CN111325437B

CN111325437B - Abnormal driving behavior recognition method and device and electronic equipment

Info

Publication number: CN111325437B
Application number: CN201910120692.6A
Authority: CN
Inventors: 李海波
Original assignee: Beijing Didi Infinity Technology and Development Co Ltd
Current assignee: Beijing Didi Infinity Technology and Development Co Ltd
Priority date: 2019-02-18
Filing date: 2019-02-18
Publication date: 2021-09-24
Anticipated expiration: 2039-02-18
Also published as: CN111325437A

Abstract

The application provides a method and a device for identifying abnormal driving behaviors and electronic equipment, wherein the method comprises the following steps: acquiring driving data in the order execution process; the driving data comprises driving track data and driving speed data; calculating the probability of abnormal driving behaviors according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability; and identifying whether the abnormal driving behavior exists in the order execution process based on the abnormal driving behavior probability. The method and the device can effectively identify whether the driver has abnormal driving behaviors or not in time in the order execution process of the driver.

Description

Abnormal driving behavior recognition method and device and electronic equipment

Technical Field

The application relates to the technical field of internet, in particular to a method and a device for identifying abnormal driving behaviors and electronic equipment.

Background

With the popularization of the designated driving platform, more and more users select the online car appointment for travel. However, during the process of providing the designated driving service for the passenger, the driver may conflict with the passenger. One of the main causes of the conflict is the abnormality of the driving behavior of the driver. Factors such as the driver's own behavior not meeting the platform regulations (e.g., private parking during passenger loading), poor driver driving habits (e.g., habitual rapid acceleration/deceleration, or speeding), etc., can cause the driver to behave abnormally during driving to service the passengers.

When the existing online booking platform carries out driving supervision on a driver, the driver is comprehensively evaluated mainly based on indexes such as the grade of the driver after a passenger finishes an order, the cancellation rate of the driver order and the like. However, this method is only a post analysis, and it is still difficult to timely monitor whether the driver has abnormal driving behavior during the driver's order execution process.

Disclosure of Invention

In view of the above, an object of the embodiments of the present application is to provide a method and an apparatus for identifying an abnormal driving behavior, and an electronic device, which can effectively identify whether an abnormal driving behavior occurs to a driver in time during an order execution process of the driver.

According to an aspect of the present application, there is provided a method of identifying abnormal driving behavior, the method including: acquiring driving data in the order execution process; the driving data comprises driving track data and driving speed data; calculating the probability of abnormal driving behaviors according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability; and identifying whether abnormal driving behaviors exist in the order execution process or not based on the abnormal driving behavior probability.

In some embodiments, the step of obtaining travel data during order fulfillment comprises: acquiring driving track data and driving speed data recorded by a driver terminal in an order execution process, wherein the driving track data comprises position data recorded by a GPS sensor; the travel speed data includes: acceleration data, angular velocity data and velocity data recorded by the velocity sensor.

In some embodiments, the step of calculating the probability of abnormal driving behavior from the driving data comprises: determining a parking position point in the driving process according to the driving track data; for each parking position point, acquiring road condition information corresponding to the parking position point and a corresponding POI (point of interest) identifier; judging whether the parking behavior corresponding to the parking position point belongs to abnormal parking or not according to the road condition information and the POI (point of interest) identification; and calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking.

In some embodiments, the step of determining whether the parking behavior corresponding to the parking position point belongs to abnormal parking according to the road condition information and the POI identifier includes: and if the road condition information is that the road condition is smooth and smooth, and the POI identification does not contain the preset identification, determining that the parking behavior corresponding to the parking position point belongs to abnormal parking.

In some embodiments, the step of calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking includes: calculating the sum of the parking time lengths of all the parking position points belonging to abnormal parking to obtain the total parking time length; and determining the abnormal parking probability according to the ratio of the total parking time to the driving time corresponding to the driving track data.

In some embodiments, the step of calculating the probability of abnormal driving behavior from the driving data comprises: checking whether data larger than a speed change threshold exists in the acceleration data of the running speed data; counting the number of data larger than the speed change threshold according to the checking result; and determining the abnormal speed change probability according to the ratio of the data number to the total data number of the acceleration data.

In some embodiments, the step of calculating the probability of abnormal driving behavior from the driving data comprises: checking whether data larger than an angular speed change threshold exists in the angular speed data of the running speed data; counting the number of data larger than the angular speed change threshold according to the checking result; and determining abnormal turning probability according to the ratio of the data number to the total data number of the angular velocity data.

In some embodiments, the step of calculating the probability of abnormal driving behavior from the driving data comprises: checking whether data of a preset speed limit larger than a corresponding position of the speed data exists in the speed data of the running speed data or not; counting the overspeed duration corresponding to the speed data larger than the preset speed limit; and determining the overspeed probability according to the ratio of the overspeed duration to the driving duration corresponding to the driving track data.

In some embodiments, the step of identifying whether there is an abnormal driving behavior in the order execution process based on the abnormal driving behavior probability includes: inputting the driving data and the abnormal driving behavior probability into an abnormal driving behavior recognition model obtained by pre-training, and recognizing whether abnormal driving behaviors occur or not through the abnormal driving behavior recognition model; wherein, the abnormal driving behavior recognition model is a neural network model.

In some embodiments, the abnormal driving behavior recognition model comprises an LSTM network and a fully connected network; wherein the input end of the full connection network is connected with the output end of the last hidden layer of the LSTM network.

In some embodiments, the step of inputting the driving data and the abnormal driving behavior probability to an abnormal driving behavior recognition model trained in advance includes: inputting the driving data into the LSTM network, and inputting the abnormal driving behavior probability into the fully-connected network.

In some embodiments, the step of identifying whether an abnormal driving behavior occurs by the abnormal driving behavior recognition model includes: generating driving state characteristics corresponding to the driving track data through the LSTM network, and inputting the driving state characteristics to the full-connection network; and performing classification regression on the driving state characteristics and the abnormal driving behavior probability through the full-connection network to generate an abnormal driving behavior identification result corresponding to the order execution process.

In some embodiments, the training process of the abnormal driving behavior recognition model includes: acquiring a sample data set; the sample data set comprises positive sample data and negative sample data, and the positive sample data and the negative sample data both carry tags; establishing a basic model structure; the basic model structure comprises an LSTM network and a fully connected network which are connected in sequence; and training the network parameters of the basic model structure through the sample data set to obtain an abnormal driving behavior recognition model.

In some embodiments, the step of obtaining a sample data set comprises: screening the marked behavior tags from the order database by adopting a machine screening mode, and acquiring driving data corresponding to the behavior tags; wherein the behavior tags include positive behavior tags and negative behavior tags; marking the behavior label on the corresponding driving data; taking the driving data marked with the positive behavior label as positive sample data; and taking the driving data marked with the negative behavior label as negative sample data.

