CN112650825B - Determination method and device for abnormal driving behavior, storage medium and electronic equipment - Google Patents
Determination method and device for abnormal driving behavior, storage medium and electronic equipment Download PDFInfo
- Publication number
- CN112650825B CN112650825B CN202011615630.1A CN202011615630A CN112650825B CN 112650825 B CN112650825 B CN 112650825B CN 202011615630 A CN202011615630 A CN 202011615630A CN 112650825 B CN112650825 B CN 112650825B
- Authority
- CN
- China
- Prior art keywords
- driving
- abnormal
- target driver
- probability
- order
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 606
- 238000000034 method Methods 0.000 title claims abstract description 154
- 230000006399 behavior Effects 0.000 claims abstract description 122
- 230000008569 process Effects 0.000 claims description 98
- 238000013507 mapping Methods 0.000 claims description 41
- 206010000117 Abnormal behaviour Diseases 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 16
- 238000013135 deep learning Methods 0.000 claims description 11
- 230000005856 abnormality Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000001514 detection method Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010924 continuous production Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Probability & Statistics with Applications (AREA)
- Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- Remote Sensing (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Fuzzy Systems (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Traffic Control Systems (AREA)
Abstract
The application provides a method and a device for determining abnormal driving behavior, a storage medium and electronic equipment, wherein the method comprises the following steps: acquiring track information corresponding to the cancelled target travel order; determining a first abnormal stay probability of abnormal stay of a target driver based on a difference judgment rule between the current running position and a preset running position; if the first abnormal stay probability is greater than a preset probability threshold, determining whether an abnormal driving behavior occurs to the target driver based on the first abnormal stay probability; if the first abnormal stay probability is not greater than the preset probability threshold, determining whether the target driver has abnormal driving connection behaviors according to the second abnormal stay probability output by the pre-trained abnormal stay recognition model. Therefore, the abnormal driving behavior of the target driver can be comprehensively judged according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model under different conditions, and the accuracy rate of recognition of the abnormal driving behavior of the target driver is improved.
Description
Technical Field
The application relates to the technical field of network taxi technologies, in particular to a method and a device for determining abnormal driving behaviors, a storage medium and electronic equipment.
Background
With the rapid development of internet technology, travel services based on the internet technology bring more and more convenience to people's travel, for example, users can travel by bus through a network about car service system. When the 'network about car' service system is used at present, after a user submits a travel order, the travel order is cancelled when the travel order is not normally completed, so that incomplete abnormal orders are generated, and the analysis of cancellation reasons of the abnormal orders has a good guiding effect on avoiding the cancellation of the order when the next dispatch is performed.
At present, the order cancellation reasons are judged by inputting the dimension information of the target driver into a trained recognition model, but the judgment of the model is generally related to the attribute of a training sample, the actual travel scene is complex and changeable, and the sample can not cover all scenes very accurately and completely, so that the recognition accuracy of the recognition of the abnormal driving behavior of the target driver through the recognition model is lower, and how to improve the recognition accuracy of the abnormal driving behavior of the driver in the driving receiving process is a problem to be solved urgently.
Disclosure of Invention
In view of this, the present application aims to provide a method, an apparatus, a storage medium and an electronic device for determining abnormal driving behavior of a target driver, which can comprehensively determine the abnormal driving behavior of the target driver under different conditions according to the driving positions of different times in the track information of the target driver and the driving time corresponding to each driving position, according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model, and is helpful to improve the accuracy of recognition of the abnormal driving behavior of the target driver.
The embodiment of the application provides a method for determining abnormal driving behavior, which comprises the following steps:
acquiring order information of the cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position;
determining a first abnormal stay probability of the target driver when the target driver stops abnormally based on a preset difference judgment rule between the current running position and the preset running position and the running track information of the target driver;
Detecting whether the first abnormal stay probability is larger than a preset probability threshold;
if the first abnormal stay probability is larger than a preset probability threshold, determining whether an abnormal driving behavior occurs to the target driver or not based on the first abnormal stay probability;
if the first abnormal stay probability is not greater than a preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of abnormal stay of the target driver in the order receiving process, and whether the target driver has abnormal driving receiving behaviors is determined based on the second abnormal stay probability.
Further, inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process, including:
inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process; the abnormal stay recognition model is trained based on different historical driving positions in the historical driving track information of the sample target driver in the history cancelled travel order under the non-extreme condition and the historical driving time corresponding to each historical driving position;
Determining a second sub-abnormal stay probability of abnormal driving of the target driver based on a preset driving track movement rule of the target driver in the driving process under the extreme condition and driving track information of the target driver;
and determining a second abnormal stay probability of abnormal stay of the target driver in the order receiving process based on the first abnormal stay probability and the second abnormal stay probability.
Further, the extreme case is a case where at least one of the following conditions exists:
the time difference between the time when the travel order is cancelled and the order taking time is smaller than a preset time threshold;
the distance between the order taking position of the travel order and the boarding position of the passenger is smaller than a preset distance threshold value.
Further, the determining the second sub-abnormal stay probability that the target driver has abnormal driving based on the driving track movement rule of the target driver and the driving track information of the target driver in the preset driving process under the extreme condition includes:
determining at least one abnormal driving position at which driving abnormality occurs at a plurality of driving positions included in the driving locus; the abnormal driving position is a position with the driving speed smaller than the preset driving speed or a position with the position offset planning navigation route;
And determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
Further, the step of inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver in the order receiving process, includes:
determining a plurality of running positions included in the running track information and running time corresponding to each running position;
inputting each driving position and corresponding driving time into a pre-trained abnormal stay recognition model, and determining at least one abnormal driving position;
a first sub-abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of at least one abnormal driving position among the plurality of driving positions.
Further, the second sub-abnormal stay probability that the target driver has abnormal driving is determined based on the driving track movement rule of the target driver and the driving track information of the target driver in the preset driving process under the extreme condition:
determining an updated mapping relation between a driving position and driving time in a driving track movement rule of a target driver based on a travel scene corresponding to the target travel order; wherein the travel scene includes at least one condition affecting a driving trajectory of a target driver;
For each driving position, if the mapping relation between the driving position and the corresponding driving time is inconsistent with the updated mapping relation, determining that the driving position is an abnormal driving position;
and determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
Further, the cancellation party of the target travel order is a passenger;
the travel track information of the target driver is travel track information of the target driver in a period from a time when the target travels the order to a time when the order is canceled.
Further, the determining method further includes:
acquiring a plurality of order attribute information corresponding to the historic cancelled travel orders; the order attribute information comprises at least one of order canceling time between the time when the travel order is cancelled and the order receiving time and a driving receiving distance between the order receiving position of the travel order and the boarding position of the passenger;
determining at least one order cancel time abnormal order with the cancel time smaller than the preset time and at least one order target driver driving distance smaller than a preset distance threshold from the acquired plurality of historic cancelled travel orders;
Filtering at least one order cancel time and at least one cancel time abnormal order from the plurality of historical cancelled travel orders, and determining a plurality of sample cancelled travel orders;
and training the constructed deep learning network based on different historical driving positions in the historical driving track information corresponding to the cancelled travel orders of each sample and the historical driving time corresponding to each historical driving position to obtain an abnormal stay recognition model.
Further, the determining, based on a preset gap judgment rule between the current driving position and the preset driving position and the driving track information of the target driver, the first abnormal stay probability that the target driver has abnormal stay includes:
determining that the driving position is an abnormal driving position if the distance difference between the current driving position and the preset driving position is larger than a preset difference threshold value in the driving time corresponding to each driving position;
a first abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of the at least one abnormal driving position among the plurality of driving positions.
The embodiment of the application provides a determining device for abnormal driving behavior, which comprises:
The order information acquisition module is used for acquiring order information of the cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position;
the first probability determining module is used for determining a first abnormal stay probability of the target driver when the target driver stops abnormally based on a preset difference judging rule between the current running position and the preset running position and the running track information of the target driver;
the probability detection module is used for detecting whether the first abnormal stay probability is larger than a preset probability threshold value or not;
the first abnormal behavior determining module is used for determining whether the target driver generates abnormal driving connection behaviors or not based on the first abnormal stay probability if the first abnormal stay probability is larger than a preset probability threshold;
the second abnormal behavior determining module is used for inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver in abnormal stay in the order receiving process if the first abnormal stay probability is not larger than a preset probability threshold, and determining whether the target driver has abnormal driving behavior or not based on the second abnormal stay probability.
The embodiment of the application also provides electronic equipment, which comprises: the system comprises a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, the processor and the memory are communicated through the bus when the electronic device runs, and the machine-readable instructions are executed by the processor to execute the steps of the method for determining abnormal driving behavior.
The embodiment of the application also provides a computer readable storage medium, and a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to execute the steps of the method for determining abnormal driving behavior.
According to the method, the device, the storage medium and the electronic equipment for determining the abnormal driving receiving behavior, the cancelled target travel order comprises order information of travel track information of a target driver for receiving the target travel order; determining a first abnormal stay probability of abnormal stay of the target driver according to a preset difference judgment rule between the current running position and the preset running position, detecting whether the first abnormal stay probability is larger than a preset probability threshold, and determining whether abnormal driving connection behaviors of the target driver occur according to the first abnormal stay probability if the first abnormal stay probability is larger than the preset probability threshold; if the first abnormal stay probability is not greater than the preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine second abnormal stay probability, whether the target driver has abnormal driving connection behaviors or not is determined according to the second abnormal stay probability, and the abnormal driving connection behaviors of the target driver can be comprehensively judged according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model under different conditions, so that the accuracy of recognition of the abnormal driving connection behaviors of the target driver is improved.
