WO2021057134A1

WO2021057134A1 - Scenario identification method and computing device

Info

Publication number: WO2021057134A1
Application number: PCT/CN2020/097886
Authority: WO
Inventors: 李登宇
Original assignee: 华为技术有限公司
Priority date: 2019-09-27
Filing date: 2020-06-24
Publication date: 2021-04-01
Also published as: CN110796007A; CN110796007B

Abstract

The present application provides a scenario identification method. The method comprises: using an identification rule to perform detection on a state sequence that is determined according to driving data of a vehicle, the identification rule being determined according to a target scenario; and determining, according to a detection result, whether or not driving scenarios corresponding to the driving data comprise the target scenario. The application provides a solution in which driving scenarios can be automatically identified according to driving data of vehicles.

Description

Scene recognition method and computing equipment

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office, the application number is 201910927376.X, and the application name is "Scene Recognition Method and Computing Equipment" on September 27, 2019. The entire content is incorporated herein by reference. Applying.

Technical field

This application relates to the field of autonomous driving, and more specifically, to a method and computing device for scene recognition.

Background technique

As the society's demand for driving intelligence, economy, safety and other aspects increases, autonomous driving technology has become one of the key development directions of the automotive industry, and has received more and more attention from Internet companies.

At present, the industry usually uses simulation test methods to verify the functions of the autonomous driving system. By using simulation software, the real traffic environment is generated or reproduced in the simulation software in the form of simulation, so as to test whether the automatic driving system can correctly identify the surrounding environment. , And whether it can make timely and accurate responses to the surrounding environment and take appropriate driving behaviors.

The data required to build a simulation scene in the simulation software consists of high-precision map data and simulated traffic flow data. The high-precision map data provides roads, static traffic information (for example, traffic lights, road signs, etc.), and static object models (for example, Information such as buildings, trees), and simulated traffic flow data provide dynamic traffic flow (for example, traffic participants such as vehicles, pedestrians, etc.) information. The simulation software realizes the function of projecting the real world to the virtual world by loading and running this information, and copies the real scene in the automatic driving to the simulation software.

Real vehicle driving data and road information related to vehicle driving data are one of the main sources of data required to build a simulation scene. By combining the road information related to vehicle driving data (for example, lane line information, traffic sign information, Red street light information, static object information) is restored to high-precision map data, and vehicle driving data is restored to simulated traffic flow data, so that vehicle driving data and road information related to vehicle driving data are restored to what is needed to build a simulation scene data. Before that, the driving scene needs to be recognized first.

Summary of the invention

The present application provides a method and computing device for scene recognition, which can automatically recognize a driving scene according to the driving data of a vehicle.

In a first aspect, a method for scene recognition is provided, including: determining a first state sequence according to driving data of a vehicle, the first state sequence representing the first state of the vehicle at different times; In the first state sequence, the recognition rule is determined according to the target scene; according to the detection result, it is determined whether the driving scene corresponding to the driving data includes the target scene.

Based on the above technical solution, the first state sequence is acquired from the driving data of the vehicle, and the first state sequence is detected using the recognition rule determined according to the target scene to be recognized, and finally the driving scene corresponding to the driving data is determined according to the detection result. Whether to include the target scene, so as to realize the purpose of automatically identifying the driving scene according to the driving data of the vehicle.

In a possible implementation manner, the determining the first state sequence according to the driving data of the vehicle includes: determining the first state sequence according to the driving data of the vehicle and the target scene.

Based on the above technical solution, the first state sequence is determined according to the two factors of the driving data of the vehicle and the target scene to be recognized, so that the determined first state sequence can be more matched with the target scene, thereby improving the recognition of the target according to the first state sequence. The efficiency of the scene.

In a possible implementation manner, the determining whether the driving scene corresponding to the driving data includes the target scene according to the detection result includes: if the first state sequence includes a first subsequence, determining The driving scene corresponding to the driving data includes the target scene, and the first subsequence is a subsequence that satisfies the recognition rule; or, if the first state sequence does not include the first subsequence, it is determined The driving scene corresponding to the driving data does not include the target scene, and the first subsequence is a subsequence that satisfies the recognition rule.

Based on the above technical solution, the first sub-sequence is determined according to the recognition rule, thereby determining whether the first state sequence meets the recognition rule by detecting whether the first sub-sequence is included in the first state sequence, so as to determine whether the driving scene corresponding to the driving data is Including the target scene, and then realize the driving scene is automatically recognized according to the driving data of the vehicle.

In a possible implementation manner, the determining the first state sequence according to the driving data of the vehicle and the target scene includes: determining the second state sequence according to the driving data of the vehicle and the target scene Unlike the third state sequence, the second state sequence represents the second state of the vehicle at a different time, and the third state sequence represents the third state of the vehicle at a different time; according to the second state sequence and The third state sequence generates the first state sequence.

Based on the above technical solution, for some more complex target scenarios, at least two state sequences (for example, the second state sequence and the third state sequence) can be determined first, and then the first state sequence is generated according to the at least two state sequences, namely , By generating a first state sequence according to at least two state sequences, and detecting the first state sequence using a recognition rule to determine whether the driving scene corresponding to the driving data includes the target scene, thereby realizing the recognition of the more complex target scene.

It should be understood that in specific implementation, the second state sequence and the third state sequence may be state sequences that describe different information. For example, the target scene is a left-turning scene at a traffic light intersection, and the second state sequence may describe whether the vehicle is at a traffic light at different times. In the intersection, the third state sequence can describe the turning state of the vehicle at different moments.

In a possible implementation, the first state sequence is an m×n matrix, and the element in the i-th row and j-th column in the matrix represents the first state of the vehicle with index i at time j, m Is an integer greater than or equal to 1, n is an integer greater than or equal to 2, i is an integer greater than or equal to 1 and less than m, and j is an integer greater than or equal to 1 and less than n.

In a possible implementation manner, the driving scene corresponding to the driving data includes the target scene, and the method further includes: determining the second state sequence in the fourth state sequence according to the moment corresponding to the first sub-sequence Sub-sequence, the fourth state sequence represents the associated state of the vehicle, the fourth state sequence is determined according to the driving data of the vehicle; according to the second sub-sequence, the complexity of the target scene is determined .

In a possible implementation manner, the determining the complexity of the target scene according to the second subsequence includes: determining the complexity of the associated state of the vehicle according to the second subsequence; The complexity of the associated state of the vehicle is weighted to determine the complexity of the target scene.