In some embodiments, the step of obtaining the sample data set further comprises: acquiring the driving data marked with the complaint label, and taking the driving data marked with the complaint label as negative sample data; wherein the complaint labels are manually labeled.

In some embodiments, the ratio of the negative sample data to the positive sample data is not less than 1: 5.

In some embodiments, the method further comprises: if the abnormal driving behavior is identified in the order execution process, initiating an abnormal reminding notification in a preset mode; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification.

According to another aspect of the present application, there is also provided an abnormal driving behavior recognition apparatus, characterized in that the apparatus includes: the driving data acquisition module is used for acquiring driving data in the order execution process; the driving data comprises driving track data and driving speed data; the abnormal probability calculation module is used for calculating the abnormal driving behavior probability according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and abnormal overspeed probability; and the abnormal behavior identification module is used for identifying whether the abnormal driving behavior exists in the order execution process based on the abnormal driving behavior probability.

In some embodiments, the travel data acquisition module is to: acquiring driving track data and driving speed data recorded by a driver terminal in an order execution process, wherein the driving track data comprises position data recorded by a GPS sensor; the travel speed data includes: acceleration data, angular velocity data and velocity data recorded by the velocity sensor.

In some embodiments, the anomaly probability calculation module is to: determining a parking position point in the driving process according to the driving track data; for each parking position point, acquiring road condition information corresponding to the parking position point and a corresponding POI (point of interest) identifier; judging whether the parking behavior corresponding to the parking position point belongs to abnormal parking or not according to the road condition information and the POI (point of interest) identification; and calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking.

In some embodiments, the anomaly probability calculation module is further configured to: and if the road condition information is that the road condition is smooth and the POI identification does not contain the preset identification, determining that the parking behavior corresponding to the parking position point does not belong to abnormal parking.

In some embodiments, the anomaly probability calculation module is further configured to: calculating the sum of the parking time lengths of all the parking position points belonging to abnormal parking to obtain the total parking time length; and determining the abnormal parking probability according to the ratio of the total parking time to the driving time corresponding to the driving track data.

In some embodiments, the anomaly probability calculation module is to: checking whether data larger than a speed change threshold exists in the acceleration data of the running speed data; counting the number of data larger than the speed change threshold according to the checking result; and determining the abnormal speed change probability according to the ratio of the data number to the total data number of the acceleration data.

In some embodiments, the anomaly probability calculation module is to: checking whether data larger than an angular speed change threshold exists in the angular speed data of the running speed data; counting the number of data larger than the angular speed change threshold according to the checking result; and determining abnormal turning probability according to the ratio of the data number to the total data number of the angular velocity data.

In some embodiments, the anomaly probability calculation module is to: checking whether data of a preset speed limit larger than a corresponding position of the speed data exists in the speed data of the running speed data or not; counting the overspeed duration corresponding to the speed data larger than the preset speed limit; and determining the overspeed probability according to the ratio of the overspeed duration to the driving duration corresponding to the driving track data.

In some embodiments, the abnormal behavior identification module is to: inputting the driving data and the abnormal driving behavior probability into an abnormal driving behavior recognition model obtained by pre-training, and recognizing whether abnormal driving behaviors occur or not through the abnormal driving behavior recognition model; wherein, the abnormal driving behavior recognition model is a neural network model.

In some embodiments, the abnormal behavior identification module is further to: inputting the driving data into the LSTM network, and inputting the abnormal driving behavior probability into the fully-connected network.

In some embodiments, the abnormal behavior identification module is further to: generating driving characteristics corresponding to the driving track data through the LSTM network, and inputting the driving characteristics into the full-connection network; and carrying out classification regression on the driving characteristics and the abnormal driving behavior probability through the full-connection network to generate an abnormal driving behavior identification result corresponding to the order execution process.

In some embodiments, the apparatus further comprises a training module of the abnormal driving behavior recognition model, the training module being configured to: acquiring a sample data set; the sample data set comprises positive sample data and negative sample data, and the positive sample data and the negative sample data both carry tags; establishing a basic model structure; the basic model structure comprises an LSTM network and a fully connected network which are connected in sequence; and training the network parameters of the basic model structure through the sample data set to obtain an abnormal driving behavior recognition model.

In some embodiments, the training module is further to: screening the marked behavior tags from the order database by adopting a machine screening mode, and acquiring driving data corresponding to the behavior tags; wherein the behavior tags include positive behavior tags and negative behavior tags; marking the behavior label on the corresponding driving data; taking the driving data marked with the positive behavior label as positive sample data; and taking the driving data marked with the negative behavior label as negative sample data.

In some embodiments, the training module is further to: acquiring the driving data marked with the complaint label, and taking the driving data marked with the complaint label as negative sample data; wherein the complaint labels are manually labeled.

In some embodiments, the apparatus further comprises: the reminding module is used for initiating an abnormal reminding notification in a preset mode if the abnormal driving behavior is identified in the order execution process; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification.

According to another aspect of the present application, there is also provided an electronic device including: the system comprises a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, when an electronic device runs, the processor and the storage medium are communicated through the bus, and the processor executes the machine-readable instructions to execute the steps of the method for identifying the abnormal driving behavior.

According to another aspect of the present application, there is also provided a computer-readable storage medium having stored thereon a computer program which, when being executed by a processor, performs the steps of the method for identifying abnormal driving behavior as set forth in any one of the above.

In the embodiment of the application, the driving data can be acquired in the order execution process, then the abnormal driving behavior probability is calculated according to the driving data, and whether the abnormal driving behavior exists in the order execution process is identified based on the abnormal driving behavior probability. The mode can effectively identify whether the driver has abnormal driving behavior in time according to the driving data of the driver in the order execution process, thereby achieving better driver supervision effect.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 is a block diagram illustrating an identification system for abnormal driving behavior provided by an embodiment of the present application;

FIG. 2 shows a schematic diagram of an electronic device provided by an embodiment of the application;

fig. 3 is a flowchart illustrating a method for identifying abnormal driving behavior according to an embodiment of the present application;

fig. 4 is a schematic structural diagram illustrating an abnormal driving behavior recognition model provided in an embodiment of the present application;

fig. 5 is a schematic structural diagram illustrating a specific structure of an abnormal driving behavior recognition model provided in an embodiment of the present application;

fig. 6 is a block diagram illustrating a structure of an abnormal driving behavior recognition apparatus according to an embodiment of the present application;

fig. 7 is a block diagram illustrating another abnormal driving behavior recognition apparatus according to an embodiment of the present application.

Detailed Description

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, 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 should be understood that the drawings in the present application are for illustrative and descriptive purposes only and are not used to limit the scope of protection of the present application. Additionally, it should be understood that the schematic drawings are not necessarily drawn to scale. The flowcharts used in this application illustrate operations implemented according to some embodiments of the present application. It should be understood that the operations of the flow diagrams may be performed out of order, and steps without logical context may be performed in reverse order or simultaneously. One skilled in the art, under the guidance of this application, may add one or more other operations to, or remove one or more operations from, the flowchart.