Further, in the processing process of the abnormal stay recognition model, a first sub abnormal stay probability of abnormal stay of the target driver in the process of receiving the driving is determined in advance through the abnormal stay recognition model, a second sub abnormal stay probability of abnormal stay of the target driver in the process of receiving the driving is determined according to a driving track movement rule of the target driver in the process of traveling under the opposite terminal condition, the abnormal stay probability output by the abnormal stay recognition model is determined for the target driver according to the first sub abnormal stay probability and the second sub abnormal stay probability, whether abnormal receiving driving behaviors occur or not is judged according to the abnormal stay probability output by the abnormal stay recognition model, and the situation that the abnormal stay recognition model cannot completely cover all situations and further misjudgment occurs can be made up through a mode of combining judgment under the non-extreme condition with judgment of rules under the extreme condition, so that the accuracy of recognition of the abnormal receiving the driving behaviors of the target driver is improved.
In order to make the above objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a method for determining abnormal driving behavior according to an embodiment of the present application;
FIG. 2 is a flowchart of another method for determining abnormal driving behavior according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a determining device for abnormal driving behavior according to an embodiment of the present application;
FIG. 4 is a second schematic structural diagram of a determining device for abnormal driving behavior according to an embodiment of the present disclosure;
fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
Detailed Description
For the purposes of making the objects, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are only some embodiments of the present application, but not all embodiments. The components of the embodiments of the present application, which are generally described and illustrated in the figures herein, may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, as provided in the accompanying drawings, is not intended to limit the scope of the application, as claimed, but is merely representative of selected embodiments of the application. Based on the embodiments of the present application, every other embodiment that a person skilled in the art would obtain without making any inventive effort is within the scope of protection of the present application.
It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the presence of the features stated hereinafter, but not to exclude the addition of other features.
The terms "passenger," "requestor," "service requestor," and "customer" are used interchangeably herein to refer to a person, entity, or tool that may request or subscribe to a service. The terms "target driver," "provider," "service provider," and "provider" are used interchangeably herein to refer to a person, entity, or tool that can provide a service. The term "user" in this application may refer to a person, entity, or tool requesting, subscribing to, providing, or facilitating the provision of a service. For example, the user may be a passenger, driver, operator, etc., or any combination thereof. In this application, "passenger" and "passenger terminal" may be used interchangeably, and "driver" and "driver terminal" may be used interchangeably.
The terms "service request" and "order" are used interchangeably herein to refer to a request initiated by a passenger, service requester, driver, service provider, or vendor, etc., or any combination thereof. Accepting the "service request" or "order" may be a passenger, a service requester, a driver, a service provider, a vendor, or the like, or any combination thereof. The service request may be either fee-based or free.
With the rapid development of internet technology, travel services based on the internet technology bring more and more convenience to people's travel, for example, users can travel by bus through a network about car service system. When the 'network about car' service system is used at present, after a user submits a travel order, the travel order is cancelled when the travel order is not normally completed, so that incomplete abnormal orders are generated, and the analysis of cancellation reasons of the abnormal orders has a good guiding effect on avoiding the cancellation of the order when the next dispatch is performed.
It should be noted that, before the application is filed in the present application, the abnormal driving behavior recognition of the target driver side is generally judged by setting a plurality of judgment rules according to different travel scenes, and because a plurality of travel scenes exist in the travel process and travel problems encountered under each travel scene are different, different judgment rules need to be set for different travel scenes and different travel problems, for example, different road congestion conditions corresponding to different travel time need to be set for judgment, and rule 1: under the condition that a road is not congested, the stay time of a target driver at the same position is too long, so that the abnormal driving behavior is realized; rule 2: under the condition that a road is congested, a target driver stays at a congested position for a long time, the congestion condition of the road is referred, the long-time stay of the target driver at the congested position is determined to be normal, and the number of judging rules which are required to be set for different travel scenes is large;
Along with the development of neural network deep learning technology, the abnormal driving behavior of the target driver can be identified through a pre-trained identification model, but the pre-trained identification model learns the data features in a large amount of sample data to give out the corresponding judgment result on the statistical level, the identification result of the identification model depends on the features of the sample data, and for the accuracy of the model in the model training process, sample data under extreme scenes (such as scenes where the existence time of a travel order is short and the driving behavior of the target driver cannot be estimated) which possibly affect the accuracy of the model are not included in the model training process, and due to the complexity of the travel scene, the judgment accuracy of the abnormal driving behavior of the target driver is low under the situation that misjudgment can occur on the driving data under some extreme scenes in the specific application process of the identification model.
Based on the above, the embodiment of the application provides a method for determining abnormal driving behavior, which can comprehensively determine the abnormal driving behavior of a target driver under different conditions according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model, and is beneficial to improving the accuracy of recognition of the abnormal driving behavior of the target driver.
Referring to fig. 1, fig. 1 is a flowchart of a method for determining abnormal driving behavior according to an embodiment of the present application. As shown in fig. 1, the method for determining abnormal driving behavior provided in the embodiment of the present application includes:
s101, acquiring order information of a cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position.
S102, determining a first abnormal stay probability of the target driver based on a preset difference judgment rule between the current running position and the preset running position and the running track information of the target driver.
S103, detecting whether the first abnormal stay probability is larger than a preset probability threshold.
And S104, if the first abnormal stay probability is larger than a preset probability threshold, determining whether an abnormal driving behavior occurs to the target driver based on the first abnormal stay probability.
S105, if the first abnormal stay probability is not greater than a preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver in abnormal stay in the order receiving process, and whether the target driver has abnormal driving approaching behaviors is determined based on the second abnormal stay probability.
In step S101, the target order to be cancelled may be an order in which the passenger issues a cancellation request or an order in which the target driver issues a cancellation request. When determining that the order is cancelled, the running positions at different times after receiving the target travel order during the cancellation of the order and the running time corresponding to each running position need to be acquired.
Here, the cancellation party of the target travel order may be the target driver or the passenger who places an order, when the cancellation party is the target driver end, there may be a situation that the target driver finds that the passenger cannot reach the passenger point after receiving the order, and the cancellation from the target driver end will not be the cancellation of the order by the target driver only in a long period of time, so the judgment of the abnormal stay behavior of the target driver will not be analyzed from the cancellation from the target driver end; in consideration of the perception of passengers on service and objective analysis of abnormal stay of a target driver, the method and the device mainly analyze abnormal driving receiving behaviors of the target driver under the condition that the passengers cancel orders subjectively.
Here, the travel track information of the target driver is travel track information of the target driver in a period from a time when the target travels the order to a time when the order is canceled.
Wherein, when the travel order is generated to cancel the order, the following five time nodes sequentially appear: the method includes the steps of placing an order (a time node when a passenger submits an order request including a travel destination and an origin of the travel), taking the order (a time scene when a target driver receives the target travel order and determines to take the order), canceling the order (an event scene when the passenger cancels the order through a handheld intelligent terminal), completing the order (a time node when the target driver sends the passenger to the travel destination in the order request made by the passenger after taking the order and the passenger finishes paying the order normally), and judging the responsibility of the order (a time node when the responsibility is judged (the target driver or the passenger) after the responsibility is not normally finished on an order platform), wherein canceling the order and judging the responsibility of the order by the existence of abnormal disputes can generally occur in the order canceling node and the responsibility of the order, and the travel track information of the target driver acquired on each node is different according to the difference of the time interval between each node and the order receiving the order.
Here, the driving track information includes driving positions of the target driver at different times after receiving the target travel order and driving times corresponding to each driving position, so that the driving progress of the target driver in the order execution time can be reflected, and further, whether the target driver has abnormal stay abnormal behaviors in the order receiving process or not is determined.
The driving positions of the target driver at different times in the driving process of the target driver and the driving time corresponding to each driving position are space-time characteristics, the driving positions of the target driver at a certain driving time are indicated, and the driving condition of the target driver can be determined through the space-time characteristics.
Wherein, the driving condition comprises normal driving, abnormal driving and the like; the normal driving condition comprises the condition that the driving speed of a target driver in a driving receiving section is consistent with the historical driving speed of the target driver in a similar historical driving receiving section, and the target driver does not wander at one or a plurality of positions within a certain distance from the target driver to the boarding point of a passenger after receiving a bill; the abnormal driving condition comprises the condition that a target driver is loitered at one or a plurality of positions within a certain distance from the getting-on point of a passenger after receiving a bill; or after the system plans the driving route from the target driver position to the passenger boarding position for the target driver, the target driver does not drive according to the driving route (yaw driving). Through the definition of normal running and abnormal running, whether the target driver is in normal running or abnormal running can be determined directly according to the driving behaviors of the target driver in the follow-up, so that the abnormal driving behavior of the target driver can be analyzed more accurately.
Here, the driving location may be an absolute location during driving, for example, a location identified by coordinates on a map or a location area divided by different street areas; the present position may be a plurality of travel positions determined by the distance between the current position and the destination driver's destination position, for example, a position 100 meters away from the destination position, a position 500 meters away from the destination position, or the like; it may also be a plurality of travel positions determined by the distance between the current position and the boarding position of the passenger, for example, a position 100 meters from the boarding position, a position 500 meters from the boarding position, etc.
Here, step S101 obtains order information of the order cancelled layer when the order is cancelled, further obtains travel track information of the order receiving target driver from the order information, and determines the probability that the target driver has abnormal travel conditions (such as abnormal stay) in the process of receiving the order and cancelling the order by analyzing travel positions of the target driver at different times indicated in the travel track information and travel times corresponding to each travel position, thereby determining the abnormal stay condition of the target driver, so that the platform determines the responsibility of the target driver according to the abnormal stay condition of the target driver when the order is cancelled later.
