Method, device and equipment for road state detection and storage medium

Abstract

The embodiment of the invention provides a road state detection method and device, equipment and a storage medium. The method comprises the steps that in response to detection that a vehicle allocated to execute a driving task has an abnormal driving behavior, a road section associated with the abnormal driving behavior is determined, the abnormal driving behavior indicates that the actual driving track of the vehicle is not matched with the state of the road section, and the state comprises a passing state or a closed state; determining feature information associated with the road section at least based on a plurality of historical driving tasks associated with the road section; using a trained machine learning model to determine an index indicating the possibility that an abnormal driving behavior will occur on the road section at least based on the feature information; and determining an adjustment to the state of the road segment based on the indicator.

Description

Technical Field

[0001] Example embodiments of the present disclosure generally relate to the field of computers, and in particular, to methods, devices, apparatuses, computer-readable storage media, and computer program products for road state detection. Background Art

[0002] With the widespread application of electronic navigation technology, drivers are increasingly relying on navigation maps during driving, and at the same time, higher requirements are placed on the accuracy and timeliness of navigation maps. The actual state of the road changes in real time. For example, large-scale events, temporary restrictions, temporary construction, etc. will lead to long-term or short-term road closures. After the event, regulation, or construction is over, the road will resume traffic. If it is not possible to determine in a timely and accurate manner that the road is in a closed section, the vehicle will not be able to pass when it reaches the closed section and will be forced to take a detour. If it is not possible to determine in a timely and accurate manner that the road section has resumed traffic, the vehicle may be guided to take an unnecessary detour. It can be seen that it is of great significance to determine the state of the road in a timely and accurate manner. Summary of the invention

[0003] In a first aspect of the present disclosure, a method for road state detection is provided. The method includes: in response to detecting that a vehicle assigned to perform a driving task has an abnormal driving behavior, determining a road segment associated with the abnormal driving behavior, wherein the abnormal driving behavior indicates that the actual driving trajectory of the vehicle does not match the state of the road segment, and the state includes a traffic state or a closed state; determining feature information associated with the road segment based at least on a plurality of historical driving tasks associated with the road segment; determining an indicator indicating the possibility that the abnormal driving behavior will occur on the road segment based at least on the feature information using a trained machine learning model; and determining an adjustment to the state of the road segment based on the indicator.

[0004] In a second aspect of the present disclosure, a device for road state detection is provided. The device includes: a road segment determination module, configured to determine a road segment associated with abnormal driving behavior in response to detecting abnormal driving behavior of a vehicle assigned to perform a driving task, wherein the abnormal driving behavior indicates that the actual driving trajectory of the vehicle does not match the state of the road segment, and the state includes a traffic state or a closed state; an information determination module, configured to determine feature information associated with the road segment based on at least a plurality of historical driving tasks associated with the road segment; an indicator determination module, configured to determine an indicator indicating the possibility of abnormal driving behavior occurring on the road segment based on at least the feature information using a trained machine learning model; and a state adjustment module, configured to determine the adjustment of the state of the road segment based on the indicator.

[0005] In a third aspect of the present disclosure, an electronic device is provided. The device includes at least one processing unit; and at least one memory, the at least one memory is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit. When the instructions are executed by the at least one processing unit, the electronic device executes the method of the first aspect.

[0006] In a fourth aspect of the present disclosure, a computer-readable storage medium is provided, wherein a computer program is stored on the medium, and when the computer program is executed by a processor, the method of the first aspect is implemented.

[0007] In a fifth aspect of the present disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, the method of the first aspect is implemented.

[0008] It should be understood that the content described in this section is not intended to limit the key features or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements, wherein:

[0010] Figure 1 A schematic diagram showing an example environment in which embodiments of the present disclosure can be implemented;

[0011] Figure 2 A flowchart showing a process for road state detection according to some embodiments of the present disclosure is shown;

[0012] Figure 3 A schematic diagram showing an example scenario for road state detection according to some embodiments of the present disclosure;

[0013] Figure 4 A schematic diagram showing an example architecture for road state detection according to some embodiments of the present disclosure is shown;

[0014] Figure 5 A schematic structural block diagram of an apparatus for road state detection according to some embodiments of the present disclosure is shown; and

[0015] Figure 6 The present invention is a block diagram of an electronic device that can implement one or more embodiments of the present disclosure. DETAILED DESCRIPTION

[0016] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0017] In the description of the embodiments of the present disclosure, the term "including" and similar terms should be understood as open inclusion, that is, "including but not limited to". The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The term "some embodiments" should be understood as "at least some embodiments". Other explicit and implicit definitions may also be included below.

[0018] Herein, unless explicitly stated, executing a step “in response to A” does not mean executing the step immediately after “A” but may include one or more intermediate steps.

[0019] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition, use, storage or deletion of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0020] It is understandable that before using the technical solutions disclosed in the various embodiments of the present disclosure, the types, scopes of use, usage scenarios, etc. of the information involved in the present disclosure should be informed to relevant users and their authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations. The relevant users may include any type of right holders, such as individuals, enterprises, and groups.