In addition, the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application.

To enable those skilled in the art to utilize the present disclosure, the following embodiments are presented in conjunction with a specific application scenario, "a network appointment scenario". It will be apparent to those skilled in the art that the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the application. Although the present application is described primarily in the context of a network appointment scenario, it should be understood that this is only one exemplary embodiment. The present application may be applied to the following vehicles: may include a taxi, a private car, a tailgating, a bus, an unmanned vehicle, or the like, or any combination thereof.

It should be noted that in the embodiments of the present application, the term "comprising" is used to indicate the presence of the features stated hereinafter, but does not exclude the addition of further features.

The terms "passenger," "requestor," "service person," "service requestor," and "customer" are used interchangeably in this application to refer to an individual, entity, or tool that can request or order a service. The terms "driver," "provider," "service provider," and "provider" are used interchangeably in this application to refer to an individual, entity, or tool that can provide a service. The term "user" in this application may refer to an individual, entity or tool that requests a service, subscribes to a service, provides a service, or facilitates the provision of a service. For example, the user may be a passenger, a driver, an operator, etc., or any combination thereof. In the present application, "passenger" and "passenger terminal" may be used interchangeably, and "driver" and "driver terminal" may be used interchangeably.

The term "order" in this application refers to a request initiated by a passenger, a service requester, a driver, a service provider, or any combination thereof. Accepting the "order" may be a passenger, a service requester, a driver, a service provider, or any combination thereof. The service request may be charged or free.

One aspect of the present application relates to an abnormal driving behavior recognition system, which is capable of acquiring driving data during order execution, calculating a probability of an abnormal driving behavior according to the driving data, and recognizing whether an abnormal driving behavior exists during the order execution based on the probability of the abnormal driving behavior. The mode can effectively identify whether the driver has abnormal driving behaviors according to the driving data of the driver in the order execution process, thereby achieving better driver supervision effect.

It is worth noting that, before the application is proposed, the existing network car booking platform only adopts a post-affair mode to supervise the driver, for example, after the driver order is executed, the driver driving behavior can be roughly known through passenger evaluation, and timely recognition and intervention can not be achieved when abnormal driving behavior occurs in the current order execution process of the driver. However, the abnormal driving behavior identification method, the abnormal driving behavior identification device and the electronic equipment can identify the abnormal driving behavior of the driver in the order execution process, so that the network appointment platform can effectively monitor the driving behavior of the driver in time.

Fig. 1 is a block diagram of an abnormal driving behavior recognition system 100 according to some embodiments of the present application. For example, the identification system 100 of abnormal driving behavior may be an online transportation service platform for transportation services such as taxi, designated driving service, express, carpool, bus service, driver rental, or regular service, or any combination thereof. The system 100 for identifying abnormal driving behavior may include one or more of a server 110, a network 120, a service requester terminal 130, a service provider terminal 140, and a database 150, and the server 110 may include a processor 112 for performing an instruction operation therein.

In some embodiments, the server 110 may be a single server or a group of servers. The set of servers can be centralized or distributed (e.g., the servers 110 can be a distributed system). In some embodiments, the server 110 may be local or remote to the terminal. For example, the server 110 may access information and/or data stored in the service requester terminal 130, the service provider terminal 140, or the database 150, or any combination thereof, via the network 120. As another example, the server 110 may be directly connected to at least one of the service requester terminal 130, the service provider terminal 140, and the database 150 to access stored information and/or data. In some embodiments, the server 110 may be implemented on a cloud platform; by way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud (community cloud), a distributed cloud, an inter-cloud, a multi-cloud, and the like, or any combination thereof. In some embodiments, the server 110 may be implemented on an electronic device 200 having one or more of the components shown in FIG. 2 in the present application.

In some embodiments, the server 110 may include a processor 112. Processor 112 may process information and/or data related to the service request to perform one or more of the functions described herein. For example, the processor 112 may determine the target vehicle based on a service request obtained from the service requester terminal 130. In some embodiments, the processor 112 may include one or more processing cores (e.g., a single-core processor (S) or a multi-core processor (S)). Merely by way of example, the Processor 112 may include a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), an Application Specific Instruction Set Processor (ASIP), a Graphics Processing Unit (GPU), a Physical Processing Unit (PPU), a Digital Signal Processor (DSP), a Field Programmable Gate Array (FPGA), a Programmable Logic Device (PLD), a controller, a microcontroller Unit, a Reduced Instruction Set computer (Reduced Instruction Set computer), a microprocessor, or the like, or any combination thereof.

Network 120 may be used for the exchange of information and/or data. In some embodiments, one or more components (e.g., server 110, service requestor terminal 130, service provider terminal 140, and database 150) in system 100 may send information and/or data to other components. For example, the server 110 may obtain a service request from the service requester terminal 130 via the network 120. In some embodiments, the network 120 may be any type of wired or wireless network, or combination thereof. Merely by way of example, Network 130 may include a wired Network, a Wireless Network, a fiber optic Network, a telecommunications Network, an intranet, the internet, a Local Area Network (LAN), a Wide Area Network (WAN), a Wireless Local Area Network (WLAN), a WLAN, a Metropolitan Area Network (MAN), a Wide Area Network (WAN), a Public Switched Telephone Network (PSTN), a bluetooth Network, a ZigBee Network, a Near Field Communication (NFC) Network, or the like, or any combination thereof. In some embodiments, network 120 may include one or more network access points. For example, network 120 may include wired or wireless network access points, such as base stations and/or network switching nodes, through which one or more components of system 100 may connect to network 120 to exchange data and/or information.

In some embodiments, the user of the service requestor terminal 130 may be someone other than the actual demander of the service. For example, the user a of the service requester terminal 130 may use the service requester terminal 130 to initiate a service request for the service actual demander B (for example, the user a may call a car for his friend B), or receive service information or instructions from the server 110. In some embodiments, the user of the service provider terminal 140 may be the actual provider of the service or may be another person than the actual provider of the service. For example, user C of the service provider terminal 140 may use the service provider terminal 140 to receive a service request serviced by the service provider entity D (e.g., user C may pick up an order for driver D employed by user C), and/or information or instructions from the server 110. In some embodiments, "service requester" and "service requester terminal" may be used interchangeably, and "service provider" and "service provider terminal" may be used interchangeably.

In some embodiments, the service requester terminal 130 may comprise a mobile device, a tablet computer, a laptop computer, or a built-in device in a motor vehicle, etc., or any combination thereof. In some embodiments, the mobile device may include a smart home device, a wearable device, a smart mobile device, a virtual reality device, an augmented reality device, or the like, or any combination thereof.