In step S102, according to the rule of determining the difference between the current driving position and the preset driving position, the mapping relationship between the difference between the current driving position and the preset driving position and the abnormal stay behavior may be determined, so as to determine the first abnormal stay probability of the target driver.
Here, for different travel scenes, the difference judgment rule between the current travel position and the preset travel position may be appropriately adjusted, so as to improve the accuracy of the abnormal driving judgment of the target driver.
Here, the travel scenario corresponding to the target travel order may be determined by the road congestion condition of the current target driver, the travel time of the target travel order, and the travel area where the target travel order is located.
The road congestion situation of the current target driver may cause that the driving speed of the target driver at one or a plurality of driving positions is slower in the driving receiving process, and further, the corresponding driving time node does not reach the corresponding driving position, and when the difference judgment rule between the current driving position and the preset driving position is set, the driving position where congestion is likely to occur and the driving time interval between the adjacent driving positions need to be lengthened.
In step S103, it is detected whether the first abnormal stay probability output by the judgment rule is greater than the preset probability threshold, so as to determine whether the abnormal driving behavior of the target driver has occurred under the judgment condition of the judgment rule, and further determine the judgment basis for finally confirming the abnormal driving behavior of the target driver according to that determination mode (rule judgment or model judgment).
In step S104, if the first abnormal stay probability is greater than the preset probability threshold, it is indicated that the target driver has abnormal stay under the rule of difference judgment between the current driving position and the preset driving position, so that it can be determined whether the target driver has abnormal driving behavior through the first abnormal stay probability.
Here, the abnormal driving behavior of the target driver can be determined earlier according to the judgment of the difference judgment rule between the current driving position and the preset driving position, so that the efficiency is high; meanwhile, the judgment based on the difference judgment rules between the current running position and the preset running position is aimed at the judgment of the corresponding scene, unlike the abnormal stay recognition model which needs to rely on historical labels, the judgment accuracy of the abnormal stay behavior of the target driver is higher.
In step S105, if the first abnormal stay probability is not greater than the preset probability threshold, it is indicated that the target driver does not have abnormal stay under the rule of determining the difference between the current driving position and the preset driving position, so that the abnormal stay recognition model is required to determine the abnormal stay behavior of the target driver, and whether the target driver has abnormal contact driving behavior is determined according to the fourth abnormal stay probability output by the abnormal stay recognition model.
Here, after the driving track information of the target driver receiving the cancelled order acquired in step S101 is input into the pre-trained abnormal stay recognition model, the second abnormal stay probability of the target driver in the abnormal stay process is output through the abnormal stay recognition model, and the second abnormal stay probability of the target driver in the abnormal stay process can be determined through the second abnormal stay probability output by the model.
Here, the abnormal stay recognition model may be any one of deep learning networks, and the abnormal stay recognition model is trained by the correspondence relationship between the acquired historic cancelled order and the different historic travel positions included in the order and the historic travel time corresponding to each of the historic travel positions.
In the process of the abnormal stay recognition model, the non-extreme situation is processed first, so that the first sub-abnormal stay probability is determined, meanwhile, the second sub-abnormal stay probability is determined through the judging rule under the extreme situation, the first sub-abnormal stay probability is judged and corrected through the second sub-abnormal stay probability, and the second abnormal stay probability with higher accuracy is determined to be used as the output of the abnormal stay recognition model.
Here, the extreme case means that there is a case where travel track information for performing driving behavior analysis on the target driver cannot be acquired, and it may be defined by a distance between the target driver order receiving position and the passenger boarding position, or may be determined by a time difference between the target driver order receiving time and the order canceling time, and the non-extreme case means a case where travel track information for performing driving behavior analysis on the target driver can be acquired, and generally a length of a time period from an order start (order placing) time to an order end of a travel order is relatively long in the non-extreme case.
When the distance between the pick-up point of the target driver and the boarding point of the passenger is defined as the non-extreme condition, a preset distance threshold is required to be set, the preset distance threshold is the minimum distance that the target driver can normally drive and move, and when the distance between the pick-up position of the target driver and the boarding position of the passenger is greater than the preset distance threshold, the non-extreme condition is determined; for the non-extreme case, when the time difference between the order taking time and the order canceling time of the target driver is defined, a preset time threshold is required to be set, wherein the preset time threshold is the minimum time period in which the target driver can normally drive and move, and when the time difference between the order taking time and the order canceling time of the target driver is greater than the preset time threshold, the non-extreme case is determined.
When the abnormal stay recognition model starts to recognize the abnormal stay behavior of the target driver, after the travel order is cancelled, the travel track information of the target driver in the period from taking the order to canceling the order is collected, and when the abnormal stay recognition model is used for judging the abnormal stay behavior of the target driver, the abnormal stay recognition model is uniform in time dimension, more observation time is given to the travel behavior of the target driver, and the accuracy of recognizing the abnormal stay behavior of the target driver under the non-extreme condition is ensured to a certain extent.
Further, in the method for determining abnormal driving behavior, the step of inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver for abnormal stay in the order receiving process, and determining whether the target driver has abnormal driving behavior based on the second abnormal stay probability includes:
a1: inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process; the abnormal stay recognition model is trained based on different historical driving positions in the historical driving track information of the sample target driver in the history cancelled travel order under the non-extreme condition and the historical driving time corresponding to each historical driving position.
b1: and determining a second sub-abnormal stay probability that the target driver is abnormally driven based on the driving track movement rule of the target driver and the driving track information of the target driver in the driving process under the preset extreme condition.
c1: and determining a second abnormal stay probability of abnormal stay of the target driver in the order receiving process based on the first abnormal stay probability and the second abnormal stay probability.
In step a1, after the acquired driving track information of the target driver receiving the cancelled order is input into an abnormal stay recognition model trained in advance, a first sub-abnormal stay probability of the target driver in abnormal stay in the order receiving process is output through the abnormal stay recognition model, and the first sub-abnormal stay probability of the target driver in abnormal stay in the order receiving process can be determined through the first sub-abnormal stay probability output by the model.
The first sub-abnormal stay probability under the non-extreme condition is determined first, the first sub-abnormal stay probability is obtained based on the historical sample information training of the abnormal stay recognition model, the determination of the abnormal stay behavior of the target driver based on the first sub-abnormal stay probability in the process can ensure the accuracy under the non-extreme condition, but under some extreme scenes, the judgment of the first sub-abnormal stay probability on the abnormal stay behavior of the target driver has a certain error judgment rate, and after the first sub-abnormal stay probability is output, the first sub-abnormal stay probability is judged according to a preset rule and then corrected, so that the output of the abnormal stay recognition model with higher accuracy is determined jointly.
In the step b1, the driving track movement rule for the target driver under the extreme condition is determined through the historical behavior track of the target driver under the extreme condition, the abnormal stay behavior of the target driver is judged according to the driving track movement rule and the driving track information of the target driver under the extreme condition, the probability of misjudgment of the abnormal stay recognition model on the abnormal stay behavior of the target driver under the extreme condition is avoided, the inaccuracy of the model judgment under the extreme condition of the abnormal stay recognition model is made up through the rule judgment, and the accuracy of the final determination of the abnormal stay behavior of the target driver in the driving receiving section is improved.
Here, the extreme case may include judging features in both time and space, and the extreme case in the time dimension means that the time difference between the time when the travel order is cancelled and the travel order taking time is smaller than a preset time threshold; the extreme case in the space dimension is that the distance between the order receiving position of the target driver and the boarding position of the passenger is smaller than a preset distance threshold value; and the extreme case may also include a case where the temporal feature and the spatial feature are combined, i.e., the time difference between the time when the travel order is cancelled and the time when the travel order is taken is less than a preset time threshold while the distance between the pick-up position of the target driver and the boarding position of the passenger is less than a preset distance threshold.
The preset distance threshold is the minimum distance from which the driving track of the target driver can be obtained; the preset time threshold is the length of the minimum period of time that the target driver can acquire the target driver's travel track. The case where the target driver's travel locus can be acquired is a case where the target driver starts driving and does not encounter any interruption (road congestion, vehicle failure, etc.) of the travel operation.
For example, in the time dimension, after the passenger makes an order through the platform, the target driver receives the target order through the platform, and when the target driver receives that the target order does not start to travel to the boarding point where the passenger is located, the passenger suddenly cancels the situation of the order; under the space dimension, after the target driver receives the passenger, determining that the receiving point is the same as the boarding position of the passenger, and after the target driver receives the passenger normally, clicking a button for confirming arrival, so that the target driver is considered to not arrive at the boarding point of the passenger for order receiving; in the case of a combination of the time dimension and the space dimension, the order taking position of the target driver is near the boarding position of the passenger, and the passenger cancels the order shortly after the passenger places the order, resulting in the target driver being cancelled before the passenger is not received.
Here, when the abnormal stay of the target driver under the non-extreme condition is determined, the driving track movement rule corresponding to the target order needs to be determined by combining the driving track movement rule of the target driver and considering the driving condition of the target driver under the current order, the travel time of the target order, the road congestion condition and the travel scene of the target order.
Here, when the abnormal stay behavior of the target driver is judged by the rule, the rule is not a conclusion obtained by the statistical probability like the recognition model, but the rule is judged according to the specific business scene setting, and the method has the characteristics of simple calculation, interpretation, light weight and convenient deployment, so that the rule can be adjusted according to the judgment rules of different travel scenes and the definition scale of the abnormal stay behavior of the target driver.
For example, the travel time of the target order belongs to an early peak period, and the travel route of the passenger passes through a stage of relatively crowded roads, and a long-time stay behavior of the target driver may occur during driving, but the stay cannot be considered as an abnormal stay condition of the target driver, and when judgment is performed by rules, misjudgment conditions caused by travel scenes and the like are required.