In a possible implementation manner, the method further includes: determining a third sub-sequence in a fifth state sequence according to a time corresponding to the first sub-sequence, and the fifth state sequence indicates that the vehicle is in a different state. The position information at the time, the fifth state sequence is determined according to the driving data of the vehicle; the determining according to the detection result whether the target scene is included in the driving scene corresponding to the driving data includes: according to the The detection result and the third sub-sequence determine whether the driving scene corresponding to the driving data includes the target scene.

Based on the above technical solution, in order to further improve the accuracy of scene recognition, it is possible to obtain the position information of the vehicle at different times (ie, an example of the third subsequence), and determine the vehicle according to the position information of the vehicle at different times and road network topology information At least one of the roads, lanes and intersections at different times, and finally according to the detection results obtained in the front and at least one of the roads, lanes and intersections where the vehicle is at different times to comprehensively determine the corresponding driving data Whether the target scene is included in the driving scene, so as to improve the accuracy of scene recognition.

In a second aspect, the present application provides a computing device, including: a determining module, configured to determine a first state sequence according to the driving data of the vehicle, the first state sequence representing the first state of the vehicle at different times; processing The module is used to detect the first state sequence using a recognition rule, the recognition rule is determined according to the target scene; the determination module is also used to determine whether the driving scene corresponding to the driving data includes The target scene.

In a possible implementation manner, when the first state sequence is determined according to the driving data of the vehicle, the determining module is specifically configured to determine the first state sequence according to the driving data of the vehicle and the target scene.

In a possible implementation manner, when determining whether the driving scene corresponding to the driving data includes the target scene according to the detection result, the determining module is specifically configured to: if the first state sequence includes the first state sequence A sub-sequence, it is determined that the driving scene corresponding to the driving data includes the target scene, and the first sub-sequence is a sub-sequence that satisfies the recognition rule; or, if the first state sequence does not include the first state sequence A sub-sequence, it is determined that the driving scene corresponding to the driving data does not include the target scene, and the first sub-sequence is a sub-sequence that satisfies the recognition rule.

In a possible implementation manner, when the first state sequence is determined according to the driving data of the vehicle and the target scene, the determining module is specifically configured to: according to the driving data of the vehicle and the target scene , Determine a second state sequence and a third state sequence, the second state sequence represents the second state of the vehicle at different moments, and the third state sequence represents the third state of the vehicle at different moments; The second state sequence and the third state sequence are used to generate the first state sequence.

In a possible implementation manner, when the driving scene corresponding to the driving data includes the target scene, the determining module is further configured to: according to the moment corresponding to the first subsequence, in the fourth state sequence The second sub-sequence is determined in the, the fourth state sequence represents the associated state of the vehicle, the fourth state sequence is determined according to the driving data of the vehicle; the target is determined according to the second sub-sequence The complexity of the scene.

In a possible implementation manner, when the complexity of the target scene is determined according to the second subsequence, the determining module is specifically configured to: determine the association of the vehicle according to the second subsequence The complexity of the state; performing a weighted operation on the complexity of the associated state of the vehicle to determine the complexity of the target scene.

In a possible implementation manner, the determining module is further configured to: determine a third subsequence in a fifth state sequence according to a time corresponding to the first subsequence, and the fifth state sequence represents the vehicle For location information at different moments, the fifth state sequence is determined according to the driving data of the vehicle; when it is determined whether the target scene is included in the driving scene corresponding to the driving data according to the detection result, the determining The module is specifically configured to determine whether the target scene is included in the driving scene corresponding to the driving data according to the detection result and the third subsequence.

In a third aspect, a computing device is provided. The computing device includes a processor and a memory. The memory is used to store computer execution instructions. When the computing device is running, the processor executes the computer execution in the memory. The instructions are to execute the method steps in the first aspect and any one of the possible implementation manners of the first aspect through the computing device.

In a fourth aspect, a non-transitory readable storage medium is provided, including program instructions. When the program instructions are executed by a computing device, the computing device executes any one of the first aspect and the first aspect. The method in the implementation.

In a fifth aspect, a computer program product is provided, which includes program instructions. When the program instructions are executed by a computing device, the computing device executes any one of the first aspect and any one of the possible implementation manners of the first aspect. method.

On the basis of the implementation manners provided in the above aspects, this application can be further combined to provide more implementation manners.

Description of the drawings

FIG. 1 is a schematic block diagram of a scene recognition system provided by an embodiment of the present application;

FIG. 2 is a schematic flowchart of a method for scene recognition provided by an embodiment of the present application;

Figure 3 is a schematic diagram of the formal route of the vehicle in scene #2;

Fig. 4 is a schematic diagram of the formal route of the vehicle in scene #4;

FIG. 5 is a schematic structural diagram of a computing device 500 provided by an embodiment of the present application;

FIG. 6 is a schematic structural diagram of a computing device 600 provided by an embodiment of the present application.

detailed description

The technical solution in this application will be described below in conjunction with the accompanying drawings.

Real vehicle driving data and road information related to vehicle driving data are one of the main sources of data required to build a simulation scene. By combining the road information related to vehicle driving data (for example, lane line information, traffic sign information, Red street light information, static object information) is restored to high-precision map data, and vehicle driving data is restored to simulated traffic flow data, so that vehicle driving data and road information related to vehicle driving data are restored to what is needed to build a simulation scene data.

Since the function of the automatic driving system corresponds to the driving scene, the corresponding driving scene needs to be used when verifying the function of the automatic driving system. For example, the automatic driving system provides the automatic braking system (Autonomous Emergency Braking, AEB) function. When verifying this function, you need to verify this function in the AEB scenario. Therefore, it can be seen that before the driving data of the vehicle and the road information related to the driving data of the vehicle are restored to the data required to build a simulation scene, the driving scene needs to be identified first, so as to be associated with the identified driving scene. The driving data of the vehicle and the road information related to the driving data of the vehicle build a simulation scene, so as to verify the function of the automatic driving system in the simulation scene in the simulation software.

At present, a method of scene recognition is known. The method collects road data, analyzes the collected data, and uses a semi-automatic labeling method to recognize the scene. For example, most scenes require manual observation of video data to be recognized, and the recognition of driving scenes cannot be fully automated. The video data appearing here and in the following may refer to the video data acquired by the camera installed on the vehicle during the driving of the vehicle.