In some embodiments, the service requester terminal 130 may be a device having a location technology for locating the location of the service requester and/or service requester terminal. The service provider terminal 140 may be a similar or identical device as the service requester terminal 130. In some embodiments, the service provider terminal 140 may be a device with location technology for locating the location of the service provider and/or the service provider terminal. In some embodiments, the service requester terminal 130 and/or the service provider terminal 140 may communicate with other locating devices to determine the location of the service requester, service requester terminal 130, service provider, or service provider terminal 140, or any combination thereof. In some embodiments, the service requester terminal 130 and/or the service provider terminal 140 may transmit the location information to the server 110.

Database 150 may store data and/or instructions. In some embodiments, the database 150 may store data obtained from the service requester terminal 130 and/or the service provider terminal 140. In some embodiments, database 150 may store data and/or instructions for the exemplary methods described herein. In some embodiments, database 150 may include mass storage, removable storage, volatile Read-write Memory, or Read-Only Memory (ROM), among others, or any combination thereof. By way of example, mass storage may include magnetic disks, optical disks, solid state drives, and the like; removable memory may include flash drives, floppy disks, optical disks, memory cards, zip disks, tapes, and the like; volatile read-write Memory may include Random Access Memory (RAM); the RAM may include Dynamic RAM (DRAM), Double data Rate Synchronous Dynamic RAM (DDR SDRAM); static RAM (SRAM), Thyristor-Based Random Access Memory (T-RAM), Zero-capacitor RAM (Zero-RAM), and the like. By way of example, ROMs may include Mask Read-Only memories (MROMs), Programmable ROMs (PROMs), Erasable Programmable ROMs (PERROMs), Electrically Erasable Programmable ROMs (EEPROMs), compact disk ROMs (CD-ROMs), digital versatile disks (ROMs), and the like. In some embodiments, database 150 may be implemented on a cloud platform. By way of example only, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, across clouds, multiple clouds, or the like, or any combination thereof.

In some embodiments, a database 150 may be connected to the network 120 to communicate with one or more components of the abnormal driving behavior recognition system 100 (e.g., the server 110, the service requester terminal 130, the service provider terminal 140, etc.). One or more components of the system 100 for identifying abnormal driving behavior may access data or instructions stored in the database 150 via the network 120. In some embodiments, the database 150 may be directly connected to one or more components (e.g., the server 110, the service requester terminal 130, the service provider terminal 140, etc.) in the abnormal driving behavior recognition system 100; alternatively, in some embodiments, database 150 may also be part of server 110.

In some embodiments, one or more components (e.g., server 110, service requester terminal 130, service provider terminal 140, etc.) in the identification system 100 of abnormal driving behavior may have access to the database 150. In some embodiments, one or more components in the abnormal driving behavior identification system 100 may read and/or modify information related to a service requester, a service provider, or the public, or any combination thereof, when certain conditions are met. For example, server 110 may read and/or modify information for one or more users after receiving a service request.

Fig. 2 illustrates a schematic diagram of exemplary hardware and software components of an electronic device 200 of a server 110, a service requester terminal 130, a service provider terminal 140, which may implement the concepts of the present application, according to some embodiments of the present application. For example, the processor 112 may be used on the electronic device 200 and to perform the functions herein.

The electronic device 200 may be a general-purpose computer or a special-purpose computer, both of which may be used to implement the method of identifying abnormal driving behavior of the present application. Although only a single computer is shown, for convenience, the functions described herein may be implemented in a distributed fashion across multiple similar platforms to balance processing loads.

For example, the electronic device 200 may include a network port 210 connected to a network, one or more processors 220 for executing program instructions, a communication bus 230, and a different form of storage medium 240, such as a disk, ROM, or RAM, or any combination thereof. Illustratively, the computer platform may also include program instructions stored in ROM, RAM, or other types of non-transitory storage media, or any combination thereof. The method of the present application may be implemented in accordance with these program instructions. The electronic device 200 also includes an Input/Output (I/O) interface 250 between the computer and other Input/Output devices (e.g., keyboard, display screen).

For ease of illustration, only one processor is depicted in the electronic device 200. However, it should be noted that the electronic device 200 in the present application may also comprise a plurality of processors, and thus the steps performed by one processor described in the present application may also be performed by a plurality of processors in combination or individually. For example, if the processor of the electronic device 200 executes steps a and B, it should be understood that steps a and B may also be executed by two different processors together or separately in one processor. For example, a first processor performs step a and a second processor performs step B, or the first processor and the second processor perform steps a and B together.

Referring to a flowchart of a method for identifying abnormal driving behavior shown in fig. 3, the method may be executed by a device such as a platform server, and specifically may include the following steps:

step S302, acquiring driving data in the order execution process; the travel data includes travel track data and travel speed data. The travel speed data in the present embodiment may include one or more types of data related to speed, such as vehicle speed, vehicle angular velocity, and vehicle acceleration.

The platform server obtains the running data during the order execution process of the driver, such as obtaining the running track data and the running speed data recorded by the driver terminal during the order execution process. For example, when a driver starts to execute an order, the driver terminal monitors the driving behavior (also called driving behavior) of the driver during the order execution process, wherein the driver terminal may be a mobile phone, a central control device of an automobile, or the like. It is understood that various sensors, such as a position sensor, a speed sensor, a gyroscope, etc., are generally provided at the driver's terminal, and can be used to acquire driving data of the driver. In one embodiment, the travel track data may include position data recorded by a GPS sensor, or the like; the travel speed data may include: acceleration data, angular velocity data, and the like recorded by the velocity sensor.

In practical applications, the driving data may be acquired in real time by the driver's terminal, or may be acquired periodically, such as every 2 minutes or 5 minutes.

Step S304, calculating the probability of abnormal driving behaviors according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability.

In this embodiment, the abnormal driving behavior probability is mainly used for indicating the frequency of the abnormal driving behavior of the driver in the order on the road section that has been traveled, and in the specific implementation, the frequency of the abnormal driving behavior may be used for representation, and specifically, the probability is a normalized representation form of the frequency, so that the frequency of the abnormal driving behavior of the driver can be more clearly and intuitively represented.

In specific implementation, if one or more behaviors such as abnormal parking, abnormal speed change, abnormal turning or overspeed and the like occur to a driver, the driver can be considered to have abnormal driving behaviors. The platform can preset judgment standards of various abnormal driving behaviors, for example, if the driver is monitored not to stop at a passenger destination or at a special place such as a road jam or a gas station, the driver can be judged to stop abnormally; if the driver is monitored to have behaviors of rapid acceleration/rapid deceleration and the like, the abnormal speed change of the driver can be judged; if the situation that the angular speed of the driver terminal is changed excessively is monitored, the driver can be judged to turn abnormally; if the current speed of the driver is monitored to exceed the road section speed limit, the driver can be judged to overspeed. The foregoing is illustrative only and is not to be construed as limiting.