Here, since the rule judgment is that a judgment process is given from the data layer, that is, a result is given according to the judgment rule when the rule judgment is performed, the reason that the target driver generates abnormal stay can be seen, and the method has better interpretation than the method that the recognition model directly outputs the probability result of the abnormal stay of the target driver.
For example, the rules determine that continuous movement is required for a certain period of time, but the target driver does not move continuously for the certain period of time and stays at a certain position for too long, and then it can be seen that the target driver is determined that the target driver has abnormal stay when receiving a bill because of too long stay time for a certain period of time.
In step c1, a second abnormal stay probability output by the abnormal stay recognition model is determined according to the first sub abnormal stay probability determined in step a1 and the second sub abnormal stay probability determined in step Sb 2. The first sub-abnormal stay probability determined under the non-extreme condition by the abnormal stay recognition model and the second sub-abnormal stay probability under the extreme condition by the rule are used for comprehensively determining the second abnormal stay probability output by the abnormal stay recognition model, the abnormal driving behavior occurrence probability of the target driver is considered from the non-extreme condition and the extreme condition, the abnormal driving behavior occurrence condition of the target driver is more comprehensively analyzed, and the accuracy of the second abnormal stay probability output by the abnormal stay recognition model can be improved through the judgment of the non-extreme condition and the complementation interpretation of the driving track movement rule of the target driver under the extreme condition.
Here, the abnormal driving behavior refers to an abnormal situation that a driving track of the target driver occurs in the driving process, and the abnormal driving behavior may be that the target driver does not travel according to a planned driving route (i.e. yaw behavior occurs); or the target driver wanders in a certain area or stays too long in the process of taking over the drive, etc.
After the target driver receives the order, a plurality of driving positions in the driving track of the target driver in the normal driving process and the driving time corresponding to each driving position are planned according to the current road condition and the driving speed of the target driver for driving the vehicle, and when the target driver does not reach the corresponding driving position at a certain time point, the target driver is considered to have abnormal driving receiving behaviors with a high probability.
For example, according to the distance between the order receiving position and the boarding position of the passenger, the current road condition and the speed of the target driver when driving the vehicle to normally run, three driving positions are determined: the position A, the position B and the position C are respectively connected with the target driver according to the distance between the position A, the position B and the position C, and the running time when the position A is reached is determined to be 30 minutes after the bill is received, the running time when the position B is reached is determined to be 70 minutes after the bill is received, and the running time when the position C is reached is determined to be 100 minutes after the bill is received; according to the fact that the running time when the target driver arrives at the position A is 30 minutes after receiving the bill in the actual running process, the running time when the target driver arrives at the position B is 120 minutes after receiving the bill, and the running time when the target driver arrives at the position C is 500 minutes after receiving the bill, the target driver can be known not to arrive at the position B and the position C at the corresponding time, and abnormal driving receiving behaviors of the target driver after the target driver arrives at the position A can be determined.
Here, the second abnormal stay probability of the target driver for abnormal stay in the order receiving process according to the first sub abnormal stay probability and the second sub abnormal stay probability may be a comprehensive abnormal stay probability determined by weighted summation of the first sub abnormal stay probability and the second sub abnormal stay probability, so that the second abnormal stay probability of the target driver for abnormal stay in the order receiving process is determined according to the comprehensive abnormal stay probability.
Here, in addition to determining the second sub-abnormal stay probability in the extreme case by the first sub-abnormal stay probability in the non-extreme case and the driving locus movement rule of the target driver, the determination result of the driving locus movement rule of the machine may be set to be yes or no and the classification result is output, and the second abnormal stay probability of abnormal stay occurring in the order receiving process is determined by the determined first sub-abnormal stay probability and the rule hit condition.
For example, when the first sub-abnormal stay probability is determined to be greater than the preset probability threshold and the rule result is that abnormal stay occurs, then it may be determined that the second abnormal stay probability of the target driver that abnormal stay occurs in the process of receiving the driver is the first sub-abnormal stay probability; when the first sub-abnormal stay probability is determined to be larger than the preset probability threshold value and the rule result is that abnormal stay does not occur, the second abnormal stay probability that the target driver has abnormal stay in the process of taking over driving can be determined to be 0.
The method for determining the second abnormal stay probability by the first abnormal stay probability and the second abnormal stay probability may be that the first abnormal stay probability and the second abnormal stay probability are weighted and summed by a first weight coefficient corresponding to the first abnormal stay probability and a second weight coefficient corresponding to the second abnormal stay probability, so as to determine the second abnormal stay probability; or the first sub-abnormal stay probability and the second sub-abnormal stay probability are averaged to further determine the second abnormal stay probability.
The method for determining the target abnormal stay probability is preferably a method for carrying out weighted summation on the first sub abnormal stay probability and the second sub abnormal stay probability, in which different weight coefficients can be set under different conditions, so that the influence degree of the first sub abnormal stay probability and the second sub abnormal stay probability on the final second abnormal stay probability determination is regulated, and the accuracy of the target abnormal stay probability determination is improved.
Here, after determining the first sub-abnormal stay probability of the target driver in the abnormal stay process in the order receiving process through the non-extreme case, it may also be determined whether the first sub-abnormal stay probability is greater than a preset probability threshold, if the first sub-abnormal stay probability is not greater than the preset probability threshold, the target driver is considered to have no abnormal stay behavior at this time, and at this time, the determination process of the second sub-abnormal stay probability of the target driver in the abnormal stay process according to the driving track movement rule of the target driver and the driving track information of the target driver in the extreme case is not performed.
Further, in the method for determining abnormal driving behavior provided in the embodiment of the present application, the determining, based on the driving track movement rule of the target driver and the driving track information of the target driver in the preset driving process under the extreme condition, the second sub-abnormal stay probability that the target driver has abnormal driving includes:
a2: determining at least one abnormal driving position at which driving abnormality occurs at a plurality of driving positions included in the driving locus; the abnormal driving position is a position with the driving speed smaller than the preset driving speed or a position with the position offset planning navigation route.
b2: and determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
In step a2, considering that the time difference between the time when the travel order is cancelled and the order taking time in the extreme case is smaller than the preset time threshold, when the time from the order placement to the cancellation is very short, it is very difficult to acquire the travel time corresponding to the track point from the track information of the target driver, so in this case, it is necessary to determine at least one abnormal travel position of the travel abnormality at a plurality of travel positions, and thus determine the second sub-abnormal stay track in the extreme case according to the abnormal travel position.
Here, the at least one abnormal travel position at which the traveling abnormality occurs may be a position at which the traveling speed of the target driver is slow, a position at which the target driver deviates from a preset order taking route, or the like.
Wherein, when the driving is slower at a certain position, the driving speed of the target driver is smaller than the normal driving speed when the road condition does not influence the normal driving of the target driver; the deviation of the target driver from the preset pick-up route refers to the situation that the target driver is obviously deviated from the pick-up route on the premise that the road does not cause the target driver to have to detour.
Here, in the case where the time difference between the time when the travel order is canceled and the order taking time is smaller than the preset time threshold, the target driver may not travel a long distance even during normal travel, and the division of the plurality of travel positions may be a division according to respective proportions so that the plurality of travel positions are divided in the connecting routes of different lengths.
For example, it may be set to 10 travel positions into which the junction road section is divided when the junction travel distance is less than 1000 m; when the driving distance is larger than 1000m and smaller than 3000m, the driving road section is divided into 15 driving positions and the like.
The dividing of the plurality of driving positions may be to divide the driving route equally, or may be to adjust the distance length between every two driving positions in real time according to the real-time road condition of the driving route. For example, on a more congested road section, there may be no movement of the vehicle for a long period of time, in which case the distance division length between two adjacent travel positions may be lengthened.
In step a2, considering that, in the extreme case that the distance between the pick-up position of the travel order and the boarding position of the passenger is smaller than the preset distance threshold, when the distance between the pick-up position of the target driver and the boarding position of the passenger is smaller than the preset threshold, the target driver may not make an effective moving track on the arrival boarding position for the situation that the pick-up position of the target driver is extremely close to the boarding position of the passenger, at this time, the driving position to be acquired may be an absolute position, which may include the pick-up position of the target driver and the boarding position of the passenger, and it is required to determine that the distance between the pick-up position and the boarding position of the passenger, if the distance between the two positions is smaller than the preset distance threshold, and the arrival command sent by the target driver end is not received, this is to determine that the target driver is an abnormal driving position at the pick-up position, where the target driver is not displaced and has no driving speed.
Here, when the distance between the order receiving position of the target driver and the boarding position of the passenger is smaller than the preset threshold, a certain driving track exists between the order receiving position of the target driver and the boarding position of the passenger, and at this time, it is required to determine the abnormal stay behavior of the target driver according to the corresponding relationship between the driving position and the driving time, and the determining manner is consistent with the abnormal driving position determining step in which the time difference between the time when the travel order is cancelled and the order receiving time is smaller than the preset time threshold, which is not repeated herein.
In step b2, a second sub-abnormal stay probability of the abnormal driving of the target driver is determined according to the proportion of the at least one abnormal driving position determined in step a2 in the plurality of driving positions. The abnormal driving position represents the abnormal driving tendency of the target driver during driving, and the existence of the individual abnormal driving position only represents the possible abnormal situation of the target driver at the position, but the abnormal driving situation possibly does not exist during the whole driving process, so that the proportion of the abnormal driving position in a plurality of driving positions needs to be determined, and the second sub abnormal stay probability is further determined.
