longitude and latitude data and the recording time of the Nth longitude and latitude data, RecordTime.

[0092] S11, calculate the average speed V1 of the supervised vehicle Ci within 5 minutes, in km / h, and determine if V1 is less than F, the vehicle is considered to be in a stopped state, otherwise, the vehicle is considered to be in a moving state. F is preset to 1-3, and this embodiment takes 2 as an example for explanation.

[0093] The formula for judging the motion state of the supervised vehicle is as follows:

[0094] V1=M1 / T1

[0095]

[0096] For the determined stopped vehicles, determine whether the vehicle journey is completed, and perform an operation on the vehicles whose journey is completed. For stopped vehicles that are in the process of transporting large items, determining whether their journey is completed helps to infer the transportation behavior of the supervised vehicles and reduce the misjudgment of abnormal operating behavior. Since the existing data does not provide a record of the actual large-item transportation time, after determining the start time of the large-item transportation through the weighing data, it is also necessary to infer the end time of the large-item transportation, so as to adjust the large-item transportation status of the vehicle and avoid misjudging that the vehicle is carrying out large-item transportation and thus misjudging that it has abnormal behavior. The present invention infers whether the vehicle has completed large-item transportation by calculating the continuous parking time of the vehicle. The specific steps are as follows:

[0097] S12: If the continuous parking time of the vehicle does not exceed G minutes, G is preset to 120-240 minutes, and this embodiment takes 180 minutes as an example to explain, and the vehicle is continued to be supervised. If the continuous parking time of the vehicle exceeds 180 minutes, it is determined that the current trip of the vehicle is over.

[0098] For vehicles that have completed their trips, extract the destination text information from the designated route for large-scale transportation. For example, the designated route is , extract "company" as the destination for large-scale transportation. Use geocoding algorithm to convert the text information of the destination for large-scale transportation into the longitude and latitude information with the same coordinate system as the longitude and latitude of the vehicle location and .

[0099] Calculate the Manhattan distance M2 between the latitude and longitude recorded at the last point of the real-time positioning data of the supervision vehicle and the latitude and longitude of the destination for large-scale transportation, and convert it approximately into a plane distance in km. The calculation formula is as follows;

[0100]

[0101] In the formula and The last longitude and latitude recorded for real-time positioning data, and are the longitude and latitude of the destination for large-scale transportation, and 100 is the coefficient value for approximately converting the longitude and latitude distance into the plane distance.

[0102] If M2 is not greater than H, the vehicle is considered to have arrived at the destination for large-scale transportation, and the vehicle's large-scale transportation is recorded as completed, and no further monitoring is performed; otherwise, the vehicle is considered to have not arrived at the destination for large-scale transportation, and the vehicle's large-scale transportation status and continuous parking time are updated to "no longer continuing large-scale transportation" and 0. H is preset to 1-3km, and this embodiment takes 2km as an example for explanation. The specific judgment formula for whether the large-scale transportation destination has been reached is as follows:

[0103]

[0104] Determine the road the vehicle is on and whether it is driving according to the route specified in the pass. For vehicles that are continuing to transport large items, the most important thing is to determine whether they are driving according to the specified route. Since the record format of the specified route is road text, and the record format of the vehicle location is longitude and latitude, the format needs to be converted to achieve location matching. The specific steps are as follows:

[0105] S13. For the supervised vehicle in driving state, select the latest latitude and longitude information and As the vehicle's current latitude and longitude and , determine the road R where the vehicle is located based on the longitude and latitude information of the vehicle's current position; extract all road name information from the specified route text Extract all road name information from the specified route text , the starting point of the large-scale transportation is the airport, is G328, is Chang'an South Road, and the end point of the large-scale transportation is the company.

[0106] Select the current road R and the j-th road Rj (j takes ) in sequence, and perform the following steps:

[0107] S131. Compare the character lengths of the current road R and the j-th road Rj. Mark the one with the longer length as L and the shorter one as S, and record the character length of S as X;

[0108] Define the formula for calculating the character length of the text as .

[0109]

[0110] For example, when the road R where the vehicle is located is Chang'an Road, len(R) = 3, when it is Chang'an South Road, len( ) = 4.

[0111] The judgment formulas for L and S are as follows:

[0112]

[0113] For example, for Chang'an Road R and Chang'an South Road , since len( ) > len(R) = 3, so "Chang'an South Road" is marked as L, and "Chang'an Road" R is marked as S, .

[0114] Judge whether each character s of S is in L in sequence. If so, store the character s in the empty set O. After the judgment is completed, if the set O is empty, the matching rate MatchRate between L and S is 0, and go to step S134;

[0115] For example, the three characters "Chang", "An", and "Road" in the short character "Chang'an Road" all appear in the long character "Chang'an South Road", and "Chang", "An", and "Road" are stored in the set O.

[0116] S132. Sequentially extract each character s in set O, determine its position in L, and starting from this position, continuously extract X characters from L (including character s) to form a string Lx, and store it in set U. If the number of characters extracted from this position is less than X, then only extract to the last character of L. After completion, calculate the number of strings in set U as u.