And step S306, identifying whether the abnormal driving behavior exists in the order execution process based on the abnormal driving behavior probability.

It can be understood that a driver may have multiple abnormal driving behaviors in the driving process, each abnormal driving behavior may also occur more than once, and finally, whether the driver has the abnormal driving behavior can be comprehensively identified by calculating the probability of the abnormal driving behavior occurring in the driven road section of the driver.

The method provided by the embodiment of the application can acquire the driving data in the order execution process, then calculate the probability of the abnormal driving behavior according to the driving data, and further identify whether the abnormal driving behavior exists in the order execution process based on the probability of the abnormal driving behavior. The mode can effectively identify whether the driver has abnormal driving behavior in time according to the driving data of the driver in the order execution process, thereby achieving better driver supervision effect.

In the present embodiment, when calculating the abnormal driving behavior probability from the driving data, for the sake of understanding, the calculation manners of the abnormal parking probability, the abnormal speed change probability, the abnormal turning probability, and the abnormal overspeed probability are respectively described as follows:

(I) calculation method of abnormal parking probability

In calculating the abnormal parking probability, the following steps may be referred to:

(1) and determining a parking position point in the driving process according to the driving track data.

(2) For each parking position point, acquiring road condition information corresponding to the parking position point and a corresponding POI (point of interest) identifier; and judging whether the parking behavior corresponding to the parking position point belongs to abnormal parking or not according to the road condition information and the POI (point of interest) identification.

Particularly, if the road condition information is that the road condition is smooth and easy, and the POI identification does not contain the preset identification, it is determined that the parking behavior corresponding to the parking position point belongs to abnormal parking. The preset type mark can comprise marks of reasonable parking spots such as a gas station mark, a market mark and an intersection mark. Or if the road condition information is road condition congestion and/or the POI identification comprises a preset identification, determining that the parking behavior corresponding to the parking position point does not belong to abnormal parking.

(3) And calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking.

For example, the sum of the parking time lengths of each parking position point belonging to the abnormal parking may be calculated to obtain the total parking time length; and then determining the abnormal parking probability according to the ratio of the total parking time to the driving time corresponding to the driving track data. Specifically, the time ratio of the abnormal parking on the traveled distance, that is, the abnormal parking probability, is mainly calculated. Let T be_aTime of use for the starting point to the current point of the order, T_stopT1+ t2+ t3 … tn; wherein T1-tn is T_aIf N abnormal parking points are detected in the time period, the abnormal parking probability is T_stop/T_a。

(II) method for calculating abnormal speed change probability

It can be understood that the driver often has the acceleration and deceleration condition in the driving process, but if the speed changes too much at one time and exceeds the preset speed change threshold, the driver can consider that the vehicle is in rapid acceleration/deceleration, so that not only is the potential safety hazard easily generated, but also the bad sitting experience can be brought to the passengers. In order to measure the frequency of abnormal speed change of a driver accurately and objectively, the method can be realized according to the ratio of the number of rapid speed change in all the speed change numbers.

When calculating the abnormal speed change probability, it may first check whether there is data greater than the speed change threshold in the acceleration data of the running speed data, then count the number of data greater than the speed change threshold according to the check result, and finally determine the abnormal speed change probability according to the ratio of the number of data to the total number of data of the acceleration data.

(III) calculation method for abnormal turning probability

Turning is a driving behavior which often occurs in the order execution process, but sharp turning brings about more serious potential safety hazard and is easy to cause passenger dissatisfaction, and similar to the calculation of the abnormal speed change probability, the frequent degree of abnormal turning of a driver can be measured according to the occupation ratio of the times of sharp turning in all turning times.

When calculating the abnormal turning probability, it may be first checked whether there is data larger than an angular velocity change threshold in the angular velocity data of the travel velocity data; then, counting the number of data larger than the angular speed change threshold according to the checking result; and finally, determining the abnormal turning probability according to the ratio of the number of the data to the total number of the data of the angular velocity data.

(IV) calculation method of overspeed probability

In this embodiment, if the current driving speed of the driver is higher than the speed limit of the road segment where the driver is located, the driver is considered to be speeding. In specific implementation, the overspeed probability of the driver can be judged according to the ratio of the overspeed duration to the total duration.

When calculating the overspeed probability, whether data of a preset speed limit larger than the corresponding position of the speed data exists in the speed data of the running speed data can be checked; then, counting the overspeed duration corresponding to the speed data larger than the preset speed limit; and finally, determining the overspeed probability according to the ratio of the overspeed duration to the corresponding travel duration of the travel track data.

In practical applications, when the abnormal behavior probability is calculated, one or more of the above-described abnormal parking probability calculation method, abnormal speed change probability calculation method, abnormal turning probability calculation method, and overspeed probability calculation method may be used.

After the abnormal driving behavior probability of the driver is calculated, whether the abnormal driving behavior exists in the order execution process can be identified based on the abnormal driving behavior probability. In order to obtain the recognition result quickly and effectively, in one embodiment, the driving data and the abnormal driving behavior probability can be input into an abnormal driving behavior recognition model obtained by training in advance, and whether the abnormal driving behavior occurs or not can be recognized through the abnormal driving behavior recognition model; the abnormal driving behavior recognition model is a neural network model. The neural network model generally includes one or more network layers, and network parameters are gradually adjusted through supervised or unsupervised learning until a desired prediction result can be finally output.

A specific implementation manner of an abnormal driving behavior recognition model is provided in this embodiment, and reference is made to a schematic structural diagram of the abnormal driving behavior recognition model shown in fig. 4, where the model includes an LSTM (Long Short-Term Memory) network and a full-connection network; where the input of the fully connected network is connected to the output of the last hidden layer (not shown in fig. 4) of the LSTM network.

In particular, the LSTM network is a time-recursive neural network, which allows information to persist, is suitable for processing and predicting important events with relatively long intervals and delays in time series, and can be used for processing sequence data, such as inputting driving data of a driver on a traveled distance into the LSTM network at different times, and in particular, the driving data can be arranged into a sequence form, such as a sequence of driving track coordinate points, data collected by a speed sensor, and the like. In one embodiment, the travel track coordinate point sequence may be understood as a GPS coordinate point sequence, each GPS point comprising two values (x, y) representing coordinate point longitude and latitude; the data collected by the speed sensor may include: the acceleration of the current point and the direction of the acceleration, and the roll angle, the pitch angle, the yaw angle and other angle data of the current driving direction. In order to enable the LSTM network to better process the travel data, the travel data may also be converted to floating point numbers.

A fully connected (Full Connection) network is mainly composed of a plurality of neurons, and can be used for completing the final binary task.