Here, the proportion of the abnormal travel position among the plurality of travel positions may be directly regarded as the second sub abnormal stay probability in the extreme case, that is, the proportion of the abnormal travel position among the plurality of travel positions is equal to the second sub abnormal stay probability; the mapping relation between the proportion of the abnormal driving position in the plurality of driving positions and the second sub-abnormal stay probability may be preset, so that the second sub-abnormal stay probability is determined from the proportion of the abnormal driving position in the plurality of driving positions through the mapping relation.
The method for determining the second sub-abnormal stay probability is preferably to determine the second sub-abnormal stay probability according to a mapping relation between the proportion of the plurality of driving positions and the second abnormal stay probability, because the second sub-abnormal stay probability which is similar and is based on the proportion of the determined at least one abnormal driving position in the plurality of driving positions and represented by the proportion of the target driver is consistent, and the second sub-abnormal stay probability is determined more accurately according to the mapping relation.
Here, since the abnormal stay behavior is a continuous process, it may also be to determine the number of continuously occurring abnormal travel positions, and determine the second sub-abnormal stay probability that the target driver has abnormal driving according to the number of abnormal travel positions and the preset mapping relationship between the number of abnormal travel positions and the second abnormal stay probability.
In the process of judging the driving track movement rule of the target driver and the driving track information of the target driver in the driving process, the abnormal driving position of the target driver is determined according to the time-space characteristics (the corresponding relation between the driving position and the driving time) of the successive driving of the target driver in the history process, so that the second sub-abnormal stay probability of abnormal driving of the target driver is determined, the judgment is carried out according to the time-space characteristics in the history driving track, the dependence of the abnormal stay behavior on real-time track service is reduced, and the usability of the abnormal stay behavior identification can be improved to a certain extent.
Further, in the method for determining abnormal driving behavior in the embodiment of the present application, the inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a single receiving process includes:
a3: a plurality of travel positions included in the travel locus information and travel times corresponding to each of the travel positions are determined.
b3: and inputting each driving position and corresponding driving time into a pre-trained abnormal stay recognition model, and determining at least one abnormal driving position.
c2: a first sub-abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of at least one abnormal driving position among the plurality of driving positions.
In step a3, a plurality of traveling positions included in the traveling locus information and traveling time of each traveling position are determined, and a first sub-abnormal stay probability is determined by determining a plurality of abnormal traveling positions among the plurality of traveling positions.
Here, the division of the plurality of travel positions may be a division according to a corresponding ratio such that the plurality of travel positions are divided in the joint driving routes of different lengths; the dividing of the plurality of driving positions can be to divide the driving route evenly, and can also be to adjust the distance length between every two driving positions in real time according to the real-time road condition of the driving route. For example, on a more congested road section, there may be no movement of the vehicle for a long period of time, in which case the distance division length between two adjacent travel positions may be lengthened.
In step b3, each driving position and corresponding driving time are input into a pre-trained abnormal stay recognition model to obtain at least one abnormal driving position, and then a first sub abnormal stay probability is determined according to the abnormal driving position.
Here, since the determination of whether or not the travel position is the abnormal travel position by the abnormal stay recognition model needs to be made in combination with the travel position before and after the travel position and the travel time reference determination, when each travel position is input into the abnormal stay recognition model, the travel positions need to be sequentially input into the abnormal stay recognition model.
The order of inputting each driving position may be from near to far, or from first to last, according to the driving time.
In step c2, a first sub-abnormal stay probability of the abnormal driving of the target driver is determined according to the proportion of the at least one abnormal driving position determined in step b3 in the plurality of driving positions. The abnormal driving position represents the abnormal driving tendency of the target driver during driving, and the existence of the individual abnormal driving position only represents the possible abnormal situation of the target driver at the position, but the abnormal driving situation possibly does not exist during the whole driving process, so that the proportion of the abnormal driving position in a plurality of driving positions needs to be determined, and the first sub abnormal stay probability is further determined.
Here, the ratio of the abnormal travel position among the plurality of travel positions may be directly regarded as the first sub abnormal stay probability in the extreme case, that is, the ratio of the abnormal travel position among the plurality of travel positions is equal to the first sub abnormal stay probability; the mapping relation between the proportion of the abnormal driving position in the plurality of driving positions and the first sub-abnormal stay probability may be preset, so that the first sub-abnormal stay probability is determined by the proportion of the abnormal driving position in the plurality of driving positions through the mapping relation.
The first sub-abnormal stay probability is preferably determined according to a mapping relation between the proportion of the plurality of driving positions and the first sub-abnormal stay probability, because the first sub-abnormal stay probability, which is represented by the proportion of the determined at least one abnormal driving position in the plurality of driving positions and is more similar and is based on abnormal driving of the target driver, is consistent, and the first sub-abnormal stay probability is determined according to the mapping relation more accurately.
Further, in the method for determining abnormal driving behavior provided in the embodiment of the present application, the determining, based on the driving track movement rule of the target driver and the driving track information of the target driver in the preset driving process under the extreme condition, the second sub-abnormal stay probability that the target driver has abnormal driving includes:
a4: determining an updated mapping relation between a driving position and driving time in a driving track movement rule of a target driver based on a travel scene corresponding to the target travel order; wherein the travel scene includes at least one condition affecting a driving trajectory of the target driver.
b4: and determining that each driving position is an abnormal driving position when the mapping relation between the driving position and the corresponding driving time is inconsistent with the updated mapping relation.
c3: a second abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of the at least one abnormal driving position among the plurality of driving positions.
In step a4, according to the travel scenario corresponding to the determined target travel order, an updated mapping relationship between the travel position and the travel time in the driving track movement rule of the target driver is determined, and considering that the target driver may have interference of different travel time conditions in the process of receiving the driving, the stay time of the target driver at a certain travel position may be excessively long, but the stay cannot be considered as the stay of subjective consciousness of the target driver, so that the stay cannot be determined as the abnormal stay behavior of the target driver, and therefore, the mapping relationship between the travel position and the travel time in the driving track movement rule of the target driver needs to be updated in combination with the real-time travel road condition, so as to improve the accuracy of judging the abnormal stay behavior of the target driver.
Here, the travel scenario corresponding to the target travel order may be determined by the road congestion condition of the current target driver, the travel time of the target travel order, and the travel area where the target travel order is located.
The current road congestion condition of the target driver may cause slower running speed of the target driver at one or more running positions in the driving receiving process, and further cause that the corresponding running time node does not reach the corresponding running position, and when the updated mapping relation is set, the running position where congestion is likely to occur and the running time interval between the adjacent running positions need to be lengthened; under the condition of the travel time of the target travel order, as the road congestion conditions of different travel times have certain difference (the road congestion conditions of the travel roads in the morning and evening peaks are relatively congested in one day, and the road congestion conditions of the travel roads in the peaked period are common), the road congestion conditions can be determined through the different travel times, so that the updated mapping relation between the travel position and the travel time in the driving track movement rule of the target driver is determined, and the determination process is consistent with the determination step under the road congestion conditions, which is not repeated herein; the situation of the travel area represents the specificity of travel time of different areas, and this specificity may affect the established mapping relationship between the travel position and travel time of the target driver, for example, in the european country, the duration of each traffic light is longer than that of the domestic traffic light, so when the mapping relationship is calculated by migration of different areas, the modification is needed, and for the above example, the modification manner is that: it is necessary to lengthen the travel positions corresponding to possible traffic posts and the travel time interval between adjacent travel positions.
In step b4, for each driving position in the driving receiving process of the target driver, if the driving time corresponding to the driving position is inconsistent with the updated mapping relationship, determining that the driving position is an abnormal driving position.
Here, the mapping relationship between the running position and the corresponding running time is inconsistent with the updated mapping relationship, which is reflected in a case where the running time of the running position is inconsistent with the running time corresponding to the running position in the updated mapping relationship and is later than the running time in the updated mapping relationship.
For example, the travel time for the target driver to reach the a travel position during the following driving is indicated as 8 in the updated map: 00, but the travel time for the target driver to reach the a travel position during the following is 9:30, at this time, the traveling position may be determined to be an abnormal traveling position.
In step c3, the ratio of the abnormal travel position among the plurality of travel positions may be directly used as the second sub abnormal stay probability in the extreme case, that is, the ratio of the abnormal travel position among the plurality of travel positions is equal to the second sub abnormal stay probability; it is also possible to set in advance a mapping relationship between the proportion of the abnormal travel position in the plurality of travel positions and the second abnormal stay probability, so that the second sub abnormal stay probability is determined from the proportion of the abnormal travel position in the plurality of travel positions by the mapping relationship.
The method for determining the second sub-abnormal stay probability preferably determines the second abnormal stay probability according to the mapping relation between the proportion of the plurality of driving positions and the second abnormal stay probability, because the second sub-abnormal stay probability which is represented by the proportion of the determined at least one abnormal driving position in the plurality of driving positions and is generated by abnormal driving of the target driver is consistent, and the second sub-abnormal stay probability is determined according to the mapping relation more accurately.
Further, referring to fig. 2, fig. 2 is a flowchart of another method for determining abnormal driving behavior according to an embodiment of the present application, where the determining method further includes:
s201, acquiring a plurality of order attributes corresponding to the history cancelled travel orders; the order attribute comprises at least one of order canceling time between the time when the travel order is canceled and the order receiving time and a driving receiving distance between the order receiving position of the travel order and the boarding position of the passenger.
S202, determining at least one order cancel time abnormal order with the cancel time smaller than the preset time and at least one order target driver driving distance smaller than a preset distance threshold from the acquired plurality of historic cancelled travel orders.
S203, filtering at least one order cancel time and at least one cancel time abnormal order from the plurality of historic cancelled travel orders, and determining a plurality of samples cancelled travel orders.