[0117] For example, sequentially extract "Chang", "An", "Lu" from set O. Since , starting from the position corresponding to the long character "Chang'an South Road", extract 3 characters, namely "Chang'an South", "An Nan Lu", and "Lu". Since "Lu" is the last character of "Chang'an South Road", the obtained Lx is "Lu".

[0118] S133. Traverse the k-th Lx in set U (k takes ), respectively extract the consecutive first 1, 2,..., X characters in Lx to form X strings with lengths of 1, 2,..., X respectively, and sequentially determine whether they exist in S, and record the longest length of the strings existing in S. The matching length X of the k-th Lx and S k . After traversing Lx, accumulate all the matching lengths of each Lx and S, and divide by the number of characters in S to obtain the matching rate MatchRate of L and S.

[0119] The matching rate of the current road R and the j-th road Rj is calculated as follows:

[0120]

[0121] X k is the matching length of S and the k-th Lx, and k takes .

[0122] For example, first sequentially extract three strings with lengths of 1, 2, and 3, namely "Chang", "Chang'an", and "Chang'an South", from "Chang'an South", and determine whether "Chang", "Chang'an", and "Chang'an South" exist in the short character "Chang'an Road". Among them, "Chang" and "Chang'an" exist, while "Chang'an South" does not exist. Record the longest number of characters existing in the short character as 2. Then perform the same processing on "An Nan Lu" and "Lu", and obtain 1 and 1 respectively. The final matching rate is max{2, 1, 1} / len(Chang'an Road) = 0.67.

[0123] S134. Calculate the current road R and the j-th road Rj (j takes ) is the maximum match rate MatchRate. If the maximum MatchRate is greater than Y, the current road of the vehicle is considered to match the road specified in the pass; otherwise, the current road of the vehicle is considered to not meet the specified route. Y is preset to 0.9-0.99, and this embodiment takes 0.95 as an example for explanation.

[0124] The calculation formula for the maximum match rate MatchRate is as follows:

[0125]

[0126] The formula for determining whether the vehicle's current driving road meets the specified route based on the matching rate MatchRate is as follows:

[0127]

[0128] If the vehicle's current driving road does not meet the specified route, when the large-scale transport status TransferStatus is "large-scale transport that meets the specified weight", it is determined that it is not transporting large-scale items according to the designated route; if its large-scale transport status TransferStatus is "large-scale transport that exceeds the specified weight", it is determined that it is not transporting large-scale items according to the designated route and exceeds the specified weight.

[0129] S14: After determining whether all vehicles carrying large items are traveling according to the designated routes, the real-time identification of the four abnormal operation behaviors of large items transportation mentioned above has been achieved. In order to maintain continuous and uninterrupted supervision of large items transportation vehicles, the next execution time will start from S2 to complete the next round of identification of the four abnormal behaviors.

[0130] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for real-time identification of abnormal operation behavior of large-scale transport vehicles, used in a real-time identification system for abnormal operation behavior of large-scale transport vehicles, the real-time identification system includes a processing module, a pass acquisition module, a weighing data acquisition module, a license plate information acquisition module and a positioning data acquisition module; the pass acquisition module is used to obtain large-scale transport pass information, the weighing data acquisition module is used to obtain weighing station weighing data, highway weighing data and / or off-site weighing data, the positioning data acquisition module is used to obtain positioning information of the transport vehicle, and the license plate information acquisition module is used to obtain license plate information of the transport vehicle; the processing module determines the transport vehicle that needs to be supervised according to the expiration date of the pass, determines the load status of the corresponding transport vehicle according to the weight information, license plate information and weighing data specified in the pass, and determines the transport information of the corresponding transport vehicle according to the route information, license plate information and positioning data specified in the pass, characterized in that The following steps are involved: S1. According to the expiration date ExpireTime of the pass, it is extended by A days, and the passes with t≤ExpireTime+A are included in the supervision scope, where t is the current time and the pass entry time Ts is recorded; S2. Determine the interval between the current time t and the pass entry time Ts. If t-Ts≥1 day, access the pass data PassData whose pass verification time VerifyTime is between the pass entry time Ts and the current time t, and update the pass entry time Ts to the current time t. S3. Determine in sequence whether the expiration time of the pass of the supervised vehicle Ci (i=1, 2…, n) exceeds A days. If t>ExpireTime+A, the vehicle Ci is no longer supervised; S4, determine whether there is still a supervised vehicle, if not, go to S2; S5, determining the upper limit WTMax and the lower limit WTMin of the weighing time WeightTime of the weighing data to be accessed, and accessing the weighing data WeightData whose weighing time is greater than or equal to WTMin and less than or equal to WTMax; S6. Delete the weighing data WeightData whose weight is not more than B tons. For the supervised vehicles with multiple weighing records, keep the most recent weighing record; S7, traverse the supervised vehicles Ci, match the weighing data WeightData with the weight Ws specified in the pass based on the license plate information; if the supervised vehicle has weighing data, update the vehicle's large-scale transport status TransferStatus; S8. For a supervised vehicle that exceeds the designated weight, the supervised vehicle is deemed to be transporting large items exceeding the designated weight; for a supervised vehicle that is currently transporting large items, whether its pass is valid is determined. If the pass is invalid, the supervised vehicle is deemed not to be transporting large items at the designated time; S9, determining the upper limit RTMax and the lower limit RTMin of the recording time RecordTime of the real-time positioning data to be accessed, and accessing the positioning data LocationData of the supervised vehicle with a recording time greater than or equal to RTMin and less than or equal to RTMax; S10, arranging the positioning data of the supervised vehicle Ci within D minutes in positive order according to the recording time RecordTime, calculating the Manhattan distances of the two adjacent positioning data in sequence, summing them up and approximately converting them into plane distances, and obtaining the vehicle travel distance M1 within D minutes; calculating the difference between the recording time RecordTime of the first positioning data and the recording time RecordTime of the last positioning data in the positioning data sequence of the supervised vehicle Ci within D minutes, and obtaining the running time T1; S11, calculating the average driving speed V1 of the supervised vehicle Ci within D minutes, and determining that if V1 is less than F, the vehicle is considered to be in a stopped state; otherwise, it is considered to be in a moving state; S12. If the stopped supervised vehicle Ci is stopped for more than G minutes, the current trip of the supervised vehicle is deemed to be over; the text information of the transport destination is extracted from the designated driving route, and the destination text information is converted into the longitude and latitude information having the same coordinate system as the supervised vehicle positioning data; the distance M2 between the longitude and latitude recorded at the last time in the supervised vehicle positioning data and the longitude and latitude of the destination is calculated; if M2 is not greater than H, it is deemed that the supervised vehicle has arrived at the destination, and the heavy-duty transport of the supervised vehicle is recorded to be completed; otherwise, the transport status of the supervised vehicle is changed to no longer continuing heavy-duty transport, and the stop status duration StopTime is changed to 0; S13. For the supervised vehicle in driving state, the latest positioning data is selected, and the road R where the vehicle is located is determined based on the current positioning data of the supervised vehicle; all road name text information is extracted from the designated driving route, and the current road R is selected to be matched with the j-th road Rj in sequence to determine whether the current driving road of the vehicle meets the designated route; if the current driving road of the vehicle does not match the designated route, it is determined that the vehicle is not transporting large items according to the designated route; S14: Go to S2 to perform the next round of recognition.

2. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: In S13, the matching method includes: S131, compare the character lengths of the current road R and the j-th road Rj, mark the longer one as L, and the shorter one as S, and record the character length of S as X; determine in turn whether each character s of S is in L; if so, store the character s in the empty set O; after the determination is completed, if the set O is empty, the match rate MatchRate of L and S is 0, and go to step S134; S132, extract each character s in the set O in turn, determine the position in L, and start from the position, continuously extract X characters from L to form a character string Lx, and store it in the set U; if the number of characters extracted from the position is insufficient, only extract the last character of L; after completion, calculate the number of character strings in the set U as u; S133, traverse the kth Lx in the set U, extract the first 1, 2, ..., X characters in Lx respectively, form X strings of length 1, 2, ..., X respectively, determine whether they exist in S in turn, and record the longest length of the string existing in S as the matching length X between the kth Lx and S k ; After traversing Lx, add up all the matching lengths of each Lx and S, and divide by the number of characters in S to get the matching rate MatchRate of L and S; S134. Calculate the maximum matching rate MatchRate between the current road R and the j-th road Rj. If the maximum MatchRate is greater than Y, it is determined that the current road of the supervised vehicle matches the road specified in the pass; otherwise, it is determined that the current road of the supervised vehicle does not meet the specified route.

3. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: S8 also includes: if the ratio of the weighing data W of the supervised vehicle to the weight Ws specified in the pass is less than Z, it is determined that the vehicle is not currently transporting large items.

4. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: S3 also includes: deleting the pass data with t>ExpireTime+A.

5. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: In S5, for the weighing data WeightData accessed for the first time, WTMin is t-30, and WTMax is t; for the weighing data WeightData not accessed for the first time, WTMin is Tw, and WTMax is t; Tw is the vehicle weighing data access time.

6. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: In S9, for the positioning data LocationData accessed for the first time, RTMin is t-5 and RTMax is t; for the positioning data LocationData not accessed for the first time, RTMin is Tq and RTMax is t; Tq is the access time of the real-time positioning data.

7. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to any one of claims 1 to 6, characterized in that: A is preset to 20-30, B is preset to 20-50, D is preset to 2-10, F is preset to 1-3, G is preset to 120-240, H is preset to 1-3km, Y is preset to 0.9-0.99, and Z is preset to 0.7-0.

99.

8. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 is characterized in that: The real-time identification system also includes a weighing station weighing data uploading module, a highway weighing data uploading module and / or an off-site weighing data uploading module, which is used to upload the weighing station weighing data, highway weighing data and / or off-site weighing data corresponding to the license plate information.

9. The method for real-time identification of abnormal operation behavior of large-scale transport vehicles according to claim 1 or 8, characterized in that: The real-time recognition system also includes a license plate information and positioning data uploading module for uploading the license plate information and positioning information of the corresponding transport vehicle.

Images