In the application of the abnormal driving behavior recognition model, driving data acquired from a driver's terminal may be input into the LSTM network, and the calculated abnormal driving behavior probability may be input into the fully-connected network. It can be understood that, because the output end of the last hidden layer of the LSTM network is further connected to the input end of the fully connected network, the fully connected network performs classification processing based on the probability of the abnormal driving behavior inputted from the outside and the data processing result inputted from the LSTM network, and finally obtains the result of identifying whether the abnormal driving behavior is present or not.

For the sake of understanding, reference may also be made to a specific structural diagram of an abnormal driving behavior recognition model shown in fig. 5, which illustrates a plurality of sub-units (cells) included in the LSTM network. Assuming that the driving data of the driver is acquired every 1 minute, the data acquired at the time t1 is input into the first subunit Cell1, the data acquired at the next minute of t1, that is, at the time t2 is input into the second subunit Cell2, and so on, and the data acquired at the time tn is input into the nth subunit Cell. In the calculation process of the LSTM network, the Cell1 processes the data input at the time t1 and inputs the processed data to the Cell2, the Cell2 performs comprehensive processing on the data input at the time t2 and the data transmitted by the Cell1, the processed data is input to the next subunit, and so on until the last subunit Celln outputs the processing result, and the processing result output by the last subunit Celln is regarded as the output of the LSTM network. Of course, it can also be understood that the above subunits are all hidden layers, each hidden layer correspondingly outputs a hidden state, and the hidden state of the last hidden layer is considered as the output of the LSTM network. In addition, fig. 5 illustrates that the data input to each subunit of the LSTM network includes not only GSP data and sensor speed, but also road segment attributes, which may include road data (such as whether the road condition is congested) of a GPS point corresponding to the current time, so that the LSTM network can perform better analysis processing on the data input at each time.

Finally, the output of the LSTM network and the abnormal driving behavior probability are both used as the input of the fully-connected network, and the fully-connected network performs a classification task to obtain a classification result (i.e., an identification result). In fig. 5, the circles shown in the fully connected network all represent network neurons.

In summary, when the abnormal driving behavior recognition model recognizes whether the abnormal driving behavior occurs, the driving state feature corresponding to the driving trajectory data (i.e., the last output of the LSTM network) may be generated through the LSTM network, the driving state feature may be input to the full-connection network, and then the driving state feature and the abnormal driving behavior probability may be subjected to classification regression through the full-connection network, so as to generate the abnormal driving behavior recognition result corresponding to the order execution process. The output of the LSTM network can be used as an intermediate value, and is used as an input value of the full-connection network together with abnormal driving behavior probabilities such as abnormal parking probability, abnormal speed change probability, abnormal turning probability, overspeed probability and the like, and finally, the full-connection network executes a classification task to obtain an identification result of the abnormal driving behavior.

If an abnormal driving behavior recognition model capable of outputting an accurate recognition result is obtained, the model needs to be trained in advance, the training process refers to that a sample data set is sent into a network, and the connection weight of the neural network model is adjusted according to the difference between the actual output and the expected output of the neural network model, namely the network parameters of the neural network model are adjusted. The specific training process can be seen as follows:

(1) acquiring a sample data set; the sample data set comprises positive sample data and negative sample data, and the positive sample data and the negative sample data both carry tags. The label can be understood as the expected output, and can be used as a reference to measure the difference between the actual output and the expected output of the model, so as to guide the model to perform reverse adjustment according to the difference. In a specific application, the label attached to the sample data can be determined according to the post-event feedback evaluation of the passenger.

In one mode, a machine screening mode is adopted to screen out marked behavior tags from an order database, and driving data corresponding to the behavior tags are obtained; wherein the behavior labels comprise positive behavior labels and negative behavior labels; marking the behavior label on the corresponding driving data; taking the driving data marked with the positive behavior label as positive sample data; and taking the driving data marked with the negative behavior label as negative sample data. The positive behavior label may be "driving smoothly" or the like, and the negative behavior label may be "abnormal driving" or the like, or may be more specific to detailed abnormal behaviors such as "sudden braking", "sudden acceleration", "abnormal parking", or the like.

Further, in order to more comprehensively acquire sample data and a corresponding label thereof, the driving data marked with a complaint label can be acquired, and the driving data marked with the complaint label is used as negative sample data; wherein the complaint labels are manually labeled. For example, the event of a telephone complaint is manually screened, and a complaint label is correspondingly marked, and the complaint label can also comprise 'abnormal driving' and the like, or detailed abnormal behaviors such as 'sharp turn' and the like.

It can be understood that the proportion of positive and negative sample data has certain influence on model training, and if the proportion of positive and negative sample data is larger, the classification capability of the model may be influenced, so that the model identification is deviated. For example, if the negative sample data is too little, the model may have difficulty in normally identifying abnormal driving behavior. After a plurality of experiments of the inventor, the proportion of the negative sample data to the positive sample data is not lower than 1:5, so that the model can obtain a more accurate identification result.

(2) Establishing a basic model structure; the infrastructure model structure includes sequentially connected LSTM networks and fully connected networks. Firstly, a basic model structure is built through an LSTM network and a full-connection network, and network parameters of the model structure can be preset initial values.

(3) And training the network parameters of the basic model structure through the sample data set to obtain an abnormal driving behavior recognition model. In specific implementation, a back propagation algorithm can be adopted to train the network parameters until the loss function value (representing the difference between the actual output and the expected output of the model) of the model converges to a preset threshold value, and it is determined that the output of the model at the moment meets the expected result, that is, the model is successfully trained.

If the abnormal driving behavior exists in the order execution process, in order to intervene the driver behavior in time and avoid safety accidents or conflict between the driver and passengers as much as possible, a preset mode can be adopted to initiate an abnormal reminding notification; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification. Specifically, an abnormal reminding notification can be sent to the driver terminal to warn the driver of the standard driving behavior. In addition, it is also possible to initiate an abnormal alert notification to the passenger and let the driver and/or passenger confirm whether there is an abnormal behavior.

In correspondence with the foregoing method for identifying an abnormal running behavior, the present embodiment provides an apparatus for identifying an abnormal running behavior, see a block diagram of the structure of an apparatus for identifying an abnormal running behavior shown in fig. 6, the apparatus including:

a driving data obtaining module 602, configured to obtain driving data in an order execution process; the driving data comprises driving track data and driving speed data;

an abnormal probability calculation module 604 for calculating an abnormal driving behavior probability according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability;

and an abnormal behavior identification module 606, configured to identify whether an abnormal driving behavior exists in the order execution process based on the abnormal driving behavior probability.

In the device provided by the embodiment of the application, the driving data can be acquired in the order execution process, the probability of the abnormal driving behavior is calculated according to the driving data, and whether the abnormal driving behavior exists in the order execution process is identified based on the probability of the abnormal driving behavior. The mode can effectively identify whether the driver has abnormal driving behavior in time according to the driving data of the driver in the order execution process, thereby achieving better driver supervision effect.