[0021] For example, in response to receiving an active request from a user, a prompt message is sent to the relevant user to clearly prompt the relevant user that the operation requested to be performed will require obtaining and using the information of the relevant user, so that the relevant user can independently choose whether to provide information to software or hardware such as an electronic device, application, server or storage medium that executes the operation of the technical solution of the present disclosure based on the prompt message.

[0022] As an optional but non-limiting implementation, in response to receiving an active request from a relevant user, a prompt message is sent to the relevant user, for example, in the form of a pop-up window, in which the prompt message may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide information to the electronic device.

[0023] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0024] As used herein, the term "model" can learn the association between the corresponding input and output from the training data, so that after the training is completed, the corresponding output can be generated for a given input. The generation of the model can be based on machine learning technology. Deep learning is a machine learning algorithm that processes inputs and provides corresponding outputs by using multi-layer processing units. A neural network model is an example of a model based on deep learning. In this article, "model" may also be referred to as "machine learning model", "learning model", "machine learning network" or "learning network", and these terms are used interchangeably in this article.

[0025] A "neural network" is a machine learning network based on deep learning. A neural network is capable of processing inputs and providing corresponding outputs, and typically includes an input layer and an output layer and one or more hidden layers between the input layer and the output layer. Neural networks used in deep learning applications typically include many hidden layers, thereby increasing the depth of the network. The layers of a neural network are connected in sequence so that the output of the previous layer is provided as input to the next layer, where the input layer receives the input of the neural network and the output of the output layer serves as the final output of the neural network. Each layer of a neural network includes one or more nodes (also called processing nodes or neurons), each of which processes input from the previous layer.

[0026] Generally, machine learning can be roughly divided into three stages, namely the training stage, the testing stage, and the application stage (also called the inference stage). In the training stage, a given model can be trained using a large amount of training data, and the parameter values ​​are continuously updated iteratively until the model can obtain consistent inferences that meet the expected goals from the training data. Through training, the model can be considered to be able to learn the association from input to output (also called the mapping of input to output) from the training data. The parameter values ​​of the trained model are determined. In the testing stage, the test input is applied to the trained model to test whether the model can provide the correct output, thereby determining the performance of the model. In the application stage, the model can be used to process the actual input based on the parameter values ​​obtained from the training to determine the corresponding output.

[0027] As mentioned above, with the widespread application of electronic navigation technology, drivers are increasingly dependent on navigation maps during driving, and at the same time, higher requirements are placed on the accuracy and timeliness of navigation maps. The actual state of the road changes in real time. For example, large-scale events, temporary controls, temporary construction, etc. will lead to long-term or short-term road closures. After the event, control, or construction is over, the road will resume traffic. If it is not possible to determine in a timely and accurate manner that the road is in a closed section, the vehicle will not be able to pass when it reaches the closed section and will be forced to take a detour. If it is not possible to determine in a timely and accurate manner that the road section has returned to a passable state, the vehicle may be guided to take an unnecessary detour. It can be seen that it is of great significance to determine the state of the road in a timely and accurate manner.

[0028] In view of this, an embodiment of the present disclosure provides an improved solution for road state detection. In this solution, if a vehicle assigned to perform a driving task is detected to have abnormal driving behavior, a road section associated with the abnormal driving behavior is determined. The abnormal driving behavior here indicates that the actual driving trajectory of the vehicle does not match the state of the road section, and the state of the road section includes an open state or a closed state. Based on at least a plurality of historical driving tasks associated with the road section, characteristic information associated with the road section is determined. Based on the characteristic information, a trained machine learning model is used to determine an indicator indicating the possibility that abnormal driving behavior will occur on the road section. Thereafter, based on the determined indicator, the adjustment of the state of the road section is determined.

[0029] The embodiments of the present disclosure can predict the possibility of abnormal driving behavior on a road section based on abnormal driving behaviors that do not match the state of the road section in historical driving tasks, and then determine whether to adjust the state of the road section. In this way, the accuracy and timeliness of road state detection can be improved, providing a reference for the planning and selection of vehicle driving routes.

[0030] Example Environment

[0031] Figure 1 1 is a schematic diagram of an example environment 100 in which embodiments of the present disclosure can be implemented. Figure 1 As shown, the environment 100 involves a vehicle service platform 110, which can provide a service environment for a vehicle 130 or a terminal device 140, for example, it can provide navigation services, taxi services, carpooling services, freight services, etc. for the vehicle 130.

[0032] In some embodiments, the vehicle service platform 110 may be deployed on the server device 120. For example, a vehicle terminal corresponding to the vehicle service platform 110 may be running on the vehicle 130, and the vehicle service platform 110 may provide services to the vehicle terminal running on the vehicle 130 based on the communication connection between the server device 120 and the vehicle 130. For another example, the terminal device 140 may be running a client of the vehicle service platform 110, and the vehicle service platform 110 may provide services to the client running on the terminal device 140 based on the communication connection between the server device 120 and the terminal device 140. In some embodiments, the vehicle service platform 110 may call the machine learning model 150 to support the provision of services to the vehicle 130 or the terminal device 140 based on the output of the machine learning model 150.