Therefore, the present application provides a method for scene recognition, which can automatically recognize the driving scene according to the driving data of the vehicle. The following describes in detail the method for scene recognition provided in the present application with reference to FIGS. 1 to 4.

FIG. 1 is a schematic block diagram of a scene recognition system 100 provided by the present application. The system 100 may include a collection device 101 and a computing device 102.

The collection device 101 is mainly responsible for collection functions. In specific implementation, the collection device 101 may be a vehicle or an urban traffic monitoring device, and the vehicle here may be a vehicle equipped with an automatic driving system. Hereinafter, the data obtained during the driving of the vehicle is referred to as road acquisition data, and the road traffic data obtained by the urban traffic monitoring device is referred to as urban traffic flow monitoring data.

Among them, a variety of sensors can be installed on the vehicle. This application does not specifically limit the sensors installed on the vehicle, which may include but are not limited to: several cameras, at least one radar, at least one positioning system, and at least one inertial measurement unit (IMU).

Among them, several cameras can be respectively deployed around the vehicle and collect environmental parameters around the vehicle. For example, at least one camera may be installed on the front and rear bumpers, side-view mirrors, and windshield of the vehicle.

The radar may include at least one of ultrasonic radar, laser radar, and millimeter wave radar. The radar can measure parameter information such as the distance and speed of the vehicle. Radar can also use radio signals to sense objects in the surrounding environment of the vehicle. Optionally, in some embodiments, in addition to sensing the object, the radar can also be used to sense the forward direction of the object.

The positioning system may be a global positioning system (GPS), Beidou system or other positioning systems, which are used to receive satellite signals and locate the current position of the vehicle.

The IMU can sense the position and orientation changes of the vehicle based on the inertial acceleration. Optionally, in one embodiment, the IMU may be a combination of an accelerometer and a gyroscope, and is used to measure the angular velocity and acceleration of the vehicle.

The collection device 101 and the computing device 102 can communicate with each other through a network or storage medium. For example, the collection device 101 can transmit road collection data and/or urban traffic flow monitoring data to the computing device 102 through a transmission method such as a network or storage medium. The computing device 102 recognizes the driving scene based on road acquisition data and/or urban traffic flow monitoring data.

Among them, road mining data can include but not limited to: camera, millimeter wave radar, lidar, ultrasonic radar, GPS, IMU and other sensor data, high-precision map data, algorithm output data, vehicle-to-everything communication (vehicle-to-everything) , V2X) data and vehicle control data and other data that can be collected; urban traffic flow monitoring data can include but not limited to: vehicle trajectory information and vehicle information in the traffic flow.

Fig. 2 is a schematic flowchart of a method for scene recognition provided by the present application. The method includes steps 210-230, and steps 210-230 are described in detail below.

Step 210: Determine the state sequence #1 (ie, an example of the first state sequence) according to the driving data of the vehicle. The state sequence #1 represents the state #1 of the vehicle at different times (ie, an example of the first state). The driving data of the vehicle here can be road collection data and/or urban traffic flow monitoring data.

After the computing device 102 obtains the driving data of the vehicle, it can determine the state #1 of the vehicle at different times according to the driving data of the vehicle. For example, the state #1 may be the driving speed of the vehicle, and the driving speed of the vehicle at different times constitutes the state. Sequence #1. Wherein, the driving data of the vehicle may include the positioning information of the vehicle obtained by the positioning system on the vehicle, and the computing device 102 may calculate the driving speed of the vehicle according to the positioning information of the vehicle.

Step 220: Use the recognition rule to detect the state sequence #1, and the recognition rule is determined according to the target scene.

The computing device 102 may determine a recognition rule according to the target scene to be recognized, and use the recognition rule to detect the state sequence #1.

Step 230: According to the detection result, it is determined whether the driving scene corresponding to the driving data includes the target scene.

The computing device 102 may determine whether the target scene is included in the driving scene corresponding to the driving data according to the detection result obtained after detecting the state sequence #1 using the recognition rule.

Optionally, step 210 can also be replaced with: determining the state sequence #1 according to the driving data of the vehicle and the target scene.

When the computing device 102 acquires the state #1 of the vehicle at different moments, it can acquire it in combination with the target scene. For example, the target scene can be the scene where the vehicle in front of the left cuts into the scene. At this time, the computing device can use the target scene from the driving of other vehicles. The data and the driving data of the own vehicle obtain the position of other vehicles relative to the own vehicle, and use the obtained position of other vehicles relative to the own vehicle as the state #1 of other vehicles at different times, and according to the state of other vehicles at different times #1, confirm status sequence #1.

Optionally, step 230 may be specifically implemented in the following manner:

For example, if the state sequence #1 includes the subsequence #1 (ie, an example of the first subsequence), it is determined that the driving scene corresponding to the driving data includes the target scene; or, if the state sequence #1 does not include the subsequence # 1. It is determined that the driving scene corresponding to the driving data does not include the target scene. Among them, subsequence #1 is a subsequence that meets the recognition rules.

For example, the target scene is a scene where the driving speed of the vehicle is greater than or equal to 30km/h, and the driving speed of the same vehicle at different times is recorded in the state sequence #1, then the recognition rule corresponding to the target scene can be in the state sequence #1 Including continuously occurring driving speeds greater than or equal to 30km/h, that is, subsequence #1 includes continuously occurring driving speeds greater than or equal to 30km/h. If the computing device 101 can detect continuously occurring driving speeds greater than or equal to 30km/h in the state sequence #1, it can be determined that the driving scene corresponding to the driving data includes the target scene; otherwise, the driving scene corresponding to the driving data is determined Does not include the target scene.

In addition, the identification can also be used to indicate the speed of the vehicle at different times. For example, “1” means that the speed of the vehicle at a certain moment is greater than or equal to 30km/h, and “0” means other conditions, for example, “0”. "It means that the speed of the vehicle at a certain moment is less than 30km/h or the vehicle is parked at a certain moment. At this time, what is recorded in the state sequence #1 is an identifier that can reflect the driving speed of the same vehicle at different moments, and the recognition rule corresponding to the target scene can be that the state sequence #1 includes consecutive "1"s, that is, a sub-sequence #1 can be a "1" that appears consecutively. If the computing device 101 can detect successive occurrences of 1s in the state sequence #1, it can be determined that the driving scene corresponding to the driving data includes the target scene; otherwise, it is determined that the driving scene corresponding to the driving data does not include the target scene.