S204, training the constructed deep learning network based on different historical driving positions in the historical driving track information corresponding to the cancelled travel orders of each sample and the historical driving time corresponding to each historical driving position to obtain an abnormal stay recognition model.
In step S201, while acquiring a plurality of cancelled travel orders, it is also required to determine order attribute information corresponding to each cancelled travel order, and screen orders under extreme conditions existing in the cancelled travel orders according to the order attribute information, so as to avoid the problem that data of the cancelled travel orders under extreme conditions affects training effects in the model training process, and further affects accuracy of model training.
Here, the order attribute information includes at least one of an order cancel time between a time when the travel order is canceled and an order taking time, and an order taking distance between an order taking position of the travel order and an boarding position of the passenger.
In step S202, according to the order attribute information of each of the historic cancelled orders determined in step S201, a cancellation time abnormal order and a cancellation route abnormal order are determined from the acquired plurality of historic cancelled orders, and further, the influence of the sample cancelled travel orders in extreme cases on the model training is removed by canceling the time abnormal order and the cancellation route abnormal order.
In step S203, the at least one cancellation time abnormal order and the at least one cancellation route abnormal order determined in step S202 are filtered from the acquired plurality of historic cancelled travel orders, so as to determine that a plurality of samples for the model in the non-extreme case are cancelled travel orders.
Here, for a travel order cancelled through one history, it may be that the travel order is cancelled at both time and distance, in order to reduce the workload of filtering the abnormal order, a step of removing duplication may be performed before filtering, and for an order identifier (such as an order number) of the travel order cancelled through the same history, orders with different order attributes under the same order identifier are duplicated (that is, the travel order is cancelled at both time and distance).
The process of removing the duplicate may be to only hold one order attribute category of a history cancelled travel order, for example, one history cancelled travel order is a cancelled time abnormal order or a cancelled route abnormal order, and then only hold the order attribute of the history cancelled travel order after removing the duplicate.
In step S204, training the constructed deep learning network according to the plurality of samples determined in step S203 by canceling the travel order, and obtaining an abnormal stay recognition model.
Here, the training process may be the following steps:
acquiring historical driving track information of sample target drivers in a plurality of sample cancelled orders, wherein the actual probability of abnormal stay of each cancelled order corresponding target driver in the driving process;
aiming at each sample cancelled order, the historical driving track information of the sample target driver is input into a constructed deep learning network, and the prediction probability of abnormal stay of the target driver corresponding to the sample cancelled order in the driving process is obtained;
determining, for each sample cancelled order, a deviation value between a predicted probability and an actual probability of the sample cancelled order;
If the deviation value corresponding to the cancelled orders of the samples is larger than the preset deviation threshold, adjusting parameters in the deep learning network until the deviation value corresponding to the cancelled orders of each sample is smaller than or equal to the preset deviation threshold, determining that the deep learning network is trained, and determining the trained deep learning network as the trained recognition model.
Further, in the method for determining abnormal driving behavior provided in the embodiment of the present application, determining, based on a preset gap judgment rule between a current driving position and a preset driving position and driving track information of the target driver, a first abnormal stay probability that the target driver has abnormal stay includes:
a5: and determining the running position as an abnormal running position if the distance difference between the current running position and the preset running position is larger than a preset difference threshold value in the running time corresponding to each running position.
b5: a first abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of the at least one abnormal driving position among the plurality of driving positions.
In step a5, according to the current travel scenario, a plurality of driving positions in the process of receiving the driving of the target driver and the driving time corresponding to each driving position are preset, and according to the difference between the current driving position and the preset driving position, whether the current driving position is an abnormal driving position or not can be determined, and then the first abnormal stay probability that the target driver has abnormal driving is determined according to the determined abnormal driving position.
After the target driver receives the order, a plurality of running positions in the running track of the target driver in the normal running process and the running time corresponding to each running position are planned according to the current road condition and the running speed of the target driver for driving the vehicle, and when the target driver does not reach the corresponding running position at a certain time point, whether the current running position is an abnormal running position or not is considered, and then the determined abnormal running position is passed.
In step b5, a first abnormal stay probability of the target driver having abnormal driving is determined according to the proportion of the at least one abnormal driving position determined in step a5 in the plurality of driving positions. The abnormal driving position characterizes the abnormal driving tendency of the target driver during driving, and the existence of the individual abnormal driving position only represents the possible abnormal situation of the target driver at the position, but the abnormal driving situation may not exist during the whole driving, so that the proportion of the abnormal driving position in a plurality of driving positions needs to be determined, and the first abnormal stay probability is further determined.
Here, the ratio of the abnormal travel position among the plurality of travel positions may be directly regarded as the first abnormal stay probability in the extreme case, that is, the ratio of the abnormal travel position among the plurality of travel positions is equal to the first abnormal stay probability; it is also possible to set in advance a mapping relationship between the proportion of the abnormal travel position in the plurality of travel positions and the first abnormal stay probability so that the first abnormal stay probability is determined from the proportion of the abnormal travel position in the plurality of travel positions by the mapping relationship.
The first abnormal stay probability is preferably determined according to a mapping relation between the proportion of the plurality of driving positions and the first abnormal stay probability, because the first abnormal stay probability which is represented by the proportion of the determined at least one abnormal driving position in the plurality of driving positions and is used for abnormal driving of the target driver is consistent, and the first abnormal stay probability is determined according to the mapping relation more accurately.
Taking a scene that an order is cancelled in a travel process as an example, a determination process of abnormal driving behavior in the technical scheme of the application is described, and the determination of the abnormal driving behavior comprises the following steps:
step 1: generating a target travel order according to the determined travel origin of the passenger and the travel destination input by the passenger, and matching a corresponding receiving driver for the target travel order;
step 2: after the target driver receives the order, monitoring the driving track of the target driver in the process of going to the travel origin where the passenger is located from the order receiving position in real time, and acquiring the driving track information of the target driver in the process of going to the travel origin where the passenger is located from the order receiving position when receiving the order canceling instruction of the passenger;
Step 3: according to a preset bottom-covering strategy: determining whether the calculated first abnormal stay probability of the target driver, which is subjected to abnormal stay under the bottom-covering strategy, is larger than a preset probability threshold value according to a difference judgment rule between the current running position and the preset running position and the running track information of the target driver;
step 4: if the calculated first abnormal stay probability of the target driver, which is subjected to abnormal stay under the bottom-approaching strategy, is larger than a preset probability threshold, it is determined that the target driver hits the bottom-approaching rule in the process of receiving the driver, so that whether the target driver has abnormal driving receiving behaviors or not is judged according to the first abnormal stay probability, and when the first abnormal stay probability is larger than the preset abnormal judgment probability, it is determined that the target driver has abnormal driving receiving behaviors in the process of receiving the driver.
Step 5: if the calculated first abnormal stay probability of the target driver after abnormal stay is not greater than a preset probability threshold under the bottom-covering strategy, the running positions at different times in the acquired running track information and the running time of each running position are input into a pre-trained abnormal stay recognition model, and the second abnormal stay probability of the target driver after abnormal stay recognition model judgment in the driving receiving process is determined according to the output result of the abnormal stay recognition model;
Step 5-1: the driving track information is input into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme asking for money in the order receiving process;
step 5-2: determining whether the target driver encounters an extreme abnormal situation in the order taking process, if the time difference between the time for canceling the order by the passenger and the time for placing the order is short, determining a second sub-abnormal stay probability that the target driver has abnormal driving under the extreme situation according to the driving track information of the target driver and the driving track movement rule of the target driver;
step 5-3: and determining a second abnormal stay probability of abnormal stay of the target driver in the receiving process according to the first sub abnormal stay probability and the second sub abnormal stay probability, and determining that the target driver has abnormal receiving driving behaviors in the receiving process when the second abnormal stay probability is larger than a preset probability threshold.
Step 6: after determining that the target driver has abnormal receiving driving behaviors in the receiving driving process, determining that the target driver is a responsible party in the canceling process of the order, recording the abnormal receiving driving behaviors of the target driver while carrying out the accountability on the target driver, and considering that resources are not inclined to the target driver when the order is distributed subsequently.
According to the method for determining the abnormal driving receiving behavior, which is provided by the embodiment of the application, the cancelled target travel order is acquired and comprises order information of travel track information of a target driver receiving the target travel order; determining a first abnormal stay probability of abnormal stay of the target driver according to a preset difference judgment rule between the current running position and the preset running position, detecting whether the first abnormal stay probability is larger than a preset probability threshold, and determining whether abnormal driving connection behaviors of the target driver occur according to the first abnormal stay probability if the first abnormal stay probability is larger than the preset probability threshold; if the first abnormal stay probability is not greater than the preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine second abnormal stay probability, whether the target driver has abnormal driving connection behaviors or not is determined according to the second abnormal stay probability, and the abnormal driving connection behaviors of the target driver can be comprehensively judged according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model under different conditions, so that the accuracy of recognition of the abnormal driving connection behaviors of the target driver is improved.
Referring to fig. 3 to fig. 4, fig. 3 is a schematic structural diagram of a determining device for abnormal driving behavior provided in an embodiment of the present application, and fig. 4 is a schematic structural diagram of a second determining device for abnormal driving behavior provided in an embodiment of the present application. As shown in fig. 3, the determining apparatus 300 includes:
an order information obtaining module 310, configured to obtain order information of the cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position;
a first probability determining module 320, configured to determine a first abnormal stay probability of the target driver that the target driver has abnormally stayed based on a preset gap judging rule between the current driving position and the preset driving position and the driving track information of the target driver;
the probability detection module 330 is configured to detect whether the first abnormal stay probability is greater than a preset probability threshold;
the first abnormal behavior determining module 340 is configured to determine whether an abnormal driving behavior occurs to the target driver based on the first abnormal stay probability if the first abnormal stay probability is greater than a preset probability threshold;
The second abnormal behavior determining module 350 is configured to input the driving track information into a pre-trained abnormal stay recognition model if the first abnormal stay probability is not greater than a preset probability threshold, so as to determine a second abnormal stay probability of the target driver during the order receiving process, and determine whether the target driver has an abnormal driving behavior based on the second abnormal stay probability.