[0033] In some embodiments, the server device 120 may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content distribution networks, and big data and artificial intelligence platforms. The server device 120 may include, for example, a computing system / server, such as a mainframe, an edge computing node, a computing device in a cloud environment, and the like.

[0034] In some embodiments, vehicle 130 may include "car", "vehicle" and "entire vehicle" or other similar terms including general motor vehicles, such as sedans, sports utility vehicles (SUV), multi-purpose vehicles (MPV), buses, trucks and other cargo or passenger vehicles, including hybrid vehicles, electric vehicles, fuel vehicles, plug-in hybrid vehicles, fuel cell vehicles and other alternative fuel vehicles, etc. A hybrid vehicle refers to a vehicle with two or more power sources, and an electric vehicle includes a pure electric vehicle, an extended-range electric vehicle, etc., which is not specifically limited in the embodiments of the present disclosure.

[0035] In some embodiments, the vehicle 130 may be provided with a positioning unit 131, and the vehicle position of the vehicle 130 may be determined by the positioning unit 131. The positioning unit 131 may include, but is not limited to, a satellite positioning system, a cellular network positioning system, a WIFI positioning system, a Bluetooth positioning system, an inertial measurement unit, etc., which is not specifically limited in the embodiments of the present disclosure.

[0036] In some embodiments, the terminal device 140 can be any type of mobile terminal, fixed terminal or portable terminal, including a mobile phone, a desktop computer, a laptop computer, a notebook computer, a netbook computer, a tablet computer, a media computer, a multimedia tablet, a personal communication system (PCS) device, a personal navigation device, a personal digital assistant (PDA), an audio / video player, a digital camera / camcorder, a positioning device, a television receiver, a radio broadcast receiver, an e-book device, a gaming device or any combination of the foregoing, including accessories and peripherals of these devices or any combination thereof.

[0037] It should be understood that the structure and function of the various elements in the environment 100 are described for exemplary purposes only and do not imply any limitation on the scope of the present disclosure.

[0038] Example Process

[0039] Some example embodiments of the present disclosure will be described in detail below with reference to examples of the accompanying drawings.

[0040] Figure 2 A flowchart of a process 200 for road state detection according to some embodiments of the present disclosure is shown. Part or all of the process 200 may be implemented by the vehicle service platform 110 or may be implemented by other devices, such as other remote devices (terminal devices or service devices) with computing capabilities. Figure 1 Some embodiments of the present disclosure are described with reference to the environment 100 in FIG. 1 and from the perspective of the vehicle service platform 110 , but this is merely exemplary.

[0041] In block 210, if abnormal driving behavior is detected in the vehicle 130 assigned to perform the driving task, the vehicle service platform 110 determines the road segment associated with the abnormal driving behavior. The driving task may include a variety of tasks related to the driving of the vehicle 130, such as navigation tasks, taxi tasks, carpooling tasks, freight tasks, etc. The abnormal driving behavior indicates that the actual driving trajectory of the vehicle does not match the state of the road segment, where the state includes a pass state or a closed state, wherein the pass state indicates that the vehicle 130 is allowed to drive on the corresponding road segment, and the closed state indicates that the vehicle 130 is prohibited from driving on the corresponding road segment.

[0042] In some embodiments, a driving task may have an associated planned driving trajectory. If the vehicle service platform 110 detects that the actual driving trajectory of the vehicle 130 deviates from the planned driving trajectory, and determines that the road section from which the vehicle 130 deviates is in a passable state, the vehicle service platform 110 determines that abnormal driving behavior of the vehicle 130 is detected. The vehicle service platform 110 may determine that the road section from which the vehicle 130 deviates and is in a passable state is a road section associated with abnormal driving behavior. In this case, abnormal driving behavior refers to the behavior of deviating from the road section in a passable state.

[0043] In some examples, Figure 3 A schematic diagram of an example scenario 300 for road state detection according to some embodiments of the present disclosure is shown. The example scenario 300 shows a road segment 302, a driving trajectory 304, and a driving trajectory 306. The planned driving trajectory provided by the vehicle service platform 110 for the vehicle 130 for the driving task may be the driving trajectory 304. If it is detected that the actual driving trajectory of the vehicle 130 is the driving trajectory 306, and it is determined that the state of the road segment 302 from which the vehicle 130 deviates is a passable state, the vehicle service platform 110 may determine that the actual driving trajectory of the vehicle 130 does not match the state of the road segment 302. The vehicle service platform 110 may determine that an abnormal driving behavior is detected, and the vehicle service platform 110 may determine that the road segment 302 is a road segment associated with the abnormal driving behavior.