In this application, the state sequence can be stored in the form of a matrix (hereinafter referred to as “state matrix” for short). For example, the state sequence can be stored as a state matrix with m rows and n columns (ie, m×n), where the i-th row The element in the jth column of can represent the state of the vehicle with index i (hereinafter referred to as "vehicle #i") at time j, where m is an integer greater than or equal to 1, n is an integer greater than or equal to 2, and i is greater than Or an integer equal to 1 and less than or equal to m, and j is an integer greater than or equal to 1 and less than or equal to n.

In the following, the state matrix is taken as an example, combining several specific scenarios to illustrate the method of scene recognition provided in this application.

Scenario #1 A scene where the vehicle is driving at a speed greater than or equal to 50km/h.

In order to identify the scene #1, the driving speed of the own vehicle may be obtained first. For example, the computing device 102 may calculate the driving speed of the own vehicle according to the positioning information of the vehicle obtained by the positioning system on the own vehicle.

At this time, the state matrix #1 (corresponding to the state sequence #1) may be a matrix with a size of 1×n, and n represents the total recorded time length for the own vehicle in the state matrix #1. _{The elements e 1, j} of the state matrix #1 represent the traveling speed of the own vehicle at time j. For example, "1" means that the driving speed of the vehicle at a certain moment is greater than or equal to 50km/h, "0" means other conditions, then _{the value of e 1, j} can be expressed as follows:

The above other conditions may indicate that the driving speed of the own vehicle at time j is less than 50 km/h or the own vehicle is in a parking state at a certain time.

The computing device 102 determines the corresponding identification rule for the scene #1 as: identifying all consecutive "1"s in the state matrix #1, and then the computing device 102 uses the identification rule to detect the state matrix #1. For example, state matrix #1 can be:

00001111111111100111111111111111111111111111000111100

It can be seen from the state matrix #1 that the driving speed of the own vehicle at time #5-time #15, time #18-time #40, and time #44-time #47 are all greater than or equal to 50 km/h. Therefore, the computing device 102 can identify 3 scenes #1 from the state matrix #1, where the time corresponding to each scene #1 is: time #5-time #15, time #18-time #40, time #44-时间#47, the computing device 102 may associate the driving data of the own vehicle corresponding to time#5-time#15, time#18-time#40, and time#44-time#47 with scene #1, thereby In the simulation software, the driving data of the own vehicle corresponding to time #5-time #15, time #18-time #40, and time #44-time #47 and road information related to the driving data of the vehicle are used to build scene #1, Thus, the function of the autonomous driving system of the vehicle in scenario #1 is verified in the simulation software.

Scene #2 The front left vehicle cuts into the scene.

In order to recognize scene #2, the computing device 102 needs to determine the location of the own vehicle and the location of other vehicles, so as to determine the location of other vehicles relative to the own vehicle. For example, the location of the own vehicle can be based on the positioning system or IMU on the own vehicle. The location information of the obtained vehicle is obtained, and the location of other vehicles can be obtained from the video data, or can also be obtained from the scanning information of the radar.

The computing device 102 can calculate the position of other vehicles relative to the own vehicle at each time according to the position of the own vehicle at each time and the position of other vehicles, and generate a state matrix of size m×n according to the position of other vehicles relative to the own vehicle at each time. #1, m represents the total number of other vehicles recorded in the state matrix #1, and n represents the total recorded time length for each vehicle in the state matrix #1.

_{The element p i, j of the} state matrix #1 represents the relative position of the vehicle #i (that is, an example of other vehicles) with respect to the own vehicle at time j. For example, "1" indicates that vehicle #i is located in the front left direction of the host vehicle at time j, "2" indicates that vehicle #i is located in the front direction of the host vehicle at time j, and "3" indicates that vehicle #i is located in front of the host vehicle at time j The front right direction. For the sake of brevity, I will not list them one by one here _{, and the values of p i,j} can be expressed as follows:

The recognition rule determined by the computing device 102 for the scenario #2 is: the element of the state matrix #1 changes from "1" to "2", and then the computing device 102 uses the recognition rule to detect the state matrix #1. For example, state matrix #1 can be:

It can be seen from the state matrix #1 that the vehicle #2 cuts in from the left front of the vehicle at time #6-time #7. Therefore, the computing device 102 can recognize scene #2 from the state matrix #1, and the computing device 102 The driving data of the host vehicle and other vehicles corresponding to time #6-time #7 can be associated with scene #2, so that the driving data of the host vehicle corresponding to time #6-time #7 and the driving data of the host vehicle can be used in the simulation software. The road information related to the driving data, the driving data of other vehicles, and the road information related to the driving data of other vehicles build scene #2, so that the function of the vehicle's automatic driving system under scene #2 (for example, the deceleration function) can be compared in the simulation software. )authenticating.

Scene #3 Turn left at a traffic light intersection.

In order to recognize scene #3, the computing device 102 may first determine whether the own vehicle is in a red intersection at different times (ie, an example of the second state), and determine the steering state of the same vehicle at different times (ie, the third An example of status). The computing device 102 can generate state sequence #2 (that is, an example of the second state sequence) according to multiple states of whether the vehicle is in a traffic light intersection at different times, and generate states based on multiple steering states of the same vehicle at different times In sequence #3 (ie, an example of the third state sequence), state sequence #1 is generated based on state sequence #2 and state sequence #3. Among them, the state sequence #2 and the state sequence #3 are both stored in the form of a matrix, that is, the state sequence #2 corresponds to the state matrix #2, and the state sequence #3 corresponds to the state matrix #3.

The computing device 102 generates a state matrix #2 of size m×n according to whether the vehicle is in multiple states at the traffic light intersection at different times, where m represents the total number of vehicles recorded in the state matrix #2, and n represents the state matrix# The total time recorded in 2 for each vehicle.

_{The element r i,j of the} state matrix #2 indicates whether the vehicle #i (that is, an example of the own vehicle) is within the traffic light intersection at time j. For example, "1" indicates that the vehicle #i is in the traffic light intersection at time j, and "0" indicates that the vehicle #i is in another state at time j. Then _{the value of r i,j} can be expressed as follows:

The computing device 102 generates a state matrix #3 with a size of m×n according to multiple steering states of the vehicle at different times, where m represents the total number of vehicles recorded in the state matrix #3, and n represents the number of vehicles recorded in the state matrix #3 for each The total time recorded by each vehicle.

_{The element s i,j of the} state matrix #3 represents the turning state of the vehicle #i at time j. For example, "1" indicates that vehicle #i is turning left at time j, "2" indicates that vehicle #i is turning right at time j, and "0" indicates that vehicle #i is in another state at time j, for example, It means that the vehicle #i did not make a steering operation at time j. Then _{the value of s i,j} can be expressed as follows:

The computing device 102 may generate the state matrix #1 according to the state matrix #2 and the state matrix #3. The elements t _i,j in the state matrix #1 represent the turning state of the vehicle #i at the traffic light intersection at time j. For example, "1" indicates that vehicle #i turns left at a traffic light intersection at time j, "2" indicates that vehicle #i turns right at a traffic light intersection at time j, and "0" indicates that vehicle #i is in another state at time j. Then _{the value of ti,j} can be expressed as follows:

When the recognition rule determined by the computing device 102 for scenario #3 is: recognize whether the state matrix #1 includes 1, and when the state matrix #1 includes 1, the computing device 102 can recognize the scenario #3 from the state matrix, and calculate The device 102 can associate the driving data of the host vehicle at the time corresponding to element 1 with scene #3, so as to construct the road information related to the driving data of the host vehicle at the time corresponding to element 1 and the driving data of the host vehicle in the simulation software. Scene #3, so as to verify the function of the self-driving system of the vehicle in scene #3 in the simulation software.

It should be understood that the foregoing only uses the storage of the state sequence in the form of a matrix as an example to exemplarily introduce the scenarios #1 to #3, but the application is not limited thereto. The state sequence can also be stored in other forms, for example, it can be stored in the form of a list. Not only that, any other storage form that can reflect the state of the vehicle at different times should fall within the protection scope of this application.

The recognition rules in this application can also be described by regular expressions. For example, the recognition rules in scenario #1 can be described by regular expressions "1+", "1+" represents a continuous 1, and in scenario #2 The recognition rule can be described by the regular expression "12", "12" means changing from "1" to "2".

The foregoing specifically explains how the computing device 102 recognizes a scene. In this application, the complexity of the scene can also be determined for the recognized scene, which will be described in detail below.

At this time, the method 200 may further include: the computing device 102 determines the state sequence #4 (that is, an example of the fourth state sequence) according to the driving data of the vehicle, and the state sequence #4 may represent the associated state of the own vehicle. For example, the associated state of the own vehicle may include at least one of the road environment state, the state of the own vehicle, and the state of other vehicles. Therefore, the state sequence #4 may include the road environment state sequence, the state sequence of the own vehicle, and the state of other vehicles. At least one of the state sequence. The computing device 102 determines the subsequence #2 (ie, an example of the second subsequence) in the state sequence #4 according to the time corresponding to the subsequence #1, and determines the complexity of the target scene according to the subsequence #2.

The road environment state sequence can describe the road environment where the vehicle is located at different moments. For example, the road environment state sequence can describe the vehicle driving on a curved road at the previous time and driving on a straight road at the next time. The vehicle state sequence can describe this The speed of the vehicle at different times and the state sequence of other vehicles can describe the distance between other vehicles and the vehicle at different times.

When the computing device 102 determines the complexity of the target scene, it can calculate the same time in the state matrix #4 (corresponding to the state sequence #4) according to the time corresponding to the subsequence #1 of the target scene in the state matrix #1 The element of is determined as sub-sequence #2, and the complexity of the target scene is determined according to sub-sequence #2.

For example, in the above scenario #1, the traveling speed of the own vehicle at time #44-time #47 is greater than or equal to 50 km/h. When determining the complexity of the scene #1, the computing device 102 may determine the element corresponding to the time #44-the time #47 in the state sequence #4 as the subsequence #2.

For example, the state sequence #4 includes the road environment state sequence, the state sequence of the own vehicle, and the state sequence of other vehicles. The computing device 120 may determine the element corresponding to the time #44-时间#47 in the road environment state sequence corresponding to the own vehicle. Is the sub-sequence #2 corresponding to the state sequence of the road environment, the element corresponding to time #44-time #47 in the state sequence of the own vehicle is determined as the sub-sequence #2 corresponding to the state sequence of the own vehicle, and the state sequence of other vehicles The element corresponding to time #44-time #47 in is determined as the sub-sequence #2 corresponding to the state sequence of other vehicles.

The computing device 102 may determine the associated state of the vehicle according to at least one of the subsequence #2 corresponding to the road environment state sequence, the subsequence #2 corresponding to the state sequence of the own vehicle, and the subsequence #2 corresponding to the state sequence of other vehicles. Determine the complexity of scene #1 according to the complexity of the associated state of the vehicle. Wherein, the complexity of the associated state of the host vehicle includes at least one of the complexity of the road environment, the complexity of the host vehicle, and the complexity of other vehicles.

For example, the computing device 102 may determine the complexity of the road environment according to the sub-sequence #2 corresponding to the state sequence of the road environment, determine the complexity of the own vehicle according to the sub-sequence #2 corresponding to the state sequence of the own vehicle, and correspond to the state sequence of other vehicles The sub-sequence #2, to determine the complexity of other vehicles.

The computing device 102 may perform a weighted operation on at least one of the complexity of the road environment, the complexity of the own vehicle, and the complexity of other vehicles, so as to determine the complexity of the scene #1.

For example, for scene #1 and scene #2, the computing device 102 respectively performs a weighted calculation on the road environment complexity, the complexity of the own vehicle and the complexity of other vehicles corresponding to scene #1 and scene #2, and the final scene #1 and The complexity of scenario #2 is shown in the following table:

Table 1

场景Scenes	道路环境复杂度Road environment complexity	其他车辆复杂度Other vehicle complexity	本车辆复杂度Complexity of own vehicle		场景复杂度Scene complexity
11	0.130.13	0.000.00	0.050.05	0.020.02
22	0.130.13	0.300.30	0.050.05	0.180.18

It should be understood that the foregoing only uses the method of weighting at least one of the complexity of the road environment, the complexity of the own vehicle and the complexity of other vehicles to determine the complexity of the scene as an exemplary description, but this does not constitute a special limitation to the application. Other methods for determining the complexity of the scene based on at least one of the complexity of the road environment, the complexity of the vehicle, and the complexity of other vehicles should all fall within the scope of protection of this application.

In the embodiment of the present application, in order to further improve the accuracy of scene recognition, the computing device 102 may obtain map information, and determine multiple roads, multiple lanes, and multiple junctions according to the map information. According to the location information of multiple roads, multiple lanes, and multiple intersections, the road network topology information can be constructed. The road network topology information can be used to determine which lane the vehicle is located on. And the spatial relationship between each road, lane, and junction. Wherein, the location information of multiple roads, multiple lanes, and multiple intersections may be the coordinates of multiple roads, multiple lanes, and multiple intersections in the map information.

After the computing device 102 obtains the driving data of the vehicle, it can determine the position information of the vehicle at different times according to the driving data of the vehicle. According to the position information of the vehicle at different times, and combined with the road network topology information, it can determine the road on which the vehicle is located. , At least one of lanes and intersections.

When determining whether the driving scene corresponding to the driving data includes the target scene, a comprehensive judgment is made according to the above detection result and at least one of the road, lane, and intersection on which the vehicle is located, so as to improve the accuracy of scene recognition.

At this time, the method 200 may further include:

According to the time corresponding to the subsequence #1, the subsequence #3 (that is, an example of the third subsequence) is determined in the state sequence #5 (that is, an example of the fifth state sequence), and the state sequence #5 indicates that the vehicle is at a different time The position information of the state sequence #5 is determined according to the driving data of the vehicle. Step 230 can be replaced with:

According to the detection result and sub-sequence #3 (that is, an example of the third sub-sequence), it is determined whether the driving scene corresponding to the driving data includes the target scene.

The above scenario #2 to scenario #3 will be further described below in conjunction with this method.

In scenario #2, when the computing device 102 recognizes that the vehicle #2 has moved from the front left direction of the host vehicle to the front direction of the host vehicle at time #6-time #7, the vehicle #2 can also be determined from the state sequence #5 Position information at time #6-time #7 (ie, an example of sub-sequence #3), and based on the position information of vehicle #2 at time #6-time #7 and road network topology information, it is determined that vehicle #2 is at time #6-时间#7 is in the lane, state sequence #5 includes the location information of vehicle #2 at different moments, where the location information of vehicle #2 at different moments can be the location of vehicle #2 at different moments on the map The coordinates in the message.

For example, as shown in FIG. 3, the computing device 102 determines that vehicle #2 has the original lane changed to the original lane at time #6-time #7 based on the location information and road network topology information of vehicle #2 at time #6-time #7 The right lane of the lane, the computing device 102 can move from the front left direction of the own vehicle to the front direction of the own vehicle according to the vehicle #2 at time #6-time #7, and the vehicle #2 at time #6-time #7 from the original With the two detection results that the lane becomes the right lane of the original lane, it is determined that the vehicle #2 cuts in from the front left of the own vehicle at time #6-time #7, thereby completing the recognition of scene #2 more accurately.

In scenario #3, when the computing device 102 recognizes that the vehicle #i turns left at a traffic light intersection at time j, it can also determine the position information of vehicle #i at time j from the state sequence #5 (ie, sub-sequence #3 Another example), and according to the location information of vehicle #i at time j and the road network topology information, it is determined that vehicle #i is located in the intersection at time j, then the computing device 102 can be left at the traffic light intersection at time j according to vehicle #i Turn, the vehicle #i is located in the intersection at time j, the two detection results, it is determined that the vehicle #i turns left at the traffic light intersection at time j, so as to more accurately complete the recognition of scene #3. The state sequence #5 includes the position information of the vehicle #i at different times, where the position information of the vehicle #i at different times may be the coordinates of the position of the vehicle #i at different times in the map information. A detailed description will be given below in conjunction with scenario #4 provided in the embodiment of the present application.

Scenario #4 The own vehicle goes straight at a traffic light intersection, and the target vehicle (ie, an example of other vehicles) turns right.

In order to identify scene #4, the computing device 102 can first determine that the vehicle is going _{straight at the traffic light intersection at time #t 1} to time #t ₂ according to the method described in scene #3, and can further determine the current vehicle from the state sequence #5. The location information of the vehicle at time #t ₁ to time #t ₂ (ie, another example of sub-sequence #3), and _{the location information of the vehicle at time #t 1} to time #t ₂ and road network topology information are determined On the road where the host vehicle is located at time #t ₁ to time #t ₂ , the state sequence #5 includes the location information of the host vehicle at different times. The location information of the host vehicle at different times may be the location information of the host vehicle at different times. The coordinates of the location in the map information.

For example, as shown, the computing device 102 according to the vehicle at the time #t #t 4 ₁ ~ time position information and the road network topology information _2, the present vehicle is determined at time #t #t ₁ ~ ₂ by the time the road # 1 Go through the red street light intersection to reach road #3.

The computing device 102 can determine that the target vehicle _{turns right at the traffic light intersection at time #t 1} to time #t ₂ according to the method described in scenario #3. After that, the computing device 102 can determine that the target vehicle is at time # from the state sequence #5. t ₁ ~ time #t ₂ location information (ie, another example of subsequence #3), and based on _{the location information of the target vehicle at time #t 1} ~ time #t ₂ and road network topology information, determine the target vehicle's location at time #t ₁ ～时间#The _{road where t 2} is located, the state sequence #5 includes the location information of the target vehicle at different moments, where the location information of the target vehicle at different moments can be the location information of the target vehicle at different moments on the map. In the coordinates.

For example, the computing device 102 determines that the target vehicle arrives at the road from the road #2 through the red street light intersection at _{time #t 1} ～time #t ₂ according to the location information and road network topology information of the target vehicle at time #t ₁ ～time #t ₂ #3.

The computing device 102 may #t ₂ ₁ to time in accordance with the present vehicle traffic lights straight at time #t, the target vehicle at time ₁ to time #t #t ₂ right turn at the traffic light and the target vehicle at time ₁ to time #t _{#t 2}由路#2 Pass the red street light intersection to reach the road #3 These three detection results, it is determined that the vehicle is going _{straight at the traffic light intersection at time #t 1} ～ time #t ₂ and the target vehicle is at time #t ₁ ～ time #t ₂ Turn right at the traffic light intersection to more accurately complete the recognition of scene #4.

It should be understood that in the various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not correspond to the embodiments of the present application. The implementation process constitutes any limitation.

The foregoing describes in detail a method for scene recognition provided by an embodiment of the present application with reference to FIGS. 1 to 4, and the following describes in detail an embodiment of the apparatus of the present application with reference to FIGS. 5 to 6. It should be understood that the description of the method embodiment and the description of the device embodiment correspond to each other, and therefore, the parts that are not described in detail may refer to the previous method embodiment.

FIG. 5 is a schematic structural diagram of a control device 500 provided by an embodiment of the present application. The control device 500 includes:

The determining module 510 is configured to determine a first state sequence according to the driving data of the vehicle, where the first state sequence represents the first state of the vehicle at different times;

The processing module 520 is configured to detect the first state sequence using a recognition rule, the recognition rule being determined according to a target scene;

The determining module 510 is further configured to determine whether the target scene is included in the driving scene corresponding to the driving data according to the detection result.

It should be understood that the computing device 500 provided in the embodiments of the present application may be implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), and the above PLD may be a complex program. Logic device (complex programmable logical device, CPLD), field-programmable gate array (field-programmable gate array, FPGA), general array logic (generic array logic, GAL) or any combination thereof. When the method of scene recognition shown in FIG. 2 can also be implemented by software, each module of the computing device 500 may also be a software module.

Optionally, in some embodiments, when the first state sequence is determined according to the driving data of the vehicle, the determining module 510 is specifically configured to: determine the first state according to the driving data of the vehicle and the target scene sequence.

Optionally, in some embodiments, when determining whether the target scene is included in the driving scene corresponding to the driving data according to the detection result, the determining module 510 is specifically configured to:

If the first state sequence includes a first subsequence, it is determined that the driving scene corresponding to the driving data includes the target scene, and the first subsequence is a subsequence that satisfies the recognition rule; or,

If the first state sequence does not include the first subsequence, it is determined that the driving scene corresponding to the driving data does not include the target scene, and the first subsequence is a subsequence that satisfies the recognition rule.

Optionally, in some embodiments, when the first state sequence is determined according to the driving data of the vehicle and the target scene, the determining module 510 is specifically configured to:

According to the driving data of the vehicle and the target scene, a second state sequence and a third state sequence are determined, the second state sequence represents the second state of the vehicle at different moments, and the third state sequence represents all State the third state of the vehicle at different moments;

The first state sequence is generated according to the second state sequence and the third state sequence.

Optionally, in some embodiments, the first state sequence is a matrix of size m×n, and the element in the i-th row and j-th column in the matrix represents the first state of the vehicle with index i at time j, m is an integer greater than or equal to 1, n is an integer greater than or equal to 2, i is an integer greater than or equal to 1 and less than m, and j is an integer greater than or equal to 1 and less than n.

Optionally, in some embodiments, when the driving scene corresponding to the driving data includes the target scene, the determining module 510 is further configured to:

According to the time corresponding to the first sub-sequence, a second sub-sequence is determined in the fourth state sequence, the fourth state sequence represents the associated state of the vehicle, and the fourth state sequence is based on the driving of the vehicle Data confirmed;

According to the second subsequence, the complexity of the target scene is determined.

Optionally, in some embodiments, when determining the complexity of the target scene according to the second subsequence, the determining module 510 is specifically configured to:

Determine the complexity of the associated state of the vehicle according to the second sub-sequence;

Perform a weighted operation on the complexity of the associated state of the vehicle to determine the complexity of the target scene.

Optionally, in some embodiments, the determining module 510 is further configured to:

According to the time corresponding to the first sub-sequence, a third sub-sequence is determined in the fifth state sequence, the fifth state sequence represents the position information of the vehicle at different moments, and the fifth state sequence is based on the The driving data of the vehicle is determined;

When determining whether the driving scene corresponding to the driving data includes the target scene according to the detection result, the determining module is specifically configured to:

According to the detection result and the third sub-sequence, it is determined whether the driving scene corresponding to the driving data includes the target scene.

The computing device 500 according to the embodiment of the present application may correspond to executing the method described in the embodiment of the present application, and the foregoing and other operations and/or functions of each unit in the computing device 500 are to implement the corresponding process of the method in FIG. 2 respectively. , For the sake of brevity, I won’t repeat it here.

FIG. 6 is a schematic structural diagram of a computing device 600 provided by an embodiment of the present application. The computing device 600 includes a processor 610, a memory 620, a communication interface 630, and a bus 650.

It should be understood that the processor 610 in the computing device 600 shown in FIG. 6 may correspond to the determining module 510 and the processing module 520 of the computing device 500 in FIG. 5, and the communication interface 630 in the computing device 600 may be used to communicate with other devices. To communicate.

Wherein, the processor 610 may be connected to the memory 620. The memory 620 can be used to store the program code and data. Therefore, the memory 620 may be a storage unit inside the processor 610, or an external storage unit independent of the processor 610, or may include a storage unit inside the processor 610 and an external storage unit independent of the processor 610. part.

Optionally, the computing device 600 may further include a bus 650. The memory 620 and the communication interface 630 may be connected to the processor 610 through the bus 650. The bus 650 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus or the like. The bus 650 can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, only one line is used in FIG. 6, but it does not mean that there is only one bus or one type of bus.

It should be understood that, in this embodiment of the present application, the processor 610 may adopt a central processing unit (CPU). The processor can also be other general-purpose processors, digital signal processors (digital signal processors, DSP), application specific integrated circuits (ASICs), ready-made programmable gate arrays (field programmable gate arrays, FPGAs) or other Programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor or the like. Or, the processor 610 adopts one or more integrated circuits to execute related programs to implement the technical solutions provided in the embodiments of the present application.

The memory 620 may include a read-only memory and a random access memory, and provides instructions and data to the processor 610. A part of the processor 610 may also include a non-volatile random access memory. For example, the processor 610 may also store device type information.

When the computing device 600 is running, the processor 610 executes the computer-executable instructions in the memory 620 to execute the operation steps of the foregoing method.

It should be understood that the computing device 600 according to the embodiment of the present application may correspond to the corresponding main body executing the method shown in FIG. 2 according to the embodiment of the present application, and the foregoing and other operations and/or functions of each module in the computing device 600 are respectively In order to realize the corresponding process of the method in FIG. 2, for the sake of brevity, details are not repeated here.

A person of ordinary skill in the art may realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

Those skilled in the art can clearly understand that, for the convenience and conciseness of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device, and method may be implemented in other ways. For example, the device embodiments described above are merely illustrative, for example, the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or It can be integrated into another system, or some features can be ignored or not implemented. In addition, the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.

In addition, the functional units in the various embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.

If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solution of the present application essentially or the part that contributes to the existing technology or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program code .

The above are only specific implementations of this application, but the protection scope of this application is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application. Should be covered within the scope of protection of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

A method for scene recognition, which is characterized in that it includes:

Determining a first state sequence according to the driving data of the vehicle, the first state sequence representing the first state of the vehicle at different times;

Detecting the first state sequence using a recognition rule, the recognition rule being determined according to a target scene;

According to the detection result, it is determined whether the driving scene corresponding to the driving data includes the target scene.
The method according to claim 1, wherein the determining the first state sequence according to the driving data of the vehicle comprises:

The first state sequence is determined according to the driving data of the vehicle and the target scene.
The method according to claim 1 or 2, wherein the determining whether the target scene is included in the driving scene corresponding to the driving data according to the detection result comprises:

If the first state sequence includes a first subsequence, it is determined that the driving scene corresponding to the driving data includes the target scene, and the first subsequence is a subsequence that satisfies the recognition rule; or,

If the first state sequence does not include the first subsequence, it is determined that the driving scene corresponding to the driving data does not include the target scene, and the first subsequence is a subsequence that satisfies the recognition rule.
The method according to claim 2 or 3, wherein the determining the first state sequence according to the driving data of the vehicle and the target scene comprises:

According to the driving data of the vehicle and the target scene, a second state sequence and a third state sequence are determined, the second state sequence represents the second state of the vehicle at different moments, and the third state sequence represents all State the third state of the vehicle at different moments;

The first state sequence is generated according to the second state sequence and the third state sequence.
The method according to any one of claims 1 to 4, wherein the first state sequence is a matrix of size m×n, and the element in the i-th row and j-th column of the matrix represents the index of i The first state of the vehicle at time j, m is an integer greater than or equal to 1, n is an integer greater than or equal to 2, i is an integer greater than or equal to 1, and less than m, j is greater than or equal to 1, and less than n Integer.
The method according to any one of claims 3 to 5, wherein the driving scene corresponding to the driving data includes the target scene, and the method further comprises:

According to the time corresponding to the first sub-sequence, a second sub-sequence is determined in the fourth state sequence, the fourth state sequence represents the associated state of the vehicle, and the fourth state sequence is based on the driving of the vehicle Data confirmed;

According to the second subsequence, the complexity of the target scene is determined.
The method according to claim 6, wherein the determining the complexity of the target scene according to the second subsequence comprises:

Determine the complexity of the associated state of the vehicle according to the second sub-sequence;

Perform a weighted operation on the complexity of the associated state of the vehicle to determine the complexity of the target scene.
The method according to any one of claims 3 to 7, wherein the method further comprises:

According to the time corresponding to the first sub-sequence, a third sub-sequence is determined in the fifth state sequence, the fifth state sequence represents the position information of the vehicle at different moments, and the fifth state sequence is based on the The driving data of the vehicle is determined;

The determining whether the target scene is included in the driving scene corresponding to the driving data according to the detection result includes:

According to the detection result and the third sub-sequence, it is determined whether the driving scene corresponding to the driving data includes the target scene.
A computing device, characterized in that it comprises:

A determining module, configured to determine a first state sequence according to the driving data of the vehicle, the first state sequence representing the first state of the vehicle at different times;

A processing module, configured to detect the first state sequence using a recognition rule, the recognition rule being determined according to a target scene;

The determining module is further configured to determine whether the target scene is included in the driving scene corresponding to the driving data according to the detection result.
The computing device according to claim 9, wherein when the first state sequence is determined according to the driving data of the vehicle, the determining module is specifically configured to:

The first state sequence is determined according to the driving data of the vehicle and the target scene.
The computing device according to claim 9 or 10, wherein when determining whether the driving scene corresponding to the driving data includes the target scene according to the detection result, the determining module is specifically configured to:

If the first state sequence includes a first subsequence, it is determined that the driving scene corresponding to the driving data includes the target scene, and the first subsequence is a subsequence that satisfies the recognition rule; or,

If the first state sequence does not include the first subsequence, it is determined that the driving scene corresponding to the driving data does not include the target scene, and the first subsequence is a subsequence that satisfies the recognition rule.
The computing device according to claim 10 or 11, wherein when the first state sequence is determined according to the driving data of the vehicle and the target scene, the determining module is specifically configured to:

According to the driving data of the vehicle and the target scene, a second state sequence and a third state sequence are determined, the second state sequence represents the second state of the vehicle at different moments, and the third state sequence represents all State the third state of the vehicle at different moments;

The first state sequence is generated according to the second state sequence and the third state sequence.
The computing device according to any one of claims 9 to 12, wherein the first state sequence is a matrix of size m×n, and an element in the i-th row and j-th column of the matrix indicates that the index is i The first state of the vehicle at time j, m is an integer greater than or equal to 1, n is an integer greater than or equal to 2, i is an integer greater than or equal to 1, and less than m, j is greater than or equal to 1, and less than An integer of n.
The computing device according to any one of claims 11 to 13, wherein when the driving scene corresponding to the driving data includes the target scene, the determining module is further configured to:

According to the time corresponding to the first sub-sequence, a second sub-sequence is determined in the fourth state sequence, the fourth state sequence represents the associated state of the vehicle, and the fourth state sequence is based on the driving of the vehicle Data confirmed;

According to the second subsequence, the complexity of the target scene is determined.
The computing device according to claim 14, wherein when determining the complexity of the target scene according to the second subsequence, the determining module is specifically configured to:

Determine the complexity of the associated state of the vehicle according to the second sub-sequence;

Perform a weighted operation on the complexity of the associated state of the vehicle to determine the complexity of the target scene.
The computing device according to any one of claims 11 to 15, wherein the determining module is further configured to:

According to the time corresponding to the first sub-sequence, a third sub-sequence is determined in the fifth state sequence, the fifth state sequence represents the position information of the vehicle at different moments, and the fifth state sequence is based on the The driving data of the vehicle is determined;

When determining whether the driving scene corresponding to the driving data includes the target scene according to the detection result, the determining module is specifically configured to:

According to the detection result and the third sub-sequence, it is determined whether the driving scene corresponding to the driving data includes the target scene.
A computing device, the computing device includes a processor and a memory, the memory is used to store computer execution instructions, when the computing device is running, the processor executes the computer execution instructions in the memory to pass the calculation The device executes the operation steps in the method according to any one of claims 1 to 8.