Further, as shown in fig. 4, the determining apparatus 300 further includes a model training module 360, where the model training module 360 is configured to:
acquiring a plurality of order attribute information corresponding to the historic cancelled travel orders; the order attribute information comprises at least one of order canceling time between the time when the travel order is cancelled and the order receiving time and a driving receiving distance between the order receiving position of the travel order and the boarding position of the passenger;
determining at least one order cancel time abnormal order with the cancel time smaller than the preset time and at least one order target driver driving distance smaller than a preset distance threshold from the acquired plurality of historic cancelled travel orders;
Filtering at least one order cancel time and at least one cancel time abnormal order from the plurality of historical cancelled travel orders, and determining a plurality of sample cancelled travel orders;
and training the constructed deep learning network based on different historical driving positions in the historical driving track information corresponding to the cancelled travel orders of each sample and the historical driving time corresponding to each historical driving position to obtain an abnormal stay recognition model.
Further, the second abnormal behavior determining module 350 is configured to input the driving track information into a pre-trained abnormal stay recognition model, so as to determine a second abnormal stay probability of the target driver during the order receiving process, where the second abnormal behavior determining module 350 is configured to:
inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process; the abnormal stay recognition model is trained based on different historical driving positions in the historical driving track information of the sample target driver in the history cancelled travel order under the non-extreme condition and the historical driving time corresponding to each historical driving position;
Determining a second sub-abnormal stay probability of abnormal driving of the target driver based on a preset driving track movement rule of the target driver in the driving process under the extreme condition and driving track information of the target driver;
and determining a second abnormal stay probability of abnormal stay of the target driver in the order receiving process based on the first abnormal stay probability and the second abnormal stay probability.
Further, the second abnormal behavior determining module 350 is configured to, when determining, based on the driving trajectory movement rule of the target driver and the driving trajectory information of the target driver in the driving process under the preset extreme condition, the second sub-abnormal stay probability that the target driver has abnormal driving, determine that the second abnormal behavior determining module 350 is configured to:
determining at least one abnormal driving position at which driving abnormality occurs at a plurality of driving positions included in the driving locus; the abnormal driving position is a position with the driving speed smaller than the preset driving speed or a position with the position offset planning navigation route;
and determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
Further, when the second abnormal behavior determining module 350 is configured to input the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of the target driver in an abnormal stay under a non-extreme condition in a bill receiving process, the second abnormal behavior determining module 350 is configured to:
determining a plurality of running positions included in the running track information and running time corresponding to each running position;
inputting each driving position and corresponding driving time into a pre-trained abnormal stay recognition model, and determining at least one abnormal driving position;
a first sub-abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of at least one abnormal driving position among the plurality of driving positions.
Further, the second abnormal behavior determining module 350 is configured to, when determining, based on the driving trajectory movement rule of the target driver and the driving trajectory information of the target driver in the driving process under the preset extreme condition, the second sub-abnormal stay probability that the target driver has abnormal driving, determine that the second abnormal behavior determining module 350 is configured to:
Determining an updated mapping relation between a driving position and driving time in a driving track movement rule of a target driver based on a travel scene corresponding to the target travel order; wherein the travel scene includes at least one condition affecting a driving trajectory of a target driver;
for each driving position, if the mapping relation between the driving position and the corresponding driving time is inconsistent with the updated mapping relation, determining that the driving position is an abnormal driving position;
and determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
Further, when the first abnormal behavior determining module 340 is configured to determine, based on a preset gap judgment rule between the current driving position and the preset driving position and the driving track information of the target driver, a first abnormal stay probability that the target driver has abnormal stays, the first abnormal behavior determining module 340 is configured to:
determining that the driving position is an abnormal driving position if the distance difference between the current driving position and the preset driving position is larger than a preset difference threshold value in the driving time corresponding to each driving position;
A first abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of the at least one abnormal driving position among the plurality of driving positions.
The determining device for abnormal driving receiving behavior obtains order information of cancelled target travel orders including travel track information of a target driver receiving the target travel orders; determining a first abnormal stay probability of abnormal stay of the target driver according to a preset difference judgment rule between the current running position and the preset running position, detecting whether the first abnormal stay probability is larger than a preset probability threshold, and determining whether abnormal driving connection behaviors of the target driver occur according to the first abnormal stay probability if the first abnormal stay probability is larger than the preset probability threshold; if the first abnormal stay probability is not greater than the preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine second abnormal stay probability, whether the target driver has abnormal driving connection behaviors or not is determined according to the second abnormal stay probability, and the abnormal driving connection behaviors of the target driver can be comprehensively judged according to the difference judgment rule between the current driving position and the preset driving position and the abnormal stay recognition model under different conditions, so that the accuracy of recognition of the abnormal driving connection behaviors of the target driver is improved.
Referring to fig. 5, fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application. As shown in fig. 5, the electronic device 500 includes a processor 510, a memory 520, and a bus 530.
The memory 520 stores machine-readable instructions executable by the processor 510, when the electronic device 500 is running, the processor 510 communicates with the memory 520 through the bus 530, and when the machine-readable instructions are executed by the processor 510, the steps of the method for determining abnormal driving behavior in the method embodiments shown in fig. 1 and fig. 2 may be executed, and detailed implementation manners may refer to the method embodiments and are not repeated herein.
The embodiment of the present application further provides a computer readable storage medium, where a computer program is stored on the computer readable storage medium, and when the computer program is executed by a processor, the steps of the method for determining abnormal driving behavior in the method embodiments shown in fig. 1 and fig. 2 may be executed, and specific implementation manners may refer to the method embodiments and are not repeated herein.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, and are not repeated herein.
In the several embodiments provided in this application, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. The above-described apparatus embodiments are merely illustrative, for example, the division of the units is merely a logical function division, and there may be other manners of division in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some communication interface, device or unit indirect coupling or communication connection, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a non-volatile computer readable storage medium executable by a processor. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Finally, it should be noted that: the foregoing examples are merely specific embodiments of the present application, and are not intended to limit the scope of the present application, but the present application is not limited thereto, and those skilled in the art will appreciate that while the foregoing examples are described in detail, the present application is not limited thereto. Any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or make equivalent substitutions for some of the technical features within the technical scope of the disclosure of the present application; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present application, and are intended to be included in the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (11)
1. The method for determining the abnormal driving behavior is characterized by comprising the following steps:
acquiring order information of the cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position;
determining a first abnormal stay probability of the target driver when the target driver stops abnormally based on a preset difference judgment rule between the current running position and the preset running position and the running track information of the target driver;
detecting whether the first abnormal stay probability is larger than a preset probability threshold;
if the first abnormal stay probability is larger than a preset probability threshold, determining whether an abnormal driving behavior occurs to the target driver or not based on the first abnormal stay probability;
if the first abnormal stay probability is not greater than a preset probability threshold, the driving track information is input into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of abnormal stay of the target driver in the order receiving process, and whether the target driver has abnormal driving receiving behaviors is determined based on the second abnormal stay probability;
The step of inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver during the order receiving process, wherein the step of determining the second abnormal stay probability comprises the following steps: inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process; the abnormal stay recognition model is trained based on different historical driving positions in the historical driving track information of the sample target driver in the history cancelled travel order under the non-extreme condition and the historical driving time corresponding to each historical driving position; determining a second sub-abnormal stay probability of abnormal driving of the target driver based on a preset driving track movement rule of the target driver in the driving process under the extreme condition and driving track information of the target driver; and determining a second abnormal stay probability of abnormal stay of the target driver in the order receiving process based on the first abnormal stay probability and the second abnormal stay probability.
2. The determination method according to claim 1, characterized in that the extreme case is a case where at least one of the following conditions exists:
The time difference between the time when the travel order is cancelled and the order taking time is smaller than a preset time threshold;
the distance between the order taking position of the travel order and the boarding position of the passenger is smaller than a preset distance threshold value.
3. The determining method according to claim 1, wherein the determining the second sub-abnormal stay probability that the target driver has abnormal driving based on the driving locus movement rule of the target driver and the driving locus information of the target driver during the driving in the extreme case set in advance includes:
determining at least one abnormal driving position at which driving abnormality occurs at a plurality of driving positions included in the driving locus; the abnormal driving position is a position with the driving speed smaller than the preset driving speed or a position with the position offset planning navigation route;
and determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
4. The method according to claim 1, wherein the step of inputting the travel track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of the target driver for abnormal stay in a non-extreme case in the order taking process includes:
Determining a plurality of running positions included in the running track information and running time corresponding to each running position;
inputting each driving position and corresponding driving time into a pre-trained abnormal stay recognition model, and determining at least one abnormal driving position;
a first sub-abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of at least one abnormal driving position among the plurality of driving positions.
5. The determination method according to claim 1, wherein the second sub-abnormal stay probability that the target driver has abnormal driving is determined based on a driving locus movement rule of the target driver and driving locus information of the target driver during driving under a preset extreme condition:
determining an updated mapping relation between a driving position and driving time in a driving track movement rule of a target driver based on a travel scene corresponding to the target travel order; wherein the travel scene includes at least one condition affecting a driving trajectory of a target driver;
for each driving position, if the mapping relation between the driving position and the corresponding driving time is inconsistent with the updated mapping relation, determining that the driving position is an abnormal driving position;
And determining a second sub-abnormal stay probability of the abnormal driving of the target driver based on the determined proportion of the at least one abnormal driving position in the plurality of driving positions.