[0044] In some embodiments, if the vehicle service platform 110 detects that the actual driving trajectory of the vehicle 130 deviates from the planned driving trajectory of the driving task, and determines that the road section traversed by the vehicle 130 is in a closed state, the vehicle service platform 110 determines that abnormal driving behavior of the vehicle 130 is detected. The vehicle service platform 110 can determine the closed road section traversed by the vehicle 130 as a road section associated with abnormal driving behavior. In this case, the abnormal driving behavior refers to the behavior of crossing a closed road section.

[0045] In some examples, such as Figure 3 As shown, the planned driving trajectory provided by the vehicle service platform 110 for the vehicle 130 for the driving task may be the driving trajectory 306. If it is detected that the actual driving trajectory of the vehicle 130 is the driving trajectory 304, and it is determined that the road segment 302 traversed by the vehicle 130 is in a closed state, the vehicle service platform 110 may determine that the actual driving trajectory of the vehicle 130 does not match the state of the road segment 302. The vehicle service platform 110 determines that an abnormal driving behavior is detected, and the vehicle service platform 110 may determine that the road segment 302 is a road segment associated with the abnormal driving behavior.

[0046] Figure 4 FIG. 4 is a schematic diagram showing an example architecture 400 for detecting user road status according to some embodiments of the present disclosure. Figure 4 As shown, the vehicle service platform 110 may include an abnormality sensing unit 410, a decision unit 420, a business unit 430, a map database 440, a recording unit 450, and an evaluation unit 460. The vehicle service platform 110 may use the abnormality sensing unit 410 to detect abnormal driving behaviors of the vehicle 130, determine the vehicle having abnormal driving behaviors, and determine the road sections associated with the abnormal driving behaviors.

[0047] Return to reference Figure 2 In box 220, the vehicle service platform 110 determines the characteristic information associated with the road section based on at least a plurality of historical driving tasks associated with the road section. The vehicle service platform 110 usually provides a planned driving trajectory for the driving task according to the state of each road section in the map. If the state of the road section recorded by the vehicle service platform 110 is the same as the actual state of the road section, in most cases the vehicle 130 will drive according to the planned driving trajectory and will not easily have abnormal driving behavior. If the state of the road section recorded by the vehicle service platform 110 is different from the actual state of the road section, the vehicle 130 performing the driving task is prone to abnormal driving behavior. The status of the historical driving tasks can indicate the driving behavior of the vehicle 130 to a certain extent, especially the historical driving tasks with abnormal driving behavior can indicate the abnormal driving behavior of the vehicle 130, and can have a certain correlation with the state of the road section, providing a reference for the detection of the state of the road section.

[0048] In some embodiments, the vehicle service platform 110 may store historical driving tasks and their corresponding road sections. In the case where a road section associated with abnormal driving behavior has been determined, the vehicle service platform 110 may determine multiple historical driving tasks associated with the road section. The vehicle service platform 110 determines feature information for the machine learning model 150 based at least on multiple historical driving tasks. In some examples, the vehicle service platform 110 may select a predetermined number of historical driving tasks before the current time, and the vehicle service platform 110 may perform feature extraction on the task information of the predetermined number (e.g., ten, fifteen, etc.) of historical driving tasks to obtain feature information related to the road section. In other words, the feature information may include information related to historical driving tasks.

[0049] In other examples, the feature information may also include a first occurrence frequency of abnormal driving behavior associated with the road segment. The first occurrence frequency may be a ratio between the number of occurrences of abnormal driving behavior associated with the road segment and the first task amount. The first task amount may indicate the total amount of driving tasks associated with the road segment over a period of time. As an example, when the road segment is in a traffic state, the vehicle service platform 110 may determine the occurrence frequency of the behavior of the road segment that deviates from the traffic state. When the road segment is in a closed state, the vehicle service platform 110 may determine the occurrence frequency of the behavior of crossing the closed state of the road segment.

[0050] It should be understood that the characteristic information is not limited to information related to historical driving tasks, but may also include other information related to road sections or abnormal driving behaviors. In some examples, the characteristic information may also include a second occurrence frequency of abnormal driving behaviors in all or part of the road sections in the vehicle service platform 110. The second occurrence frequency may be the ratio between the abnormal driving behavior detected by the vehicle service platform 110 and the second task amount. The second task amount may indicate the total amount of driving tasks triggered by the vehicle service platform 110 within a period of time. In this way, a reference can be provided for the possibility prediction of abnormal driving behaviors from the macro level of the vehicle service platform 110, which is conducive to improving the accuracy of the prediction of the possibility of abnormal driving behaviors. Of course, the above-mentioned characteristic information is only exemplary. In actual applications, any other appropriate information related to road sections or abnormal driving behaviors may also be selected as characteristic information, and the embodiments of the present disclosure are not limited to this.

[0051] Continue to refer Figure 2 In the process 200 shown, in box 230, the vehicle service platform 110 uses the trained machine learning model 150 to determine an indicator indicating the possibility of abnormal driving behavior on the road section based on at least the feature information. Usually the state of the road section is stable over a period of time. Therefore, within multiple time periods or multiple cycles within a certain time range, the driving behavior of the vehicle 130 on the road section is affected by the state of the road section. In multiple time periods or multiple cycles, the driving behavior of the vehicle 130 on the road section is correlated. On this basis, the machine learning model 150 is trained to predict the possibility of abnormal driving behavior in the next time period (i.e., the future) based on information related to historical driving tasks in the previous time period or the previous cycle. In some embodiments, the machine learning model 150 can be formed by deep learning model training. Of course, the machine learning model 150 can also be formed by other types of model training. The embodiments of the present disclosure do not limit the type of machine learning model 150.

[0052] Regarding the input and output of the machine learning model 150, in some examples, the vehicle service platform 110 provides the feature information to the trained machine learning model 150 and obtains the model output output by the machine learning model 150. The model output of the machine learning model 150 may include an indicator indicating the possibility of abnormal driving behavior on the road segment (hereinafter referred to as "indicator"). The vehicle service platform 110 may determine to indicate the indicator based on the model output. As an example, Figure 4 As shown, the decision unit 420 provides the feature information to the machine learning model 150. The decision unit 420 can then determine an indicator based on the model output of the machine learning model 150.

[0053] In other examples, the vehicle service platform 110 uses a trained machine learning model based on feature information to determine the predicted number of abnormal driving behaviors in future driving tasks of the road section. Afterwards, the vehicle service platform 110 determines the index based on at least the predicted number of abnormal driving behaviors. In some examples, the vehicle service platform 110 can determine the predicted number of abnormal driving behaviors in a predetermined number (e.g., ten or fifteen) of future driving tasks based on the model output. As an example, a reward coefficient can be set in advance for abnormal driving behaviors, for example, a reward coefficient can be set for the behavior of a road section that deviates from a traffic state and the behavior of a road section that crosses a closed state. In the case of determining the predicted number of abnormal driving behaviors in a predetermined number of future driving tasks, the vehicle service platform 110 can use a reward function to determine the index based on a pre-set reward coefficient and the predicted number of abnormal driving behaviors. In this way, the result of the index can be avoided from being affected by the traffic volume of different sections, and it can be ensured that sections with different traffic volumes can form relatively accurate detection results.

[0054] In some embodiments, the vehicle service platform 110 uses a trained machine learning model to determine an indicator indicating the possibility of abnormal driving behavior occurring on a road section based on feature information and user feedback information associated with a plurality of historical driving tasks. The feedback information here may include user complaints, comments, ratings, recommendations, and the like for historical driving tasks. In some examples, a reward coefficient may be set in advance for, for example, user complaints. The vehicle service platform 110 provides the feature information to the machine learning model 150 to obtain the model output of the machine learning model 150. Based on the model output, the predicted number of abnormal driving behaviors occurring in a predetermined number of future driving tasks is determined. The vehicle service platform 110 uses a reward function to determine an indicator based on the number of complaints and the predicted number of abnormal driving behaviors.

[0055] In other examples, the vehicle service platform 110 may also set a reward coefficient for the number of times the decision unit is triggered. The vehicle service platform 110 determines the number of times the decision unit 420 is triggered in a predetermined number of historical driving tasks. In some cases, the number of times the decision unit 420 is triggered may be the same as the number of times abnormal driving behavior occurs in a predetermined number of historical driving tasks. The vehicle service platform 110 may determine the indicator using a reward function based on the predicted number of abnormal driving behaviors, the number of complaints, the number of triggers of the decision unit 420, and the reward coefficients corresponding to the predicted number, the number of complaints, and the number of triggers.

[0056] As an example, an example of a reward coefficient for the number of predictions, the number of feedbacks, and the number of triggers of the decision unit 420 for abnormal driving behaviors is shown below, as shown in the table below.

[0057] Table 1

[0058] project Reward coefficient <![CDATA[Number of predictions of behavior on sections deviating from the passing state (F1)]]> <![CDATA[C1]]> <![CDATA[Number of predictions of behavior for traversing a section in a closed state (F2)]]> <![CDATA[C2]]> <![CDATA[The trigger count of the decision-making unit (F3)]]> <![CDATA[C3]]> <![CDATA[Number of feedbacks (F4)]]> <![CDATA[C4]]>

[0059] Combined with the reward coefficients in the above table, the indicator (I) can be calculated using the formula shown below.

[0060] I=F1×C1+F2×C2+F3×C3+F4×C4 (1)

[0061] It should be understood that the model output of the above-mentioned machine learning model 150 is only exemplary. In actual application, the machine learning model 150 may also be configured to output other data content. For example, the model output of the machine learning model 150 may also include the probability or frequency of abnormal driving behavior in the future. The vehicle service platform 110 may also determine the indicator based on the probability or frequency of abnormal driving behavior in the future. In addition, the above-mentioned method for determining the indicator is only exemplary. In actual application, the indicator may also be determined by any other appropriate method. The embodiments of the present disclosure do not specifically limit this.

[0062] return Figure 2 In the process 200 shown, in box 240, the vehicle service platform 110 determines to adjust the state of the road section based on the indicator. It should be understood that the determination of the adjustment of the state of the road section here may include not adjusting the state of the road section, and may also include adjusting the state of the road section. As described above, the consistency between the state of the road section recorded by the vehicle service platform 110 and the actual state of the road section is directly related to whether the vehicle 130 performing the driving task has abnormal driving behavior. On this basis, the indicator indicating the potential possibility of abnormal driving behavior in the road section can indicate to a certain extent whether the state of the road section recorded by the vehicle service platform 110 is the same as or matches the actual state of the road section, and then it can be determined whether the state of the road needs to be adjusted.

[0063] In some embodiments, an indicator threshold may be set in advance for the indicator. The vehicle service platform 110 may determine whether the indicator exceeds the indicator threshold. If it is determined that the indicator does not exceed the indicator threshold, the vehicle service platform 110 determines not to adjust the state of the road section. If it is determined that the indicator exceeds the indicator threshold, the vehicle service platform 110 determines to adjust the state of the road section. Specifically, if the state of the road section is a passable state, the vehicle service platform 110 may adjust the state of the road section to a closed state. If the state of the road section is a closed state, the vehicle service platform 110 may adjust the state of the road section to a passable state. Of course, the above method for determining whether the state of a road section needs to be adjusted is merely exemplary. In actual applications, any other appropriate method may also be used to determine whether the state of a road section needs to be adjusted based on an indicator, and the embodiments of the present disclosure are not limited to this.

[0064] In some embodiments, if it is determined to adjust the state of the road segment, the vehicle service platform 110 updates the map data associated with the road segment based on the adjusted state of the road segment, so that the vehicle service platform 110 can perform driving path planning according to the updated map data. Figure 4 As shown, the vehicle service platform 110 can utilize the business unit 430 to update the map data in the map database 440 based on the adjusted status of the road segment.

[0065] In other examples, the vehicle service platform 110 can also use the recording unit 450 to record the detected abnormal driving behavior, the road section related to the abnormal driving behavior, the state of the road section before adjustment, the predicted number of abnormal driving behaviors occurring on the road section, the indicators determined by the decision unit 420, the state of the road section after adjustment, and so on.

[0066] In some other examples, the evaluation unit 460 can evaluate the accuracy of the adjustment of the state of the road section by the decision unit 420 based on the predicted number of abnormal driving behaviors in the future driving tasks of the road section determined by the decision unit 420, and the number of occurrences of abnormal driving behaviors related to the road section. As an example, the recording unit 450 can record the predicted number of abnormal driving behaviors in a predetermined number of future driving tasks determined by the decision unit 420. After a period of time, the evaluation unit 460 can determine the number of occurrences of abnormal driving behaviors in the predetermined number of driving tasks related to the road section based on the completed predetermined number of driving tasks. The evaluation unit 460 can evaluate the decision effect of the decision unit 420 based on the predicted number and the number of occurrences.

[0067] The following will exemplify the training process of the machine learning model 150. It should be understood that such a training process can be performed by appropriate training equipment, which may include but is not limited to the vehicle service platform 110.

[0068] In an embodiment of the present disclosure, a training device obtains a plurality of training samples. Each training sample includes a plurality of historical driving tasks of a corresponding sample road in a first cycle and a plurality of historical driving tasks in a second cycle after the first cycle. The first cycle and the second cycle here can be understood as sampling cycles for training samples, and the first cycle and the second cycle can be time cycles or task quantity cycles. For example, a predetermined number of consecutive driving tasks can be regarded as a task quantity cycle.

[0069] In an embodiment of the present disclosure, the training device determines sample feature information associated with the sample road based on multiple historical driving tasks in the first cycle. The number of occurrences of abnormal driving behaviors in multiple historical driving tasks in the second cycle is determined. Afterwards, the training device can train the machine learning model based on the sample feature information and the number of occurrences of abnormal driving behaviors. After the training is completed, the training device can evaluate whether the model output of the machine learning model 150 meets the predetermined evaluation requirements. If it is determined that the model output of the machine learning model 150 meets the predetermined evaluation requirements, it can be determined that the machine learning model training is completed.

[0070] It is understandable that, during the use of the machine learning model, the machine learning model 150 can also be incrementally trained. It should be understood that the above training process is only exemplary, and the training device can select any appropriate method to train the machine learning model 150 according to the type of the machine learning model 150. The embodiments of the present disclosure are not limited to this.

[0071] In summary, the embodiments of the present disclosure can predict the potential for abnormal driving behaviors on a road section based on abnormal driving behaviors that do not match the state of the road section in historical driving tasks, and then determine whether to adjust the state of the road section from a passable state to a closed state, or from a closed state to a passable state. In this way, the accuracy and timeliness of road state detection can be improved, providing a reference for the planning and selection of vehicle driving routes.

[0072] Example devices and equipment

[0073] The embodiments of the present disclosure also provide corresponding devices for implementing the above methods or processes. Figure 5 A schematic structural block diagram of an apparatus 500 for road state detection according to some embodiments of the present disclosure is shown. The apparatus 500 may be implemented in or included in the vehicle service platform 110. Each module / component in the apparatus 500 may be implemented by hardware, software, firmware, or any combination thereof.

[0074] like Figure 5As shown, the device 500 includes: a road segment determination module 510, which is configured to determine a road segment associated with the abnormal driving behavior in response to detecting that a vehicle assigned to perform a driving task has abnormal driving behavior, wherein the abnormal driving behavior indicates that the actual driving trajectory of the vehicle does not match the state of the road segment, and the state includes a pass state or a closed state; an information determination module 520, which is configured to determine feature information associated with the road segment based on at least a plurality of historical driving tasks associated with the road segment; an indicator determination module 530, which is configured to determine an indicator indicating the possibility of abnormal driving behavior about to occur on the road segment based on at least the feature information using a trained machine learning model; and a state adjustment module 540, which is configured to determine the adjustment of the state of the road segment based on the indicator.

[0075] In some embodiments, the road segment determination module 510 is further configured to: in response to the actual driving trajectory of the vehicle deviating from the planned driving trajectory of the driving mission, determine the road segment in a passable state from which the vehicle deviates as a road segment associated with abnormal driving behavior.

[0076] In some embodiments, the road segment determination module 510 is further configured to: in response to the actual driving trajectory of the vehicle deviating from the planned driving trajectory of the driving mission, determine the closed road segment traversed by the vehicle as a road segment associated with abnormal driving behavior.

[0077] In some embodiments, the indicator determination module 530 is further configured to: determine the predicted number of abnormal driving behaviors that will occur in future driving tasks on the road segment based on feature information using a trained machine learning model; and determine an indicator indicating the possibility of abnormal driving behaviors that will occur on the road segment based at least on the predicted number of abnormal driving behaviors.

[0078] In some embodiments, the indicator determination module 530 is further configured to: utilize a trained machine learning model to determine an indicator indicating the possibility that abnormal driving behavior will occur on a road segment based on feature information and user feedback information associated with multiple historical driving tasks.

[0079] In some embodiments, the state adjustment module 540 is further configured to perform at least one of the following operations: in response to the indicator not exceeding the corresponding indicator threshold, determine not to adjust the state of the road segment, or in response to the indicator exceeding the indicator threshold, adjust the state of the road segment from one of the open state and the closed state to the other of the open state and the closed state.

[0080] In some embodiments, the apparatus 500 further includes: an updating module configured to, in response to determining to adjust the state of the road segment, update the map data associated with the road segment based on the adjusted state of the road segment.

[0081] In some embodiments, the device 500 also includes: a training module, configured to obtain multiple training samples, each training sample includes multiple historical driving tasks of the corresponding sample road in a first period and multiple historical driving tasks in a second period after the first period; based on the multiple historical driving tasks in the first period, determine the sample feature information associated with the sample road; determine the number of occurrences of abnormal driving behaviors in the multiple historical driving tasks in the second period; and train the machine learning model based on the sample feature information and the number of occurrences of abnormal driving behaviors.

[0082] The units and / or modules included in the device 500 can be implemented in various ways, including software, hardware, firmware, or any combination thereof. In some embodiments, one or more units and / or modules can be implemented using software and / or firmware, such as machine executable instructions stored on a storage medium. In addition to or as an alternative to machine executable instructions, some or all of the units and / or modules in the device 500 can be implemented at least in part by one or more hardware logic components. As an example and not limitation, exemplary types of hardware logic components that can be used include field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0083] Figure 6 1 shows a block diagram of an electronic device 600 in which one or more embodiments of the present disclosure may be implemented. It should be understood that Figure 6 The electronic device 600 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein. Figure 6 The electronic device 600 shown may include or be implemented as Figure 1 Vehicle service platform 110 or Figure 5 Device 500.

[0084] like Figure 6 As shown, the electronic device 600 is in the form of a general electronic device. The components of the electronic device 600 may include, but are not limited to, one or more processors or processing units 610, a memory 620, a storage device 630, one or more communication units 640, one or more input devices 650, and one or more output devices 660. The processing unit 610 may be an actual or virtual processor and is capable of performing various processes according to a program stored in the memory 620. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to improve the parallel processing capability of the electronic device 600.

[0085] The electronic device 600 typically includes a plurality of computer storage media. Such media may be any accessible media that is accessible to the electronic device 600, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 620 may be a volatile memory (e.g., registers, caches, random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 630 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data and may be accessed within the electronic device 600.

[0086] The electronic device 600 may further include additional removable / non-removable, volatile / non-volatile storage media. Figure 6 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. The memory 620 may include a computer program product 625 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0087] The communication unit 640 implements communication with other electronic devices through a communication medium. Additionally, the functions of the components of the electronic device 600 can be implemented with a single computing cluster or multiple computing machines that can communicate through a communication connection. Therefore, the electronic device 600 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0088] The input device 650 may be one or more input devices, such as a mouse, a keyboard, a tracking ball, etc. The output device 660 may be one or more output devices, such as a display, a speaker, a printer, etc. The electronic device 600 may also communicate with one or more external devices (not shown) through the communication unit 640 as needed, such as a storage device, a display device, etc., communicate with one or more devices that allow a user to interact with the electronic device 600, or communicate with any device that allows the electronic device 600 to communicate with one or more other electronic devices (e.g., a network card, a modem, etc.). Such communication may be performed via an input / output (I / O) interface (not shown).

[0089] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which computer-executable instructions are stored, wherein the computer-executable instructions are executed by a processor to implement the method described above. According to an exemplary implementation of the present disclosure, a computer program product is also provided, which is tangibly stored on a non-transitory computer-readable medium and includes computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method described above.

[0090] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices, equipment, and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0091] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0092] Computer-readable program instructions can be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, so that the instructions executed on the computer, other programmable data processing apparatus, or other device implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0093] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0094] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the various implementations disclosed herein.

Claims

1. A method for road state detection, comprising: In response to detecting abnormal driving behavior of a vehicle assigned to perform a driving task, determining a road segment associated with the abnormal driving behavior, wherein the abnormal driving behavior indicates that an actual driving trajectory of the vehicle does not match a state of the road segment, wherein the state includes a pass state or a closed state; Determining characteristic information associated with the road segment based at least on a plurality of historical driving tasks associated with the road segment; Determining, using the trained machine learning model based at least on the feature information, an indicator indicating a likelihood that abnormal driving behavior will occur on the road segment; and Based on the indicator, an adjustment to the status of the road segment is determined.

2. The method according to claim 1, wherein determining the road segment associated with the abnormal driving behavior comprises: In response to the actual driving trajectory of the vehicle deviating from the planned driving trajectory of the driving task, the road section in a trafficable state from which the vehicle deviates is determined as a road section associated with the abnormal driving behavior.

3. The method according to claim 1, wherein determining the road segment associated with the abnormal driving behavior comprises: In response to the actual driving trajectory of the vehicle deviating from the planned driving trajectory of the driving task, a road section in a closed state traversed by the vehicle is determined as a road section associated with the abnormal driving behavior.

4. The method of claim 1, wherein determining an indicator indicating a likelihood that abnormal driving behavior will occur on the road segment comprises: Determine, using the trained machine learning model based on the feature information, a predicted number of abnormal driving behaviors that will occur in future driving tasks on the road segment; as well as Based at least on the predicted number of abnormal driving behaviors, an indicator is determined that indicates the likelihood that abnormal driving behaviors will occur on the road segment.

5. The method of claim 1, wherein determining an indicator indicating a likelihood that abnormal driving behavior will occur on the road segment comprises: An indicator indicating the possibility that abnormal driving behavior will occur on the road segment is determined using a trained machine learning model based on the feature information and user feedback information associated with the multiple historical driving tasks.

6. The method of claim 1, wherein determining the adjustment to the state of the road segment comprises at least one of the following: In response to the indicator not exceeding the corresponding indicator threshold, determining not to adjust the state of the road segment, or In response to the indicator exceeding the indicator threshold, the state of the road segment is adjusted from one of the passable state and the closed state to the other of the passable state and the closed state.

7. The method according to claim 1, further comprising: In response to determining to adjust the status of the road segment, map data associated with the road segment is updated based on the adjusted status of the road segment.

8. The method according to claim 1, wherein the training of the machine learning model comprises: Acquire a plurality of training samples, each training sample including a plurality of historical driving tasks of a corresponding sample road in a first cycle and a plurality of historical driving tasks in a second cycle after the first cycle; Determining sample feature information associated with the sample road based on a plurality of historical driving tasks in the first cycle; Determining the number of occurrences of abnormal driving behaviors in a plurality of historical driving tasks in the second cycle; as well as The machine learning model is trained based on the sample feature information and the number of occurrences of the abnormal driving behavior.

9. A device for road state detection, comprising: a road segment determination module, configured to, in response to detecting abnormal driving behavior of a vehicle assigned to perform a driving task, determine a road segment associated with the abnormal driving behavior, wherein the abnormal driving behavior indicates that an actual driving trajectory of the vehicle does not match a state of the road segment, wherein the state includes a pass state or a closed state; an information determination module, configured to determine characteristic information associated with the road section based at least on a plurality of historical driving tasks associated with the road section; an indicator determination module configured to determine an indicator indicating a likelihood that abnormal driving behavior will occur on the road segment based at least on the feature information using a trained machine learning model; as well as The state adjustment module is configured to determine the adjustment of the state of the road segment based on the indicator.

10. An electronic device, comprising: at least one processing unit; as well as At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions causing the electronic device to perform the method according to any one of claims 1 to 8 when executed by the at least one processing unit.

11. A computer-readable storage medium having a computer program stored thereon, wherein the computer program can be executed by a processor to implement the method according to any one of claims 1 to 8.

12. A computer program product, comprising a computer program, wherein the computer program implements the method according to any one of claims 1 to 8 when executed by a processor.

Images