In some embodiments, the driving data acquisition module is configured to: acquiring driving track data and driving speed data recorded by a driver terminal in an order execution process, wherein the driving track data comprises position data recorded by a GPS sensor; the travel speed data includes: acceleration data, angular velocity data and velocity data recorded by the velocity sensor.

In some embodiments, the anomaly probability calculating module is configured to: determining a parking position point in the driving process according to the driving track data; for each parking position point, acquiring road condition information corresponding to the parking position point and a corresponding POI (point of interest) identifier; judging whether the parking behavior corresponding to the parking position point belongs to abnormal parking or not according to the road condition information and the POI (point of interest) identification; and calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking.

In some embodiments, the above-mentioned anomaly probability calculation module is further configured to: and if the road condition information is that the road condition is smooth and smooth, and the POI identification does not contain the preset identification, determining that the parking behavior corresponding to the parking position point belongs to abnormal parking.

In some embodiments, the above-mentioned anomaly probability calculation module is further configured to: calculating the sum of the parking time lengths of all the parking position points belonging to abnormal parking to obtain the total parking time length; and determining the abnormal parking probability according to the ratio of the total parking time to the driving time corresponding to the driving track data.

In some embodiments, the anomaly probability calculating module is configured to: checking whether data larger than a speed change threshold exists in acceleration data of the running speed data; counting the number of data larger than the speed change threshold according to the checking result; and determining the abnormal speed change probability according to the ratio of the data number to the total data number of the acceleration data.

In some embodiments, the anomaly probability calculating module is configured to: checking whether data larger than an angular speed change threshold exists in the angular speed data of the running speed data; counting the number of data larger than the angular speed change threshold according to the checking result; and determining the abnormal turning probability according to the ratio of the data number to the total data number of the angular velocity data.

In some embodiments, the anomaly probability calculating module is configured to: checking whether data of a preset speed limit larger than a corresponding position of the speed data exists in the speed data of the running speed data; counting the overspeed duration corresponding to the speed data larger than the preset speed limit; and determining the overspeed probability according to the ratio of the overspeed duration to the corresponding travel duration of the travel track data.

In some embodiments, the abnormal behavior identification module is configured to: inputting the running data and the abnormal running behavior probability into an abnormal running behavior recognition model obtained by pre-training, and recognizing whether abnormal running behaviors occur or not through the abnormal running behavior recognition model; the abnormal driving behavior recognition model is a neural network model.

In some embodiments, the abnormal driving behavior recognition model comprises an LSTM network and a fully connected network; wherein the input of the full connection network is connected to the output of the last hidden layer of the LSTM network.

In some embodiments, the above abnormal behavior identification module is further configured to: the method comprises the steps of inputting the running track data into an LSTM network, and inputting the abnormal running behavior probability into a fully-connected network.

In some embodiments, the above abnormal behavior identification module is further configured to: generating driving characteristics corresponding to the driving track data through an LSTM network, and inputting the driving characteristics into a full-connection network; and carrying out classification regression on the driving characteristics and the abnormal driving behavior probability through a full-connection network to generate an abnormal driving behavior recognition result corresponding to the order execution process.

In some embodiments, the apparatus further includes a training module of the abnormal driving behavior recognition model, and the training module is configured to: acquiring a sample data set; the sample data set comprises positive sample data and negative sample data, and the positive sample data and the negative sample data both carry tags; establishing a basic model structure; the basic model structure comprises an LSTM network and a fully connected network which are connected in sequence; and training the network parameters of the basic model structure through the sample data set to obtain an abnormal driving behavior recognition model.

In some embodiments, the training module is further configured to: screening the marked behavior tags from the order database by adopting a machine screening mode, and acquiring driving data corresponding to the behavior tags; wherein the behavior labels comprise positive behavior labels and negative behavior labels; marking the behavior label on the corresponding driving data; taking the driving data marked with the positive behavior label as positive sample data; and taking the driving data marked with the negative behavior label as negative sample data.

In some embodiments, the training module is further configured to: acquiring the driving data marked with the complaint label, and taking the driving data marked with the complaint label as negative sample data; wherein the complaint labels are manually labeled.

Referring to a block diagram of another abnormal driving behavior recognition apparatus shown in fig. 7, on the basis of fig. 6, the apparatus further includes: a reminding module 702, configured to initiate an abnormal reminding notification in a preset manner if an abnormal driving behavior is identified in the order execution process; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification.

The device provided by the embodiment has the same implementation principle and technical effect as the foregoing embodiment, and for the sake of brief description, reference may be made to the corresponding contents in the foregoing method embodiment for the portion of the embodiment of the device that is not mentioned.

It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the system and the apparatus described above may refer to corresponding processes in the method embodiments, and are not described in detail in this application. In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. The above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical division, and there may be other divisions in actual implementation, and for example, a plurality of modules or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or modules through some communication interfaces, and may be in an electrical, mechanical or other form.

The modules described as separate parts may or may not be physically separate, and parts displayed as modules 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 can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into 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 or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a U disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by 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

1. A method for identifying abnormal driving behavior, the method comprising:

acquiring driving data in the order execution process; the driving data comprises driving track data and driving speed data;

calculating the probability of abnormal driving behaviors according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability; the abnormal parking probability is calculated according to the parking time length of the abnormal parking determined from the driving data, the abnormal speed change probability is calculated according to the number of times of abnormal speed change determined from the driving data, the abnormal turning probability is calculated according to the number of times of abnormal turning determined from the driving data, and the overspeed probability is calculated according to the overspeed time length determined from the driving data;

inputting the driving data and the abnormal driving behavior probability into an abnormal driving behavior recognition model obtained by pre-training, and recognizing whether abnormal driving behaviors occur or not through the abnormal driving behavior recognition model; wherein, the abnormal driving behavior recognition model is a neural network model.

2. The method of claim 1, wherein the step of obtaining travel data during order fulfillment comprises:

acquiring driving track data and driving speed data recorded by a driver terminal in an order execution process, wherein the driving track data comprises position data recorded by a GPS sensor; the travel speed data includes: acceleration data, angular velocity data and velocity data recorded by the velocity sensor.

3. The method of claim 1, wherein the step of calculating the probability of abnormal driving behavior from the driving data comprises:

determining a parking position point in the driving process according to the driving track data;

for each parking position point, acquiring road condition information corresponding to the parking position point and a corresponding POI (point of interest) identifier; judging whether the parking behavior corresponding to the parking position point belongs to abnormal parking or not according to the road condition information and the POI (point of interest) identification;

and calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking.

4. The method according to claim 3, wherein the step of determining whether the parking behavior corresponding to the parking position point belongs to abnormal parking according to the road condition information and the POI (point of interest) identifier comprises:

and if the road condition information is that the road condition is smooth and smooth, and the POI identification does not contain the preset identification, determining that the parking behavior corresponding to the parking position point belongs to abnormal parking.

5. The method according to claim 3, wherein the step of calculating the abnormal parking probability according to the parking time length of the parking position point of the abnormal parking includes:

calculating the sum of the parking time lengths of all the parking position points belonging to abnormal parking to obtain the total parking time length;

and determining the abnormal parking probability according to the ratio of the total parking time to the driving time corresponding to the driving track data.

6. The method of claim 1, wherein the step of calculating the probability of abnormal driving behavior from the driving data comprises:

checking whether data larger than a speed change threshold exists in the acceleration data of the running speed data;

counting the number of data larger than the speed change threshold according to the checking result;

and determining the abnormal speed change probability according to the ratio of the data number to the total data number of the acceleration data.

7. The method of claim 1, wherein the step of calculating the probability of abnormal driving behavior from the driving data comprises:

checking whether data larger than an angular speed change threshold exists in the angular speed data of the running speed data;

counting the number of data larger than the angular speed change threshold according to the checking result;

and determining abnormal turning probability according to the ratio of the data number to the total data number of the angular velocity data.

8. The method of claim 1, wherein the step of calculating the probability of abnormal driving behavior from the driving data comprises:

checking whether data of a preset speed limit larger than a corresponding position of the speed data exists in the speed data of the running speed data or not;

counting the overspeed duration corresponding to the speed data larger than the preset speed limit;

and determining the overspeed probability according to the ratio of the overspeed duration to the driving duration corresponding to the driving track data.

9. The method of claim 1, wherein the abnormal driving behavior recognition model comprises an LSTM network and a fully connected network; wherein the input end of the full connection network is connected with the output end of the last hidden layer of the LSTM network.

10. The method according to claim 9, wherein the step of inputting the driving data and the abnormal driving behavior probability to an abnormal driving behavior recognition model trained in advance comprises:

inputting the driving data into the LSTM network, and inputting the abnormal driving behavior probability into the fully-connected network.

11. The method according to claim 10, wherein the step of identifying whether the abnormal driving behavior occurs by the abnormal driving behavior recognition model includes:

generating driving state characteristics corresponding to the driving track data through the LSTM network, and inputting the driving state characteristics to the full-connection network;

and performing classification regression on the driving state characteristics and the abnormal driving behavior probability through the full-connection network to generate an abnormal driving behavior identification result corresponding to the order execution process.

12. The method according to claim 9, wherein the training process of the abnormal driving behavior recognition model includes:

acquiring a sample data set; the sample data set comprises positive sample data and negative sample data, and the positive sample data and the negative sample data both carry tags;

establishing a basic model structure; the basic model structure comprises an LSTM network and a fully connected network which are connected in sequence;

and training the network parameters of the basic model structure through the sample data set to obtain an abnormal driving behavior recognition model.

13. The method of claim 12, wherein the step of obtaining a sample data set comprises:

screening the marked behavior tags from the order database by adopting a machine screening mode, and acquiring driving data corresponding to the behavior tags; wherein the behavior tags include positive behavior tags and negative behavior tags;

marking the behavior label on the corresponding driving data;

taking the driving data marked with the positive behavior label as positive sample data; and taking the driving data marked with the negative behavior label as negative sample data.

14. The method of claim 13, wherein the step of obtaining a sample data set further comprises:

acquiring the driving data marked with the complaint label, and taking the driving data marked with the complaint label as negative sample data; wherein the complaint labels are manually labeled.

15. The method of claim 12, wherein the ratio of the negative sample data to the positive sample data is not less than 1: 5.

16. The method of claim 1, further comprising:

if the abnormal driving behavior is identified in the order execution process, initiating an abnormal reminding notification in a preset mode; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification.

17. An apparatus for identifying an abnormal driving behavior, the apparatus comprising:

the driving data acquisition module is used for acquiring driving data in the order execution process; the driving data comprises driving track data and driving speed data;

the abnormal probability calculation module is used for calculating the abnormal driving behavior probability according to the driving data; the abnormal driving behavior probability includes at least one of: abnormal parking probability, abnormal speed change probability, abnormal turning probability and overspeed probability; the abnormal parking probability is calculated according to the parking time length of the abnormal parking determined from the driving data, the abnormal speed change probability is calculated according to the number of times of abnormal speed change determined from the driving data, the abnormal turning probability is calculated according to the number of times of abnormal turning determined from the driving data, and the overspeed probability is calculated according to the overspeed time length determined from the driving data;

the abnormal behavior recognition module is used for inputting the driving data and the abnormal driving behavior probability into an abnormal driving behavior recognition model obtained by pre-training and recognizing whether abnormal driving behaviors occur or not through the abnormal driving behavior recognition model; wherein, the abnormal driving behavior recognition model is a neural network model.

18. The apparatus of claim 17, wherein the travel data acquisition module is configured to:

19. The apparatus of claim 17, wherein the anomaly probability calculation module is configured to:

20. The apparatus of claim 19, wherein the anomaly probability calculation module is further configured to:

21. The apparatus of claim 19, wherein the anomaly probability calculation module is further configured to:

22. The apparatus of claim 17, wherein the anomaly probability calculation module is configured to:

23. The apparatus of claim 17, wherein the anomaly probability calculation module is configured to:

24. The apparatus of claim 17, wherein the anomaly probability calculation module is configured to:

25. The apparatus of claim 17, wherein the abnormal driving behavior recognition model comprises an LSTM network and a fully connected network; wherein the input end of the full connection network is connected with the output end of the last hidden layer of the LSTM network.

26. The apparatus of claim 25, wherein the abnormal behavior identification module is further configured to:

27. The apparatus of claim 26, wherein the abnormal behavior identification module is further configured to:

generating driving characteristics corresponding to the driving track data through the LSTM network, and inputting the driving characteristics into the full-connection network;

and carrying out classification regression on the driving characteristics and the abnormal driving behavior probability through the full-connection network to generate an abnormal driving behavior identification result corresponding to the order execution process.

28. The apparatus of claim 25, further comprising a training module of an abnormal driving behavior recognition model, the training module configured to:

29. The apparatus of claim 28, wherein the training module is further configured to:

marking the behavior label on the corresponding driving data;

30. The apparatus of claim 29, wherein the training module is further configured to:

31. The apparatus of claim 28, wherein a ratio of the negative sample data to the positive sample data is not lower than 1: 5.

32. The apparatus of claim 29, further comprising:

the reminding module is used for initiating an abnormal reminding notification in a preset mode if the abnormal driving behavior is identified in the order execution process; the preset mode comprises one or more of telephone notification, short message notification and APP message push notification.

33. An electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, when the electronic device runs, the processor and the storage medium communicate through the bus, and the processor executes the machine-readable instructions to execute the steps of the method for identifying abnormal driving behavior according to any one of claims 1 to 16.

34. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a computer program which, when being executed by a processor, carries out the steps of the method for identifying abnormal driving behavior according to any one of claims 1 to 16.