6. The determination method according to claim 1, wherein the cancellation party of the target travel order is a passenger;
the travel track information of the target driver is travel track information of the target driver in a period from a time when the target travels the order to a time when the order is canceled.
7. The determination method according to claim 1, characterized in that the determination method further comprises:
acquiring a plurality of order attribute information corresponding to the historic cancelled travel orders; the order attribute information comprises at least one of order canceling time between the time when the travel order is cancelled and the order receiving time and a driving receiving distance between the order receiving position of the travel order and the boarding position of the passenger;
determining at least one order cancel time abnormal order with the cancel time smaller than the preset time and at least one order target driver driving distance smaller than a preset distance threshold from the acquired plurality of historic cancelled travel orders;
Filtering at least one order cancel time and at least one cancel time abnormal order from the plurality of historical cancelled travel orders, and determining a plurality of sample cancelled travel orders;
and training the constructed deep learning network based on different historical driving positions in the historical driving track information corresponding to the cancelled travel orders of each sample and the historical driving time corresponding to each historical driving position to obtain an abnormal stay recognition model.
8. The determination method according to claim 1, wherein the determining the first abnormal stay probability that the target driver has abnormal stay based on the preset gap judgment rule between the current running position and the preset running position and the running track information of the target driver includes:
determining that the driving position is an abnormal driving position if the distance difference between the current driving position and the preset driving position is larger than a preset difference threshold value in the driving time corresponding to each driving position;
a first abnormal stay probability that the target driver has abnormal driving is determined based on the determined proportion of the at least one abnormal driving position among the plurality of driving positions.
9. A determination device for abnormal driving behavior, wherein the determination device comprises:
the order information acquisition module is used for acquiring order information of the cancelled target travel order; the order information comprises driving track information of a target driver for receiving a target trip order; the driving track information comprises driving positions of a target driver at different times after receiving a target travel order and driving time corresponding to each driving position;
the first probability determining module is used for determining a first abnormal stay probability of the target driver when the target driver stops abnormally based on a preset difference judging rule between the current running position and the preset running position and the running track information of the target driver;
the probability detection module is used for detecting whether the first abnormal stay probability is larger than a preset probability threshold value or not;
the first abnormal behavior determining module is used for determining whether the target driver generates abnormal driving connection behaviors or not based on the first abnormal stay probability if the first abnormal stay probability is larger than a preset probability threshold;
the second abnormal behavior determining module is used for inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver in abnormal stay in the order receiving process if the first abnormal stay probability is not greater than a preset probability threshold value, and determining whether the target driver has abnormal driving receiving behaviors or not based on the second abnormal stay probability;
The step of inputting the driving track information into a pre-trained abnormal stay recognition model to determine a second abnormal stay probability of the target driver during the order receiving process, wherein the step of determining the second abnormal stay probability comprises the following steps: inputting the driving track information into a pre-trained abnormal stay recognition model to determine a first sub-abnormal stay probability of abnormal stay of the target driver under a non-extreme condition in a bill receiving process; the abnormal stay recognition model is trained based on different historical driving positions in the historical driving track information of the sample target driver in the history cancelled travel order under the non-extreme condition and the historical driving time corresponding to each historical driving position; determining a second sub-abnormal stay probability of abnormal driving of the target driver based on a preset driving track movement rule of the target driver in the driving process under the extreme condition and driving track information of the target driver; and determining a second abnormal stay probability of abnormal stay of the target driver in the order receiving process based on the first abnormal stay probability and the second abnormal stay probability.
10. An electronic device, comprising: a processor, a storage medium and a bus, the storage medium storing machine-readable instructions executable by the processor, the processor and the storage medium communicating over the bus when the electronic device is running, the processor executing the machine-readable instructions to perform the steps of the method of determining abnormal driving behavior according to any one of claims 1 to 8.
11. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when executed by a processor, performs the steps of the method of determining abnormal driving behavior according to any one of claims 1 to 8.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011615630.1A CN112650825B (en) | 2020-12-30 | 2020-12-30 | Determination method and device for abnormal driving behavior, storage medium and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011615630.1A CN112650825B (en) | 2020-12-30 | 2020-12-30 | Determination method and device for abnormal driving behavior, storage medium and electronic equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112650825A CN112650825A (en) | 2021-04-13 |
CN112650825B true CN112650825B (en) | 2024-04-05 |
Family
ID=75364389
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011615630.1A Active CN112650825B (en) | 2020-12-30 | 2020-12-30 | Determination method and device for abnormal driving behavior, storage medium and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112650825B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113447037B (en) * | 2021-06-04 | 2023-01-20 | 上海钧正网络科技有限公司 | Stroke yaw detection method and device |
CN113704643B (en) * | 2021-09-03 | 2022-10-18 | 北京百度网讯科技有限公司 | Method and device for determining state of target object, electronic equipment and storage medium |
CN115081545B (en) * | 2022-07-22 | 2022-11-25 | 天津所托瑞安汽车科技有限公司 | Driver rotation identification method and identification model construction method |
CN115631482B (en) * | 2022-11-30 | 2023-04-04 | 广汽埃安新能源汽车股份有限公司 | Driving perception information acquisition method and device, electronic equipment and readable medium |
CN116543770B (en) * | 2023-07-05 | 2023-09-22 | 北京龙驹易行科技有限公司 | Method, device, equipment and storage medium for detecting span conflict |
CN117635403B (en) * | 2023-11-08 | 2024-06-21 | 杭州一喂智能科技有限公司 | Abnormal order alarm method, device, electronic equipment and computer readable medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR20190098092A (en) * | 2019-07-31 | 2019-08-21 | 엘지전자 주식회사 | Management method of hacking vehicle in automatic driving system and the apparatus for the method |
WO2019242259A1 (en) * | 2018-06-21 | 2019-12-26 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for determining potential malicious event |
CN111325437A (en) * | 2019-02-18 | 2020-06-23 | 北京嘀嘀无限科技发展有限公司 | Abnormal driving behavior recognition method and device and electronic equipment |
CN111814109A (en) * | 2020-04-08 | 2020-10-23 | 北京嘀嘀无限科技发展有限公司 | Vehicle track deviation detection method and device, storage medium and electronic equipment |
CN111832764A (en) * | 2019-04-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
-
2020
- 2020-12-30 CN CN202011615630.1A patent/CN112650825B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019242259A1 (en) * | 2018-06-21 | 2019-12-26 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for determining potential malicious event |
CN111325437A (en) * | 2019-02-18 | 2020-06-23 | 北京嘀嘀无限科技发展有限公司 | Abnormal driving behavior recognition method and device and electronic equipment |
CN111832764A (en) * | 2019-04-15 | 2020-10-27 | 北京嘀嘀无限科技发展有限公司 | Data processing method and device, electronic equipment and readable storage medium |
KR20190098092A (en) * | 2019-07-31 | 2019-08-21 | 엘지전자 주식회사 | Management method of hacking vehicle in automatic driving system and the apparatus for the method |
CN111814109A (en) * | 2020-04-08 | 2020-10-23 | 北京嘀嘀无限科技发展有限公司 | Vehicle track deviation detection method and device, storage medium and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN112650825A (en) | 2021-04-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112650825B (en) | Determination method and device for abnormal driving behavior, storage medium and electronic equipment | |
Gu et al. | Utilizing UAV video data for in-depth analysis of drivers’ crash risk at interchange merging areas | |
US11867517B2 (en) | Navigating to a moving target | |
US9508257B2 (en) | Road detection logic | |
US8731808B2 (en) | Road network analysis system | |
CN108444486B (en) | Navigation route sorting method and device | |
JP6247754B2 (en) | How to process vehicle measurement data to identify the start of parking hunting | |
JP2017228266A (en) | Local trajectory planning method and apparatus used for smart vehicles | |
CN112435469B (en) | Vehicle early warning control method and device, computer readable medium and electronic equipment | |
CN110363984B (en) | Traffic flow prediction method and apparatus | |
US20140365103A1 (en) | Method and system for collecting traffic data | |
US10132641B2 (en) | Automated crowd sourcing of a navigation route | |
CN112669594A (en) | Method, device, equipment and storage medium for predicting traffic road conditions | |
CN115083171B (en) | Unmanned aerial vehicle fixed-point inspection method and system based on muck vehicle monitoring | |
JP2001084479A (en) | Method and device for forecasting traffic flow data | |
CN109493606A (en) | The recognition methods and system of parking are disobeyed on a kind of highway | |
CN109489679A (en) | A kind of arrival time calculation method in guidance path | |
KR100865939B1 (en) | Traffic information guiding system and method | |
JP2003296891A (en) | Travel time estimation device, method therefor and computer program | |
CN111489460B (en) | Travel data processing method, travel data processing device, navigation equipment and computer storage medium | |
Elefteriadou et al. | Comparison of methods for measuring travel time at Florida freeways and arterials. | |
JP4957612B2 (en) | Travel pattern information acquisition device, travel pattern information acquisition method, and travel pattern information acquisition program | |
WO2015125467A1 (en) | Travel time computation device, traffic information data, and traffic information data generation device | |
Richly et al. | Predicting location probabilities of drivers to improved dispatch decisions of transportation network companies based on trajectory data | |
Hayashi et al. | Prioritization of Lane-based Traffic Jam Detection for Automotive Navigation System utilizing Suddenness Index Calculation Method for Aggregated Values |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |