Method and apparatus for identifying special road conditions, electronic device, and storage medium

Abstract

A method and apparatus, an electronic device, and a storage medium for recognizing special road conditions are provided. The method comprises: a step of recognizing a road area having special road conditions in real time based on big data of vehicle parameters through a pre-trained recognition model for special conditions, based on large-scale vehicle parameters uploaded in real time to a server by large-scale vehicles, and further recognizing a scenario type of special road condition through image features and feature modules of special road condition scenarios based on images or video frames containing the content of special road conditions uploaded in real time by vehicles; and a step of marking special road condition information, such as the feature type and scenario type of special road condition, and the duration of the special road condition of the scenario type obtained through statistical analysis of big data on map data, to support navigation route planning based on map data, reminders of approaching special road conditions, and driving decision-making by vehicles to pass through special road conditions. This ensures the real-time nature and accuracy of special road condition recognition and effectively improves the efficiency and safety of vehicle driving.

Description

Technology Field

[0001] The embodiments of the present application relate to Internet of Vehicles technologies, and in particular, to a method and apparatus for recognizing special road conditions, an electronic device, and a storage medium. Background Technology

[0002] With economic development, the number of vehicles in operation gradually increases. During driving, vehicles may encounter various special road conditions, such as potholes, manholes with loose or missing lids, speed bumps, and lane restrictions. Drivers are sometimes unable to avoid these special road conditions, which can impair the driving experience or even cause danger. Therefore, it is particularly important to have a method to effectively remind vehicles to safely pass through road sections with special road conditions.

[0003] In the prior art, a vehicle may collect images of the surroundings of the vehicle during driving, compare the collected images with preset images of special road conditions, and remind the driver when it recognizes that the images include special road conditions; or a corresponding driving command may be transmitted to the autonomous vehicle based on special road conditions. For example, if there is a pothole in front of the vehicle, the driving command may control the vehicle to decelerate, turn right into the right lane, etc.

[0004] However, conventional technology suffers from poor real-time performance. For example, when an image with a special road condition is recognized, the special road condition cannot be recognized in a timely manner because a vehicle may have passed through the special road condition at high speed.

[0005] Embodiments of the present application provide a method and apparatus for recognizing a special road condition, an electronic device, and a storage medium, wherein an area having a special road condition can be pre-acquired based on map data at the current moment, and high real-time performance.

[0006] According to a first embodiment, a method for recognizing a special road condition is provided. The method may be applied to a terminal device or to a chip within the terminal device. The method is described below using an example in which the method is applied to a terminal device. The method comprises: a terminal device obtaining map data at a current moment from a server, and the map data comprises: a first road area at a current moment, the first road area being a road area where a special road condition is located; the first road area is obtained based on a road condition model and vehicle parameters within a preset period prior to the current moment; the road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition, that is, the road condition model can recognize whether a special road condition exists in a road segment area based on the vehicle parameters of the road segment area within a set period. The terminal device determines whether a second road area exists in the vehicle's planned route based on the vehicle's planned route, and the second road area is a road area within the first road area. It should be understood that the terminal device may periodically report vehicle parameters to the server, and that vehicle parameters include dynamic parameters and static parameters of the vehicle. The vehicle's dynamic parameters include the vehicle's position, images or videos captured by the vehicle, and the vehicle's driving data. The vehicle's driving data may be speed, acceleration, and driving actions such as steering and braking. The vehicle's static parameters are the vehicle's attribute data. The vehicle's attribute data may include data such as weight, length, width, and height, and the vehicle's shock absorption. It should be noted that the vehicle reports its dynamic and static parameters when it first reports vehicle parameters, and reports dynamic parameters when it subsequently reports vehicle parameters.

[0007] It may be understood that the terminal device in this embodiment of the present application can acquire map data at the current moment in real time from a server. Unlike conventional map data, the map data at the current moment includes a first road area having a special road condition, and the first road area is acquired by the server based on reported vehicle parameters, acquired by a road condition model, and marked on a map. Since the reported vehicle parameters are large-scale data reported by multiple vehicles in real time, the accuracy and real-time performance of acquiring the first road area can be improved. Additionally, the terminal device can acquire up-to-date information representing the special road condition for the first road area on the road in real time through the server.

[0008] In a possible design, the terminal device can perform a vehicle driving application based on map data at the current moment.

[0009] According to the modality, during the driving process of a vehicle, it can be determined whether a second road area having a special road condition exists on the vehicle's planned route based on map data at the current moment. A scenario in which a special road condition exists on the vehicle's planned route is described in the following two possible ways.

[0010] One scenario: The vehicle is an autonomous vehicle, and a second road area exists on the vehicle's planned path. Since the autonomous vehicle is not driven by a user, in this embodiment of the application, a driving command (driving decision) may be generated to control the driving behavior of the vehicle. When the vehicle travels into the second road area, the vehicle's driving is controlled based on the driving command. Alternatively, when the vehicle travels to a preset distance from the second road area, the vehicle's driving may be controlled based on the driving command to effectively guide the driving action decision.

[0011] In this embodiment of the present application, a driving command may be generated based on map data at the current moment. The map data further includes descriptive information regarding special road conditions in a first road area, and the descriptive information is used to describe the scenario type of the special road condition. Optionally, there are multiple road condition models. In this embodiment of the present application, a driving command may be generated based on descriptive information regarding special road conditions in a second road area. Specifically, a driving behavior indicated by the characteristics of historical vehicle parameters used to obtain a target road condition model may be used as a driving command, and the target road condition model is a model for determining the second road area as a special road condition.

[0012] Another scenario: The vehicle is a non-autonomous vehicle, and a second road area exists on the vehicle's planned route. As the non-autonomous vehicle is driven by a user, in this embodiment of the application, a reminder message may be generated to remind the user that a special road condition exists in the second road area, so that when the vehicle attempts to drive into the second road area, the reminder message is pushed to remind the user of the special road condition in advance.

[0013] In this embodiment of the present application, driving commands may be generated based on map data at the current moment. The map data at the current moment further includes descriptive information regarding a special road condition in a first road area and / or a target image of the special road condition. The descriptive information is used to describe the scenario type of the special road condition. The target image of the first road area is an image containing the special road condition in the first road area in the vehicle parameters.

[0014] Optionally, in this embodiment of the present application, descriptive information regarding special road conditions in a second road area and / or a target image of the special road conditions are used as a reminder message. Specifically, when a vehicle attempts to drive into the second road area, descriptive information regarding special road conditions in the second road area is played, and / or a target image of the special road conditions in the second road area is displayed.

[0015] According to another aspect, in this embodiment of the present application, a pre-set route may be further planned for a vehicle based on map data at the current moment. Specifically, a terminal device may transmit a route planning request to a server, and the server obtains the planned route and then transmits the planned route to the terminal device. When designing the planned route, the server may avoid planning a route that includes special road conditions and may avoid driving on special road conditions that require a significant amount of time for the vehicle to pass through. This improves passage efficiency and user experience.

[0016] In a possible design, an identifier for a special road condition in each first road area is displayed on a map of the terminal device. A selection command for an identifier for a special road condition in any first road area is received from the user. Descriptive information regarding the special road condition in the first road area selected by the user is displayed.

[0017] In this design, users can view information regarding any specific road conditions on the map, allowing them to actively select their driving route. This enhances the user experience.

[0018] According to a second embodiment, a method for recognizing special road conditions is provided. This method is applied to a server and includes the following: The server obtains a first road area at the current moment based on a road condition model and vehicle parameters within a preset period prior to the current moment. The server obtains map data at the current moment by marking the first road area on the map data or by removing the expired first road area from the map data. The first road area is a road area where a special road condition is located. The road condition model is used to represent the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0019] In this embodiment of the present application, the server may periodically update map data based on reported vehicle parameters to ensure real-timeness of the first road area on the map. In this way, the terminal device can acquire the latest map data in real time to enable prior recognition and reminder of specific road conditions.

[0020] In a possible design, to reduce the data throughput of the server, in this embodiment of the present application, vehicle parameters of at least one vehicle received within a preset period may be preprocessed to determine an initial road area. Then, vehicle parameters corresponding to the initial road area are used as vehicle parameters within a preset period prior to the current moment. In this embodiment of the present application, the area where the location of the vehicle parameters is located that does not match the road features at the vehicle's location in the map data may be determined as the initial road area. Accordingly, vehicle parameters that do not match the road features at the vehicle's location in the map data are used as vehicle parameters within a preset period prior to the current moment.

[0021] It should be noted that the road condition model in this embodiment of the present application is obtained using a plurality of historical vehicle parameters as training parameters. The historical vehicle parameters are vehicle parameters received from at least one vehicle prior to a preset period.

[0022] In a possible design, vehicle parameters reported by different vehicles differ when the vehicles pass through road areas where different special road conditions are located. Accordingly, in this embodiment of the application, different road condition models are trained for different special road conditions to improve the accuracy of recognizing special road conditions. Accordingly, in this embodiment of the application, there are multiple road condition models.

[0023] When a plurality of road condition models are trained, in this embodiment of the present application, a plurality of historical vehicle parameters may be divided into N training data sets. Each training data set is used as training data to train a road condition model, thereby obtaining at least one road condition model. The vehicle parameters within each training data set have the same characteristics, and N is an integer greater than 1. Correspondingly, vehicle parameters within a preset period prior to the current moment are input into at least one road condition model to obtain a first road area.

[0024] In the method described above, the feature type of a first road area having a special road condition can be obtained based on at least one road condition model and vehicle parameters within a preset period prior to the current moment, but the scenario type of the special road condition in the first road area cannot be determined. In a possible design, the scenario type of the special road condition in the first road area can be determined by referring to an image or video captured by a vehicle in the reported vehicle parameters.

[0025] A target image of the first road area may be obtained. The target image of the first road area is an image containing special road conditions in the first road area in vehicle parameters. The target image is an image or video frame from a video captured by a vehicle. In this embodiment of the application, descriptive information regarding special road conditions in the first road area is generated based on the target image of the first road area. The descriptive information is used to describe the scenario type of the special road condition. Specifically, the method of generating descriptive information regarding special road conditions in the first road area may be as follows: In this embodiment of the application, the server may store a recognition model in advance. The recognition model is used to express the correspondence between the features of the image and the scenario type of the special road condition, that is, to input the image into the recognition model. The recognition model may determine the scenario type of the special road condition by recognizing whether the image is an image containing pixel blocks of the special road condition. Additionally, in this embodiment of the application, descriptive information regarding special road conditions in the first road area and / or the target image of the special road condition may be added to the map data.

[0026] In a possible design, in this embodiment of the present application, the duration of a special road condition in a first road area is determined based on a scenario type of a special road condition represented by descriptive information regarding a special road condition in a first road area. The duration of the special road condition in the first road area is added to map data.

[0027] In the two designs described above, in this embodiment of the present application, a target image including a special road condition is determined based on an image or video in the vehicle parameters, descriptive information regarding the special road condition can be generated, and the duration of the special road condition can be further determined based on the scenario type of the special road condition. The information is added to the map data. In this way, after acquiring the map data at the current moment, the terminal device can generate a driving decision or reminder message, or plan a preset route for the vehicle.

[0028] In accordance with the first embodiment, when a terminal device requests a server to obtain a planned route of a vehicle, the server receives a route planning request from the terminal device, obtains a planned route based on a start point and an end point, the duration of a special road condition in a first road area, and a scenario type of a special road condition in the first road area, and can push the planned route to the terminal device. The route planning request includes a start point and an end point.

[0029] According to a third embodiment, a device for recognizing special road conditions is provided, and the device is:

[0030] It includes a processing module, wherein the processing module is configured to: acquire map data at the current moment - the map data includes a first road area at the current moment -; determine, based on the vehicle's planned route whether a second road area exists on the vehicle's planned route, wherein the second road area is a road area within the first road area, and the first road area is a road area where a special road condition is located, and the first road area is acquired based on a road condition model and vehicle parameters within a preset period prior to the current moment, and the road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0031] Optionally, if the vehicle is an autonomous vehicle, the processing module is further configured to: generate a driving command if a second road area exists on the vehicle's planned path; and, when the vehicle travels into the second road area, to control the vehicle's driving based on the driving command, wherein the driving command is used to direct the vehicle's driving behavior.

[0032] Optionally, there are multiple road condition models, and the map data additionally includes descriptive information regarding special road conditions in the first road area, and the descriptive information is used to describe the scenario type of the special road condition.

[0033] Accordingly, the processing module is specifically configured to generate driving commands based on descriptive information regarding special road conditions in the second road area. The driving behavior indicated by the driving command is identical to the driving behavior indicated by the characteristics of the historical vehicle parameters used to acquire the target road condition model. The target road condition model is a model for determining the second road area as a special road condition.

[0034] Optionally, if the vehicle is a non-autonomous vehicle, the processing module is further configured to: generate a reminder message when a second road area exists on the vehicle's planned route; and push a reminder message when the vehicle attempts to drive into the second road area, the reminder message being used to indicate that a special road condition exists in the second road area.

[0035] Optionally, the map data additionally includes descriptive information regarding special road conditions in a first road area and / or a target image of the special road conditions. The descriptive information is used to describe the scenario type of the special road conditions. The target image of the first road area is an image containing the special road conditions in the first road area in the vehicle parameters.

[0036] In response to this, the processing module is specifically configured to use descriptive information regarding special road conditions in the second road area and / or a target image of the special road conditions as a reminder message.

[0037] The playback module is configured to play descriptive information regarding special road conditions in the second road area; and / or,

[0038] The display module is configured to display a target image of a special road condition in the second road area.

[0039] Optionally, the map data additionally includes the duration of special road conditions in the first road area.

[0040] Optionally, there are multiple first road areas. The display module is further configured to display identifiers of special road conditions in each first road area on the map.

[0041] A transceiver module is configured to receive a selection command from a user for an identifier of a special road condition in any first road area. Correspondingly, a display module is further configured to display descriptive information regarding the special road condition in the first road area selected by the user.

[0042] Optionally, the transceiver module is further configured to send a route planning request to the server and receive the planned route sent by the server.

[0043] Optionally, the transceiver module is further configured to report vehicle parameters to the server. Vehicle parameters include the vehicle's location, images or videos captured by the vehicle, and vehicle attribute data and driving data.

[0044] For the beneficial effects of the device for recognizing special road conditions provided in the third embodiment, refer to the beneficial effects caused by the first embodiment and possible designs. Details are not described again herein.

[0045] According to a fourth embodiment, a device for recognizing special road conditions is provided, and the device is:

[0046] It includes a processing module, wherein the processing module is configured to: acquire a first road area at the current moment based on a road condition model and vehicle parameters within a preset period prior to the current moment; and acquire map data at the current moment by marking the first road area on the map data. The first road area is a road area where a special road condition is located. The road condition model is used to represent the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0047] Optionally, vehicle parameters include the vehicle's position.

[0048] A transceiver module is configured to receive vehicle parameters reported by at least one vehicle within a preset period.

[0049] In response to this, the processing module is further configured to determine vehicle parameters within a preset period prior to the current moment based on vehicle parameters of at least one vehicle. The vehicle parameters within the preset period prior to the current moment are vehicle parameters that do not match the road features at the vehicle's location in the map data.

[0050] Optionally, the processing module is further configured to acquire a road state model using multiple historical vehicle parameters as training parameters. The historical vehicle parameters are vehicle parameters received from at least one vehicle prior to a preset period.

[0051] Optionally, there are multiple road condition models.

[0052] The processing module is specifically configured to divide a plurality of historical vehicle parameters into N training data sets and use each training data set as training data for training a road state model to obtain at least one road state model. The vehicle parameters within each training data set have the same characteristics, and N is an integer greater than 1.

[0053] Optionally, the processing module is specifically configured to acquire a first road area by inputting vehicle parameters within a preset period prior to the current moment into at least one road state model.

[0054] Optionally, vehicle parameters include images or videos captured by the vehicle.

[0055] The processing module is further configured to: acquire a target image of a first road area; generate descriptive information regarding a special road condition in the first road area based on the target image of the first road area; and add the descriptive information regarding the special road condition in the first road area and / or the target image of the special road condition to map data. The target image of the first road area is an image containing the special road condition in the first road area in vehicle parameters. The target image is an image or video frame from a video captured by a vehicle. The descriptive information is used to describe the scenario type of the special road condition.

[0056] Optionally, the processing module is further configured to: determine the duration of a special road condition in a first road area based on a scenario type of a special road condition represented by descriptive information regarding a special road condition in a first road area; and add the duration of the special road condition in the first road area to map data.

[0057] Optionally, the processing module is further configured to obtain a planned route based on a start point and an end point, the duration of a special road condition in a first road area, and a scenario type of a special road condition in the first road area when a route planning request is received from a terminal device. The route planning request includes a start point and an end point.

[0058] The transceiver module is additionally configured to push the planned path to the terminal device.

[0059] Optionally, vehicle parameters include vehicle attribute data and driving data.

[0060] For the beneficial effects of the device for recognizing special road conditions provided in the fourth embodiment, refer to the beneficial effects caused by the second embodiment and possible designs. Details are not described again herein.

[0061] According to a fifth embodiment, an electronic device comprising a processor, memory, and a transceiver is provided. The transceiver is coupled to the processor. The processor controls the transmission and reception actions of the transceiver. The processor performs actions performed by a processing module according to a third embodiment or a fourth embodiment. The transceiver performs actions performed by a transceiver module according to a third embodiment or a fourth embodiment.

[0062] Memory is configured to store computer-executable program code. The program code includes instructions. When the processor executes the instructions, the instructions cause the terminal device to perform a method according to a first embodiment or a second embodiment.

[0063] According to the sixth embodiment, a computer program product comprising instructions is provided. When the instructions are executed on a computer, the computer performs a method according to the first embodiment or the second embodiment.

[0064] According to the seventh embodiment, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions. When the instructions are executed on a computer, the computer performs a method according to the first embodiment or the second embodiment.

[0065] Embodiments of the present application provide a method and apparatus for recognizing a special road condition, an electronic device, and a storage medium, wherein a server can determine a first road area where the special road condition is located based on large-scale reported vehicle parameters, update map data, and acquire map data at the current moment. It should be understood that the map data at the current moment includes the first road area. After acquiring the map data at the current moment, the terminal device can recognize whether the special road condition exists within the vehicle's planned route and thus recognize the special road condition within the planned route in advance. This improves the real-time performance of recognizing the special road condition. Brief explanation of the drawing

[0066] FIG. 1 is a schematic diagram of a scenario in which a method for recognizing special road conditions according to an embodiment of the present application is applicable. FIG. 2 is a schematic flowchart 1 of a method for recognizing special road conditions according to an embodiment of the present application. FIG. 3 is a schematic diagram 1 of a map according to an embodiment of the present application. FIG. 4 is a schematic diagram 2 of a map according to an embodiment of the present application. FIGS. 5a and 5b are schematic flowchart 2 of a method for recognizing special road conditions according to an embodiment of the present application. FIG. 6 is a schematic flowchart for obtaining a road condition model according to an embodiment of the present application. FIG. 7 is a schematic diagram 1 of an interface change of a terminal device according to an embodiment of the present application. FIG. 8 is a schematic diagram 2 of an interface change of a terminal device according to an embodiment of the present application. FIG. 9 is a schematic diagram 1 of the structure of a device for recognizing special road conditions according to an embodiment of the present application. FIG. 10 is a schematic diagram 2 of the structure of a device for recognizing special road conditions according to an embodiment of the present application. FIG. 11 is a schematic diagram 1 of the structure of an electronic device according to an embodiment of the present application. FIG. 12 is a schematic diagram 2 of the structure of an electronic device according to an embodiment of the present application. Specific details for implementing the invention

[0067] Currently, during driving, vehicles identify specific road conditions primarily based on navigation prompts, recognition by onboard auxiliary driving sensors, or user perception. However, conventional navigation systems can only prompt for specific road conditions such as road construction and traffic congestion, but cannot prompt for conditions such as missing manhole covers, uneven manhole covers, potholes or potholes, road water accumulation, speed bumps, temporary closures of lanes or road sections, and traffic accidents. Therefore, the types of specific road conditions prompted by navigation are not comprehensive. Furthermore, navigation cannot provide timely prompts for unexpected specific road conditions, such as unexpected obstacles in lanes, car accidents, or debris flows. Consequently, real-time performance is poor. In the case of a method for recognizing special road conditions through an assisted driving sensor, the assisted driving sensor identifies special road conditions where obstacles exist on the road via laser radar, but cannot recognize special road conditions such as potholes or pits, missing manhole covers, and road flooding. Therefore, the types of special road conditions recognized by the assisted driving sensor are not comprehensive. The user needs to rely on their driving experience to recognize special road conditions and cannot recognize or avoid special road conditions due to inattention or slow response. It should be understood that special road conditions in this embodiment of the present application may include: missing manhole covers, uneven manhole covers, potholes or pits on the road, road flooding, speed bumps, lane closures for construction, temporary lane or road section closures, traffic accidents, traffic congestion, etc.

[0068] To address the aforementioned problems, the prior art provides a method for recognizing special road conditions using images collected during a vehicle's driving process. The vehicle recognizes images collected during driving by comparing them with images of various preset special road conditions, and if it determines that the collected images contain special road conditions, it reminds the user. However, this method suffers from the problem of low real-time performance. Due to the high driving speed of the vehicle, by the time an image is collected and a special road condition is recognized in the image, the vehicle may have passed through the special road condition at high speed, and the reminder cannot be provided in a timely manner. Furthermore, the method relies on a single image collected by the vehicle to recognize special road conditions. This can also lead to the problem of low recognition accuracy.

[0069] To address the aforementioned problem of low accuracy in recognizing special road conditions, in the prior art, a server can receive an image of a special road condition and the location of the special road condition uploaded by a user. In this way, vehicles passing through the location of the special road condition are reminded. However, the method still suffers from low real-time performance, and if the user does not upload an image of the special road condition, the server cannot remind the user of the special road condition.

[0070] In recent years, with the advancement of vehicle Internet of Things (VIT) and communication technologies, vehicles can report vehicle parameters to a vehicle Internet of Things (VIT) cloud server in real time. Based on this, to solve the problems of the prior art, an embodiment of the present application provides a method for recognizing special road conditions. Various types of special road conditions are recognized using large-scale vehicle parameters reported in real time by vehicles, marked on a map, and reminded to the driving vehicle in real time. In this embodiment of the present application, since special road conditions are recognized in real time using vehicle parameters reported by large-scale vehicles during driving, the accuracy and real-time performance of recognizing special road conditions can be guaranteed. This supports real-time reminders.

[0071] FIG. 1 is a schematic diagram of a scenario in which a method for recognizing special road conditions according to an embodiment of the present application is applicable. As illustrated in FIG. 1, the scenario includes a terminal device and a server. The terminal device is connected to the server wirelessly. The server may be a vehicle internet cloud server. The terminal device may be a vehicle, an in-vehicle terminal inside the vehicle, etc.

[0072] Hereinafter, a method for recognizing a specific road condition provided in this application is described with reference to specific embodiments. Some of the following embodiments may be combined with one another, and the same or similar concepts or processes may not be described again in some embodiments.

[0073] In FIG. 2, the method for recognizing a special road condition provided in this embodiment of the present application is described in terms of the interaction between a server and a terminal device. FIG. 2 is a schematic flowchart 1 of the method for recognizing a special road condition according to an embodiment of the present application. As illustrated in FIG. 2, the method for recognizing a special road condition provided in this embodiment of the present application may include the following steps.

[0074] S201: The server obtains a first road area at the current moment based on a road condition model and vehicle parameters within a preset period prior to the current moment, the first road area is a road area where a special road condition is located, and the road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0075] S202: The server marks the first road area on the map data and obtains the map data at the current moment.

[0076] S203: The terminal device acquires map data at the current moment.

[0077] S204: The terminal device determines whether a second road area exists on the vehicle's planned route based on the vehicle's planned route, and the second road area is a road area within the first road area.

[0078] In step S201, the road condition model is used to express the correspondence between the characteristics of vehicle parameters and special road conditions, that is, to input vehicle parameters into the road condition model. The road condition model can recognize whether special road conditions exist in a road segment area based on the vehicle parameters of the road segment area within a certain period. Optionally, the road condition model in this embodiment of the present application is obtained based on vehicle parameters historically acquired through a clustering algorithm or an AI algorithm such as machine learning. The AI ​​algorithm may be an algorithm such as a decision tree algorithm, a random forest algorithm, a logistic regression algorithm, a support vector machine algorithm, a naive Bayes algorithm, or a neural network algorithm. The type of AI algorithm used to obtain the road condition model is not limited in this embodiment of the present application.

[0079] The preset period prior to the current moment is a predefined period. In this embodiment of the present application, to ensure real-time performance in acquiring special road conditions, vehicle parameters within the preset period prior to the current moment may be used periodically to acquire the road area where the special road condition is located. For example, if the current moment is 8:00, the vehicle parameters within the preset period prior to the current moment may be vehicle parameters within the period from 7:50 to 8:00. Vehicle parameters within the period from 7:50 to 8:00 are input into the road condition model so that the road area where the special road condition is located at 8:00 can be acquired.

[0080] In this embodiment of the present application, vehicle parameters within a preset period prior to the present moment may be vehicle parameters reported to a server by the vehicle within a preset period prior to the present moment. It should be understood that the vehicle may periodically report vehicle parameters during the course of driving. Vehicle parameters include dynamic parameters and static parameters of the vehicle. Dynamic parameters of the vehicle include the vehicle's location, images or videos captured by the vehicle, and driving data of the vehicle. Driving data of the vehicle may be speed, acceleration, and driving actions such as steering and braking of the vehicle. Static parameters of the vehicle are attribute data of the vehicle. Attribute data of the vehicle may include data such as weight, length, width, and height, and shock absorption of the vehicle. It should be noted that the vehicle reports the vehicle's dynamic and static parameters when it first reports vehicle parameters, and reports the dynamic parameters when it subsequently reports vehicle parameters.

[0081] In this embodiment of the present application, the road area where the special road condition is located at the current moment is used as the first road area. The first road area is a lane-level road area. For example, if the special road condition is traffic congestion, the first road area may be three lanes between XX Road and XX Road. If the special road condition is missing manhole cover, the first road area may be the area of ​​the left lane on XX Road.

[0082] It should be understood that "vehicle parameters within a preset period prior to the current moment" may be vehicle parameters reported by at least one vehicle within the preset period.

[0083] Optionally, to reduce the data throughput of the road condition model, in this embodiment of the application, vehicle parameters of at least one vehicle received within a preset period may be preprocessed to determine an initial road area. Then, vehicle parameters corresponding to the initial road area are used as vehicle parameters within a preset period prior to the current moment. The initial road area is a preset road area where a special road condition is located, and vehicle parameters corresponding to the initial road area are vehicle parameters of a vehicle whose location is in the initial road area among the vehicle parameters of at least one vehicle received within a preset period. For example, based on vehicle parameters of N vehicles received within a preset period, Area 1 is determined as the initial road area, and then vehicle parameters of a vehicle whose location is in Area 1 are obtained from the vehicle parameters of N vehicles and are used as vehicle parameters within a preset period prior to the current moment, i.e., vehicle parameters input to the road condition model.

[0084] In this embodiment of the present application, an area where a location in vehicle parameters does not match the road features at the vehicle's location in the map data may be determined as an initial road area. Correspondingly, vehicle parameters that do not match the road features at the vehicle's location in the map data are used as vehicle parameters within a preset period prior to the current moment. For example, if the road features of Area 1 in the map data are a straight lane with a minimum vehicle speed of 60 km / h, and based on vehicle parameters, the vehicle speed of the vehicle parameters reported by the vehicle in Area 1 is determined to be 10 km / h, 0 km / h, etc., then the vehicle parameters of Area 1 may be determined not to match the road features of Area 1, and special road conditions such as traffic congestion may exist in Area 1. In this case, Area 1 may be used as an initial road area, and the vehicle parameters of a vehicle located in Area 1 are used as vehicle parameters within a preset period prior to the current moment.

[0085] In step S202, in this embodiment of the present application, a first road area may be marked on map data, and map data is updated to obtain map data at the current moment. It should be understood that marking a first road area on map data may involve marking a special road condition at a location corresponding to the first road area on the map, thereby enabling the first road area to be marked on the updated map data at the current moment.

[0086] FIG. 3 is a schematic diagram 1 of a map according to an embodiment of the present application. As shown in FIG. 1, a first road area is marked at three locations (e.g., A, B, and C) on the map. For example, FIG. 3 represents the first road area in the form of an "exclamation mark".

[0087] It should be noted that in this embodiment of the present application, map data may be further updated to remove special road conditions. For example, if the road characteristic of Area 1 in the map data is a straight lane with a vehicle speed of less than 60 km / h, and the vehicle speed of the vehicle parameters reported by the vehicle in Area 1 at the previous moment is 10 km / h, 0 km / h, etc., then Area 1 is marked as the first road area in the map data at the previous moment. However, if it is determined that the lowest vehicle speed of the vehicle parameters reported by the vehicle in Area 1 within the preset period after the current moment is 60 km / h, etc., based on the vehicle parameters within the preset period at the current moment, then the special road condition of Area 1 at the previous moment may be further determined to have disappeared, and the map data may be updated to obtain the map data at the current moment. A specific method of updating the map data is to delete Area 1 marked as the first road area from the map.

[0088] FIG. 4 is a schematic diagram 2 of a map according to an embodiment of the present application. Compared to FIG. 3, the special road condition at location A of FIG. 4 is removed, and an updated map at the current moment is obtained as shown in FIG. 4. In FIG. 4, the first road area is marked at locations B and C.

[0089] In step S203, the terminal device can obtain map data at the current moment from the server. Optionally, an application program for displaying the map, for example, a navigation application program or an autonomous driving map application program, is installed on the terminal device. The terminal device can obtain map data at the current moment and update the map data in the application program of the terminal device.

[0090] In step S204, according to step S203, the terminal device can acquire map data at the current moment, so that the autonomous vehicle can acquire a first road area in advance. Correspondingly, the terminal device can display the first road area on the map, so that the user can know the first road area. In other words, in this embodiment of the present application, the autonomous vehicle or the non-autonomous vehicle can acquire a first road area in advance based on map data at the current moment. This solves the problem of real-time performance in the prior art.

[0091] In this embodiment of the present application, when a vehicle is driving along a planned route, the terminal device can determine whether a second road area exists along the vehicle's planned route according to the vehicle's planned route, that is, can recognize a special road condition in advance along the planned route. The second road area is a road area within the first road area. In other words, in this embodiment of the present application, whether there is a road area where a special road condition is located along the vehicle's planned route can be determined in advance. If the second road area exists along the vehicle's planned route, a driving decision (in the case of an autonomous vehicle, see step S508 in the following embodiment for details) or a prior reminder (in the case of a non-autonomous vehicle) is performed in advance, so that the reminder can be implemented in a timely manner.

[0092] The planned path in this embodiment of the present application can be obtained through a terminal device based on start and end points entered by a user, or can be obtained through a terminal device by requesting a server.

[0093] In this embodiment of the present application, the server, based on reported vehicle parameters and a road condition model, acquires an area where a special road condition is located at the current moment, i.e., a first road area, and updates the map data to acquire map data at the current moment. The map data at the current moment includes the first road area. Correspondingly, after the server acquires the map data at the current moment, the terminal device recognizes whether a special road condition exists within the vehicle's planned route and can recognize the special road condition within the planned route in advance. This improves the real-time performance of recognizing special road conditions.

[0094] Based on the above-described embodiment, FIGS. 5a and 5b are schematic flowchart 2 of a method for recognizing a special road condition according to an embodiment of the present application. As illustrated in FIGS. 5a and 5b, the method for recognizing a special road condition provided in this embodiment of the present application may include the following steps.

[0095] S501: The server obtains a first road area by inputting vehicle parameters within a preset period prior to the current moment into at least one road state model.

[0096] S502: The server acquires a target image of a first road area, the target image of the first road area is an image including special road conditions in the first road area in vehicle parameters, and the target image is a video frame from an image or video captured by a vehicle.

[0097] S503: The server generates descriptive information regarding special road conditions in the first road area based on the target image of the first road area, and the descriptive information is used to describe the scenario type of the special road condition.

[0098] S504: The server determines the duration of a special road condition in the first road area based on the scenario type of the special road condition represented by the descriptive information regarding the special road condition in the first road area.

[0099] S505: The server marks a first road area on the map data, and adds descriptive information regarding a special road condition in the first road area and / or a target image of the special road condition and the duration of the special road condition in the first road area to the map data, thereby acquiring the map data at the current moment.

[0100] S506: The terminal device acquires map data at the current moment.

[0101] S507: The terminal device determines whether a second road area exists on the vehicle's planned route based on the vehicle's planned route.

[0102] S508: If the vehicle is an autonomous vehicle and a second road area exists on the vehicle's planned path, the terminal device generates a driving command, and the driving command is used to direct the vehicle's driving behavior.

[0103] S509: When the vehicle travels into the second road area, the terminal device controls the vehicle's driving based on a driving command.

[0104] S510: If the vehicle is a non-autonomous vehicle and a second road area exists on the vehicle's planned route, the terminal device generates a reminder message, and the reminder message is used to indicate that a special road condition exists in the second road area.

[0105] S511: When the vehicle attempts to drive into the second road area, the terminal device pushes a reminder message.

[0106] In step S501, in this embodiment of the present application, there are a plurality of road condition models. Each road condition model is used to recognize that vehicle parameters having different features are vehicle parameters corresponding to special road conditions. For example, road condition model 1 is used to recognize a special road condition where a manhole cover is missing, road condition model 2 is used to recognize a special road condition where the road is blocked, and road condition model 3 is used to recognize a special road condition where a vehicle is sliding.

[0107] Vehicle parameters reported by different vehicles differ when vehicles pass through road areas where different special road conditions are located. Therefore, different road condition models are trained for different special road conditions, thereby improving the accuracy of recognizing special road conditions. In this embodiment of the application, a first road area is obtained by inputting vehicle parameters within a preset period prior to the current moment into at least one road condition model. In the case of vehicle parameters, when the vehicle parameters are input into road condition model 1, the result output by road condition model 1 may be that the vehicle parameters are not vehicle parameters corresponding to a special road condition. However, when the vehicle parameters are input into road condition model 2, the result output by road condition model 2 may be that the vehicle parameters are vehicle parameters corresponding to a special road condition. In this way, the vehicle parameters may be determined to be vehicle parameters corresponding to a blocked road.

[0108] In this embodiment of the present application, it should be understood that the method of obtaining a first road area may be as follows: based on the output of at least one road condition model, vehicle parameters may be determined to be vehicle parameters corresponding to a special road condition, and the vehicle parameters include the location of the vehicle. In this embodiment of the present application, an area containing a preset quantity of vehicle parameters corresponding to a special road condition may be used as the first road area. For example, if all vehicle parameters reported by 10 vehicles in area 1 are determined to be vehicle parameters corresponding to a special road condition, area 1 may be used as the first road area.

[0109] Alternatively, in this embodiment of the present application, the method of acquiring the first road area may alternatively be as follows: vehicle parameters within a preset period prior to the current moment may be input into at least one road condition model, thereby acquiring the first road area where the special road condition is located without a vehicle parameter analysis process for vehicle parameters corresponding to the special road condition.

[0110] In this embodiment of the present application, a road condition model may be obtained using a plurality of historical vehicle parameters as training parameters. Historical vehicle parameters are vehicle parameters received from at least one vehicle prior to a predetermined period. A method for obtaining a road condition model is described in detail below with reference to FIG. 6. FIG. 6 is a schematic flowchart for obtaining a road condition model according to an embodiment of the present application. As illustrated in FIG. 6, the method for obtaining a road condition model in an embodiment of the present application includes the following steps.

[0111] S601: Split multiple historical vehicle parameters into N training data sets.

[0112] Each historical vehicle parameter may include the vehicle's location, images or videos captured by the vehicle, and the vehicle's attribute data and driving data. For details regarding the attribute data and driving data, refer to the relevant description in step S201. The vehicle parameters within each training data set have the same characteristics, and N is an integer greater than or equal to 1.

[0113] In this embodiment of the present application, special road conditions of different feature types may be obtained in advance. Special road conditions of different feature types may be features that affect a vehicle generating different vehicle parameters. For example, feature types of special road conditions include road blockage, missing manhole cover, frozen road, speed bump, etc.

[0114] It is found that vehicle parameters differ when a vehicle encounters special road conditions of different feature types. Accordingly, in this embodiment of the application, historical vehicle parameters may be divided into N training data sets based on the features of the historical vehicle parameters. The vehicle parameters within each training data set have the same features, that is, all vehicle parameters within each training data set are generated under the influence of special road conditions of the feature type. For example, the features of vehicle parameters generated under the influence of a blocked road may be: turning right after deceleration, turning left after deceleration, or turning around after deceleration. The features of vehicle parameters generated under the influence of a speed bump may be: body vibration after deceleration, etc.

[0115] In addition, in this embodiment of the present application, the degree of deceleration, body vibration, turning, etc., can be further distinguished to determine the feature type of a special road condition corresponding to the historical vehicle parameters. In this way, training data sets in which the historical vehicle parameters are located are obtained through partitioning. Deceleration may include slow deceleration, rapid deceleration, etc. Body vibration may include small amplitude vibration, large amplitude vibration (for example, the body vibration amplitude can be discretized into different integer values, e.g., 0-10, to implement amplitude partitioning), etc. Turning may include rapid turning, slow turning, etc.

[0116] S602: Each training data set is used as training data to train a road condition model, and at least one road condition model is obtained.

[0117] In this embodiment of the present application, each training data set can be used as training data to train a road condition model, thereby training a road condition mode. In this way, N training data sets are trained to obtain at least one road condition model.

[0118] When acquiring a road condition model by training a training data set, one or more vehicle parameters within the training data set may be labeled. In this case, the road condition model is trained by using the labeled training data set as training data for training the road condition model. Correspondingly, in this embodiment of the present application, N (i.e., at least one) road condition models may be acquired by training N training data sets. It should be understood that one or more vehicle parameters are labeled. For example, among one or more vehicle parameters, a vehicle parameter corresponding to deceleration is labeled as "Deceleration," and a vehicle parameter corresponding to left turn is labeled as "Left Turn."

[0119] In step S201, it should be understood that the first road area can be obtained through a road condition model. It should be understood that the road condition model in the aforementioned embodiment may be a model that incorporates at least one road condition model in this embodiment of the present application. Accordingly, a specific road condition can be recognized for vehicle parameters of different features.

[0120] In step S502, after the first road area is obtained according to step S501, the feature type of the first road area where the special road condition is located may be obtained, but the scenario type of the special road condition in the first road area cannot be determined. For example, the feature type of the special road condition is, for example, "blocked road," "road bump," or "road congestion." However, the scenario type of the special road condition, for example, "road blocked due to a car accident" or "road blocked due to flooding," cannot be determined.

[0121] Accordingly, in this embodiment of the present application, a scenario type of a special road condition in a first road area is further obtained based on vehicle parameters, thereby obtaining more detailed information regarding the special road condition in the first road area. Vehicle parameters reported by the vehicle include images or videos captured by the vehicle. In this embodiment of the present application, vehicle parameters reported by the vehicle in the first road area may be obtained based on the position of the vehicle in the first road area and the vehicle parameters, and images or videos captured by the vehicle are obtained from the vehicle parameters reported in the first road area. For convenience of explanation, the vehicle parameters reported in the first road area are referred to as target vehicle parameters in the following description.

[0122] In this embodiment of the present application, a target image of a first road area may be obtained from a target vehicle parameter. It should be understood that there may be a plurality of target vehicle parameters, and correspondingly, there may also be a plurality of images or videos in the target vehicle parameters. A method of obtaining a target image from a target vehicle parameter may be: using an image or video frame containing a specific road condition in the first road area as the image to be selected, and obtaining a target image from the selected image. It should be understood that the video in the target vehicle parameter may include a plurality of video frames.

[0123] Optionally, the server in this embodiment of the present application may store a recognition model in advance. The recognition model is used to express the correspondence between the features of an image and the scenario type of a special road condition, that is, to input an image into the recognition model. The recognition model can determine the scenario type of a special road condition by recognizing whether the image is an image containing a pixel block of a special road condition. In this embodiment of the present application, an image or video frame in the target vehicle parameters may be input into the recognition model, and an image or video frame containing a special road condition is used as the image to be selected. Additionally, the recognition model may further output the similarity of the image to be selected to express the accuracy of the image to be selected containing the special road condition. In this embodiment of the present application, the target image may be determined from the image to be selected based on image clarity and similarity. For example, the image to be selected having the highest image clarity may be used as the target image, or the image to be selected having the highest similarity may be used as the target image.

[0124] It should be understood that the recognition model in this embodiment of the present application can be acquired through training using a machine learning method with a plurality of types of images including specific road conditions as a training data set. The machine learning method for training the recognition model may be the same as the aforementioned method for training the road condition model.

[0125] In step S503, descriptive information regarding special road conditions in this embodiment of the present application is used to describe the scenario type of the special road condition. For example, the scenario type of the special road condition may be traffic congestion, missing manhole covers, or uneven manhole covers.

[0126] In this embodiment of the present application, the type of special road condition in the first road area is determined based on the target image of the first road area, and descriptive information regarding the special road condition in the first road area can be generated based on the scenario type of the special road condition in the first road area. For example, if the type of special road condition in the first road area is a missing manhole cover, the descriptive information regarding the special road condition in the first road area may be a detailed description of the scenario type of the special road condition in the first road area. For example, the descriptive information regarding the special road condition in the first road area may be that a manhole cover on the first lane on the left side of the east direction of XX road is missing.

[0127] In this embodiment of the present application, the method for determining the type of special road condition in the first road area may be as follows: a recognition model is used to express the correspondence between the features of an image and the scenario type of a special road condition, that is, the image is input into the recognition model to obtain the scenario type of a special road condition in the image.

[0128] Another method for determining the type of special road condition in the first road area may be as follows: there may be multiple recognition models. Each recognition model is used to express the correspondence between the special road condition of the scenario type and the features of the image. In this embodiment of the present application, an image or video frame in the target vehicle parameters may be input to the multiple recognition models, and the input image is the scenario type of the special road condition expressed by the special road condition recognition model, that is, the scenario type of the special road condition contained in the image. For example, recognition model 1 is used to express the correspondence between a missing manhole cover and the features of the image, that is, to recognize an image containing a missing manhole cover. Recognition model 2 is used to express the correspondence between traffic congestion and the features of the image, that is, to recognize an image containing traffic congestion. Recognition model 3 is used to express the correspondence between a speed bump and the features of the image, that is, to recognize an image containing a speed bump.

[0129] In this way, the machine learning method used to train each recognition model may be the same as the aforementioned method for training the recognition model. However, it should be noted that the training data for training each recognition model is different from the training data for training the recognition model. In this method, the training data used to train each recognition model is an image containing specific road conditions of the same scenario type, and the training data used to train the recognition model is an image containing specific road conditions of various scenario types. For example, in this embodiment of the present application, the training data for training recognition model 1 may be a plurality of images containing missing manhole covers.

[0130] In step S504, the duration of the special road condition in the first road area is the time from the present moment until the moment the first road area is removed. It should be understood that the duration of the special road condition in the first road area may be determined based on an empirical average value obtained through statistical collection of big data, or it may be the time from the occurrence of the special road condition to its removal. In this embodiment of the application, the duration of the special road condition in the first road area may be determined based on the scenario type of the special road condition in the first road area. Optionally, the server stores an empirical value of the duration of the special road condition for each scenario type. The empirical value may be entered by a user (technician) or obtained by the server based on the duration of the historical special road condition. For example, the server may use the average, maximum, or minimum value of the duration of the historical special road condition as the duration of the special road condition for which the scenario type is the same as the scenario type of the historical special road condition. For example, if the special road condition is one day for the duration of missing manhole covers, the special road condition is one where the duration of debris flow is four hours, or is similar.

[0131] In step S505, in this embodiment of the present application, after the first road area, descriptive information regarding the special road condition in the first road area and / or a target image of the special road condition, and the duration of the special road condition in the first road area are obtained, the first road area may be marked on the map data. Additionally, the descriptive information regarding the special road condition in the first road area and / or a target image of the special road condition and the duration of the special road condition in the first road area are added to the map data. That is, the current map data includes the first road area, descriptive information regarding the special road condition in the first road area and / or a target image of the special road condition, and the duration of the special road condition in the first road area.

[0132] It should be understood that for the implementations of steps S506 to S507 in this embodiment of the present application, reference is made to the relevant descriptions of steps S203 and S204 in the aforementioned embodiment. Details are not described again herein.

[0133] In step S508, where the vehicle is an autonomous vehicle and the second road area exists on the vehicle's planned path, in this embodiment of the application, a driving decision, i.e., a driving command, may be generated for the autonomous vehicle. The driving command is used to direct the driving behavior of the vehicle. For example, the driving command may be a command instructing the vehicle to decelerate and turn right, or a command instructing the vehicle to decelerate.

[0134] In this embodiment of the present application, since there are multiple road condition models and the map data additionally includes descriptive information regarding special road conditions in a first road area, in this embodiment of the present application, a driving command may be generated based on descriptive information regarding special road conditions in a second road area. It should be understood that the driving behavior indicated by the driving command is identical to the driving behavior indicated by the characteristics of the historical vehicle parameters used to acquire the target road condition model, and that the target road condition model is a model for determining the second road area as a special road condition. In other words, in this embodiment of the present application, a model for inputting the second road area as a special road condition is used as the target road condition model, and a driving behavior indicated by the characteristics of the historical vehicle parameters for training the target road condition model is used as the driving command. For example, when a missing manhole cover exists in the second road area, the model for outputting the second road area as a special road condition is Road Condition Model 2, and the characteristics of the historical vehicle parameters for training Road Condition Model 2 are to first decelerate and then turn right. In this case, as a driving command for the second road area, decelerating first and then turning right may be used.

[0135] In step S509, when the autonomous vehicle is driving into the second road area, the terminal device controls the vehicle's driving based on a driving command. Specifically, in this embodiment of the present application, a driving command for the autonomous vehicle in the second road area may be generated in advance in step S508 so that the autonomous vehicle may be reminded in advance. In this case, the autonomous vehicle may drive based on the driving command when driving into the second road area. For example, if the driving command for the second road area is to decelerate first and then turn right, the autonomous vehicle may decelerate first and then turn right when driving into the second road area.

[0136] Optionally, to further improve the safety of the autonomous vehicle, in this embodiment of the application, a driving command may be additionally executed at a preset distance before the autonomous vehicle travels into a second road area. For example, when it is still 1 meter away from the second road area, the vehicle first decelerates and then turns right.

[0137] In step S510, if the vehicle is a non-autonomous vehicle and a second road area exists on the vehicle's planned route, a reminder message may be generated in this embodiment of the present application to remind the user driving the vehicle. The map data further includes descriptive information regarding a special road condition in the first road area and / or a target image of the special road condition. In the reminder message generated in this embodiment of the present application, descriptive information regarding a special road condition in the second road area and / or a target image of the special road condition may be used as a reminder message.

[0138] In step S511, to remind the user in advance of the special road conditions in the second road area, the terminal device may push a reminder message when the vehicle attempts to drive into the second road area. Specifically, the method of pushing the reminder message may be to play explanatory information regarding the special road conditions in the second road area, and / or to display a target image of the special road conditions in the second road area.

[0139] For example, if the scenario type of a special road condition in the first road area is a missing manhole cover, the descriptive information regarding the special road condition in the first road area may be that a manhole cover on the first lane on the left side of the east direction of XX road is missing. In this embodiment of the present application, when a vehicle attempts to drive into the second road area (for example, when there is a preset distance from the second road area), a reminder message "a manhole cover on the first lane on the left side of the east direction of XX road is missing" may be played, and a target image of the special road condition in the second road area is displayed on the display screen of the terminal device.

[0140] FIG. 7 is a schematic diagram 1 of an interface change of a terminal device according to an embodiment of the present application. As illustrated in the interface (701) of FIG. 7, the interface (701) displays a navigation interface of the vehicle. When the vehicle intends to drive into a second road area, the interface (701) may jump to an interface (702). The interface (702) displays a target image of a specific road condition in the second road area. For example, the interface (702) displays an image of a "missing manhole cover" in the second road area. It should be understood that an example in which the terminal device is an in-vehicle terminal is used for the description in this embodiment of the present application.

[0141] Optionally, in this embodiment of the present application, there are a plurality of first road areas, and an identifier of a special road condition in each first road area may be additionally displayed on the map. FIG. 8 is a schematic diagram 2 of an interface change of a terminal device according to an embodiment of the present application. As illustrated in the interface (801) of FIG. 8, the interface (801) displays an identifier of a special road condition in the first road area on the vehicle's navigation interface displayed on the interface (701).

[0142] For example, the identifiers of special road conditions in the first road area may be identical, for example, all of which are icons with an exclamation mark. Alternatively, the identifier of a special road condition in the first road area may represent a scenario type of the special road condition in the first road area. As illustrated in the interface (801) of FIG. 8, when a special road condition in the first road area includes a missing manhole cover, an uneven manhole cover, or a debris flow, the corresponding identifier may be displayed at a location corresponding to the first road area on the map. For example, identifier 1 indicating a missing manhole cover is marked at location A, identifier 2 indicating an uneven manhole cover is marked at location B, and identifier 3 indicating a missing manhole cover is marked at location C.

[0143] In this embodiment of the present application, the terminal device receives a selection command from a user for an identifier of a special road condition in any first road area and displays descriptive information regarding the special road condition in the first road area selected by the user, thereby enabling the user to obtain a scenario type of the selected special road condition in the first road area. For example, when the user selects identifier 1 by tapping, the interface (801) jumps to the interface (802). The interface (802) displays descriptive information regarding the special road condition at location A, for example, that a manhole cover on the first lane on the left side of the east direction of road XX is missing.

[0144] It should be understood that steps S508-S509 or steps S510-S511 are steps selected to be performed. It should be understood that steps S508-S509 are steps performed when the vehicle is an autonomous vehicle, and steps S510-S511 are steps performed when the vehicle is a non-autonomous vehicle.

[0145] Steps S507-S511 may be a scenario in which a vehicle is traveling. In this scenario, the map data additionally includes the duration of a special road condition in a first road area. If a second road area exists on the vehicle's planned route, and the scenario type of the special road condition in the second road area is a preset scenario type, and the time the vehicle travels to the second road area is shorter than the duration of the special road condition in the second road area, the server may be requested to update the vehicle's planned route to obtain an updated planned route. Optionally, the preset scenario type is a predetermined scenario type, and the scenario type of the special road condition that the vehicle cannot pass through quickly, e.g., traffic congestion or debris flow.

[0146] For example, if a second road area exists on the vehicle's planned route, the scenario type of the special road condition in the second road area is a debris flow, it takes 30 minutes for the vehicle to travel to the second road area, and the duration of the debris flow in the second road area is 4 hours, the server may be requested to update the vehicle's planned route to avoid the second road area. In this way, the updated planned route is obtained and additionally displayed on the terminal device. Optionally, the reason for updating the planned route, for example, a text reminder message "Route updated due to debris flow ahead" may be additionally displayed on the terminal device.

[0147] The planned route of a vehicle can be obtained through a terminal device by requesting it from a server. Upon receiving a route planning request entered by a user, the terminal device can transmit the route planning request to the server. After receiving the route planning request from the terminal device, the server can obtain the planned route based on the start point and end point, the duration of the special road condition in the first road area, and the scenario type of the special road condition in the first road area. When designing the planned route, the server can avoid planning a route that includes a special road condition of a pre-set scenario type.

[0148] In this embodiment of the present application, the road condition model is obtained after a large amount of historical vehicle parameters have been trained. Therefore, the recognition of the first road area at the current moment through the road condition model has high accuracy. Furthermore, in this embodiment of the present application, the terminal device can generate a driving decision or reminder message in advance based on current map data to remind autonomous vehicles and non-autonomous vehicles. This improves the user experience. Additionally, the terminal device can further update a pre-planned, pre-set route based on current map data or set a pre-set route for the vehicle. This can further improve the user experience.

[0149] FIG. 9 is a schematic diagram 1 of the structure of a device for recognizing special road conditions according to an embodiment of the present application. As shown in FIG. 9, the device for recognizing special road conditions may be a terminal device in the aforementioned embodiments. The device for recognizing special road conditions (900) includes a processing module (901), a playback module (902), a display module (903), and a transceiver module (904).

[0150] The processing module (901) is configured to: acquire map data at the current moment - the map data includes a first road area at the current moment -; determine whether a second road area exists on the vehicle's planned route based on the vehicle's planned route, wherein the second road area is a road area within the first road area, the first road area is a road area where a special road condition is located, the first road area is acquired based on a road condition model and vehicle parameters within a preset period prior to the current moment, and the road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0151] Optionally, if the vehicle is an autonomous vehicle, the processing module (901) is further configured to: generate a driving command when a second road area exists on the vehicle's planned path; and, when the vehicle travels to the second road area, to control the vehicle's driving based on the driving command, wherein the driving command is used to direct the vehicle's driving behavior.

[0152] Optionally, there are multiple road condition models, and the map data additionally includes descriptive information regarding special road conditions in the first road area, and the descriptive information is used to describe the scenario type of the special road condition.

[0153] Accordingly, the processing module (901) is specifically configured to generate a driving command based on descriptive information regarding a special road condition in a second road area. The driving behavior indicated by the driving command is identical to the driving behavior indicated by the characteristics of the historical vehicle parameters used to obtain a target road condition model. The target road condition model is a model for determining the second road area as a special road condition.

[0154] Optionally, if the vehicle is a non-autonomous vehicle, the processing module (901) is further configured to: generate a reminder message when a second road area exists on the vehicle's planned route; and push a reminder message when the vehicle attempts to drive into the second road area, the reminder message being used to indicate that a special road condition exists in the second road area.

[0155] Optionally, the map data at the current moment additionally includes descriptive information regarding special road conditions in the first road area and / or a target image of the special road condition. The descriptive information is used to describe the scenario type of the special road condition. The target image of the first road area is an image containing the special road condition in the first road area in the vehicle parameters.

[0156] In response to this, the processing module (901) is specifically configured to use descriptive information regarding special road conditions in the second road area and / or a target image of the special road conditions as a reminder message.

[0157] The playback module (902) is configured to play explanatory information regarding special road conditions in the second road area; and / or,

[0158] The display module (903) is configured to display a target image of a special road condition in the second road area.

[0159] Optionally, the map data additionally includes the duration of special road conditions in the first road area.

[0160] Optionally, there are multiple first road areas. The display module (903) is further configured to display an identifier of a special road condition in each first road area on the map.

[0161] The transceiver module (904) is configured to receive a selection command from a user for an identifier of a special road condition in any first road area. In response, the display module (903) is further configured to display descriptive information regarding the special road condition in the first road area selected by the user.

[0162] Optionally, the transceiver module (904) is further configured to send a route planning request to the server and to receive the planned route sent by the server.

[0163] Optionally, the map data additionally includes the duration of special road conditions in the first road area.

[0164] Optionally, the transceiver module (904) is further configured to report vehicle parameters to a server. The vehicle parameters include the location of the vehicle, images or videos taken by the vehicle, and attribute data and driving data of the vehicle.

[0165] For the beneficial effects of the device for recognizing special road conditions provided in this embodiment of the present application, reference is made to the beneficial effects of the method for recognizing special road conditions described above. Details are not described again herein.

[0166] FIG. 10 is a schematic diagram 2 of the structure of a device for recognizing special road conditions according to an embodiment of the present application. As shown in FIG. 10, the device for recognizing special road conditions may be a server in the aforementioned embodiment. The device for recognizing special road conditions (1000) includes a processing module (1001) and a transceiver module (1002).

[0167] The processing module (1001) is configured to: acquire a first road area at the current moment based on a road condition model and vehicle parameters within a preset period prior to the current moment; and to acquire map data at the current moment by marking the first road area on the map data. The first road area is a road area where a special road condition is located. The road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition.

[0168] Optionally, vehicle parameters include the vehicle's position.

[0169] The transceiver module (1002) is configured to receive vehicle parameters reported by at least one vehicle within a preset period.

[0170] In response to this, the processing module (1001) is further configured to determine vehicle parameters within a preset period prior to the current moment based on vehicle parameters of at least one vehicle. Vehicle parameters within a preset period prior to the current moment are vehicle parameters that do not match the road features at the vehicle's location in the map data.

[0171] Optionally, the processing module (1001) is further configured to obtain a road condition model using a plurality of historical vehicle parameters as training parameters. The historical vehicle parameters are vehicle parameters received from at least one vehicle prior to a preset period.

[0172] Optionally, there are multiple road condition models.

[0173] The processing module (1001) is specifically configured to: divide a plurality of historical vehicle parameters into N training data sets - each training data set has the same characteristics and N is an integer greater than 1 -; and use each training data set as training data to train a road condition model to obtain at least one road condition model.

[0174] Optionally, the processing module (1001) is specifically configured to obtain a first road area by inputting vehicle parameters within a preset period prior to the current moment into at least one road state model.

[0175] Optionally, vehicle parameters include images or videos captured by the vehicle.

[0176] The processing module (1001) is further configured to: acquire a target image of a first road area; generate descriptive information regarding a special road condition in the first road area based on the target image of the first road area; and add the descriptive information regarding the special road condition in the first road area and / or the target image of the special road condition to map data. The target image of the first road area is an image containing the special road condition in the first road area in vehicle parameters. The target image is an image or video frame from a video captured by a vehicle. The descriptive information is used to describe the scenario type of the special road condition.

[0177] Optionally, the processing module (1001) is further configured to: determine the duration of the special road condition in the first road area based on the scenario type of the special road condition represented by descriptive information regarding the special road condition in the first road area; and add the duration of the special road condition in the first road area to map data.

[0178] Optionally, the processing module (1001) is further configured to obtain a planned route based on a start point and an end point, the duration of a special road condition in a first road area, and a scenario type of a special road condition in a first road area when a route planning request is received from a terminal device. The route planning request includes a start point and an end point.

[0179] The transceiver module (1002) is additionally configured to push the planned path to the terminal device.

[0180] Optionally, vehicle parameters include vehicle attribute data and driving data.

[0181] For the beneficial effects of the device for recognizing special road conditions provided in this embodiment of the present application, reference is made to the beneficial effects of the method for recognizing special road conditions described above. Details are not described again herein.

[0182] FIG. 11 is a schematic diagram 1 of the structure of an electronic device according to an embodiment of the present application. As illustrated in FIG. 11, the electronic device may be the terminal device of FIG. 9, and the electronic device may include a processor (1101), a player (1102), a display (1103), a transceiver (1104), and a memory (1105). It should be understood that the processor (1101) performs the action of the processing module (901), the player (1102) performs the action of the playback module (902), the display (1103) performs the action of the display module (903), and the transceiver (1104) performs the action of the transceiver module (904). The memory (1105) may store various instructions to complete various processing functions and implement the method steps of the present application.

[0183] A transceiver (1104) is coupled to a processor (1101), and the processor (1101) controls the transmission and reception actions of the transceiver (1104 (1202)). The memory (1105) may include high-speed random-access memory (RAM) or may additionally include non-volatile memory (NVM), for example, at least one magnetic disk memory. Optionally, the electronic device in this application may additionally include a power supply (1106), a communication bus (1107), and a communication port (1108). The transceiver (1104) may be integrated into the transceiver of a terminal device or may be an independent transceiver antenna of the terminal device. The communication bus (1107) is configured to implement communication and connection between elements. The communication port (1108) is configured to implement connection and communication between the terminal device and other peripheral devices. The display (1103) is connected to the processor (1101) and can display the setting interface of the above-described embodiment under the control of the processor (1101).

[0184] In this embodiment of the present application, the memory (1105) is configured to store computer-executable program code. The program code includes instructions. When the processor (1101) executes the instructions, the instructions enable the processor (1101) of the terminal device to perform processing actions of the terminal device in the aforementioned method embodiments, and enable the transceiver (1104) to perform transmission and reception actions of the terminal device in the aforementioned method embodiments. The implementation principles and technical effects are similar and are not described again herein.

[0185] FIG. 12 is a schematic diagram 2 of the structure of an electronic device according to an embodiment of the present application. As illustrated in FIG. 12, the electronic device may be the server of FIG. 10, and the electronic device may include a processor (1201), a transceiver (1202), and a memory (1203). It should be understood that the processor (1201) performs the actions of the processing module (1001), and the transceiver (1202) performs the actions of the transceiver module (1002). The memory (1203) may store various instructions to complete various processing functions and implement the method steps of the present application.

[0186] A transceiver (1202) is coupled to a processor (1201), and the processor (1201) controls the transmission and reception actions of the transceiver (1202). Memory (1203) may include high-speed random-access memory (RAM) or non-volatile memory (NVM), for example, at least one magnetic disk memory. Optionally, the electronic device in this application may further include a power supply (1204), a communication bus (1205), and a communication port (1206). The transceiver (1202) may be integrated into the transceiver of a terminal device or may be an independent transceiver antenna of the terminal device. The communication bus (1205) is configured to implement communication and connection between elements. The communication port (1206) is configured to implement connection and communication between the terminal device and other peripheral devices. The display (1103) is connected to the processor (1201) and can display the setting interface of the above-described embodiment under the control of the processor (1201).

[0187] In this embodiment of the present application, the memory (1203) is configured to store computer-executable program code. The program code includes instructions. When the processor (1201) executes the instructions, the instructions enable the processor (1201) of the terminal device to perform processing actions of the terminal device in the aforementioned method embodiments, and enable the transceiver (1202) to perform transmission and reception actions of the terminal device in the aforementioned method embodiments. The implementation principles and technical effects are similar and are not described again herein.

[0188] All or part of the foregoing embodiments may be implemented using software, hardware, firmware, or any combination thereof. When software is used to implement the embodiments, all or part of the embodiments may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When computer program instructions are loaded and executed on a computer, all or part of the procedures or functions according to the embodiments of the present application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions may be stored on a computer-readable storage medium or transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, computer instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, or Digital Subscriber Line (DSL)) or wireless (e.g., infrared, radio, or microwave). A computer-readable storage medium may be any available medium accessible by a computer, or a data storage device such as a server or data center that incorporates one or more available media. Available media may include magnetic media (e.g., floppy disk, hard disk, or magnetic tape), optical media (e.g., DVD), semiconductor media (e.g., SSD (solid state disk)), etc.

[0189] In this specification, the term "plural" means two or more. In this specification, the term "and / or" describes only the association relationship to explain the associated objects and indicates that three relationships may exist. For example, A and / or B may indicate the following three cases: A alone exists, both A and B exist, and B alone exists. Additionally, in this specification, the character " / " generally indicates an "or" relationship between associated objects. In formulas, the character " / " indicates a "partition" relationship between associated objects.

[0190] It will be understood that the various numbers in the embodiments of this application are used only for distinction to facilitate explanation and are not used to limit the scope of the embodiments of this application.

[0191] It should be understood that the sequence numbers of the aforementioned processes do not imply execution sequences in the embodiments of this application. Execution sequences of the processes should be determined according to the functions and internal logic of the processes and should not be interpreted as any limitation on the implementation processes of the embodiments of this application.

Claims

Claim 1 A method for recognizing a special road condition applied to a terminal device, comprising the step of acquiring map data at a current moment - said map data includes a first road area at said current moment, said first road area is a road area where said special road condition is located, said first road area is acquired based on a road condition model and vehicle parameters within a preset period prior to said current moment, said road condition model is used to express the correspondence between the characteristics of said vehicle parameters and said special road condition - ; A method comprising the step of determining whether a second road area exists in the planned route of the vehicle based on the planned route of the vehicle, wherein the second road area is a road area within the first road area, and the empirical value of the duration of the special road condition for each scenario type is obtained by a server based on the average, maximum, or minimum value of the historical duration of the special road condition for each scenario type, and the duration of the special road condition in the first road area is determined based on the scenario type of the special road condition expressed by descriptive information regarding the special road condition in the first road area, and the descriptive information regarding the special road condition in the first road area is generated based on a target image of the first road area, wherein the target image of the first road area is an image including the special road condition in the first road area in the vehicle parameters. Claim 2 The method of claim 1, wherein the vehicle is an autonomous vehicle, the method further comprises: a step of generating a driving command when the second road area exists on the planned path of the vehicle, wherein the driving command is used to direct the driving behavior of the vehicle; and a step of controlling the driving of the vehicle based on the driving command when the vehicle drives into the second road area. Claim 3 A method according to paragraph 2, wherein there are multiple road condition models, the map data further includes descriptive information regarding the special road condition in the first road area, and the step of generating the driving command includes: generating the driving command based on descriptive information regarding the special road condition in the second road area, wherein the driving behavior indicated by the driving command is identical to the driving behavior indicated by the characteristics of the historical vehicle parameters used to obtain the target road condition model, and the target road condition model is a model for determining the second road area as the special road condition. Claim 4 In claim 1, where the vehicle is a non-autonomous vehicle, the method comprises: generating a reminder message when the second road area exists on the planned path of the vehicle, wherein the reminder message is used to indicate that a special road condition exists in the second road area; and further comprising the step of pushing the reminder message when the vehicle attempts to drive into the second road area. Claim 5 In claim 4, the map data further comprises the descriptive information regarding the special road condition in the first road area and / or the target image of the special road condition, and the step of generating the reminder message comprises: using the descriptive information regarding the special road condition in the second road area and / or the target image of the special road condition as the reminder message; and the step of pushing the reminder message comprises: playing the descriptive information regarding the special road condition in the second road area, and / or displaying the target image of the special road condition in the second road area. Claim 6 A method according to claim 5, wherein there are a plurality of first road areas, and the method further comprises: a step of displaying an identifier of a special road condition in each first road area on a map; and a step of receiving a selection command from a user for an identifier of a special road condition in any first road area, and displaying descriptive information regarding the special road condition in the first road area selected by the user. Claim 7 A method according to any one of claims 1 to 6, wherein the method further comprises: a step of transmitting a path planning request to the server; and a step of receiving the planned path transmitted by the server. Claim 8 A method according to any one of claims 1 to 6, wherein the method further comprises the step of reporting the vehicle parameters to the server, wherein the vehicle parameters include the location of the vehicle, an image or video captured by the vehicle, and attribute data and driving data of the vehicle. Claim 9 A method for recognizing special road conditions applied to a server, comprising: a step of acquiring a first road area at the current moment based on a road condition model and vehicle parameters within a preset period prior to the current moment, wherein the first road area is a road area where the special road condition is located, and the road condition model is used to express the correspondence between the characteristics of the vehicle parameters and the special road condition; and a step of acquiring map data at the current moment by marking the first road area on map data, wherein the vehicle parameters include an image or video captured by a vehicle, and the method comprises: a step of acquiring a target image of the first road area, wherein the target image of the first road area is an image containing the special road condition in the first road area in the vehicle parameters, and the target image is a video frame in the image or video captured by the vehicle; a step of generating descriptive information regarding the special road condition in the first road area based on the target image of the first road area, wherein the descriptive information is used to describe the scenario type of the special road condition; A method further comprising: a step of determining the duration of the special road condition in the first road area based on a scenario type of the special road condition expressed by descriptive information regarding the special road condition in the first road area; and a step of adding descriptive information regarding the special road condition in the first road area and / or a target image of the special road condition to the map data, wherein the method further comprises: a step of obtaining an empirical value of the duration of the special road condition for each scenario type based on an average value, a maximum value, or a minimum value of the duration of the historical special road condition for each scenario type. Claim 10 In claim 9, the vehicle parameters include the location of the vehicle, and before obtaining a first road area at the present moment, the method further comprises: receiving vehicle parameters reported by at least one vehicle within the preset period; and determining vehicle parameters within a preset period prior to the present moment based on the vehicle parameters of the at least one vehicle, wherein the vehicle parameters within the preset period prior to the present moment are vehicle parameters that do not match the road features at the location of the vehicle in the map data. Claim 11 In claim 10, the method further comprises the step of obtaining the road condition model using a plurality of historical vehicle parameters as training parameters, wherein the historical vehicle parameters are vehicle parameters received from at least one vehicle prior to the preset period. Claim 12 In claim 11, the step of acquiring a road condition model using a plurality of historical vehicle parameters as training parameters comprises: a step of dividing the plurality of historical vehicle parameters into N training data sets - each training data set has the same vehicle parameters and N is an integer greater than 1 -; and a step of acquiring at least one road condition model by using each training data set as training data for training a road condition model. Claim 13 In claim 12, the step of obtaining a first road area at the present moment based on a road condition model and vehicle parameters within a preset period prior to the present moment comprises: a step of obtaining the first road area by inputting the vehicle parameters within the preset period prior to the present moment into at least one road condition model. Claim 14 delete Claim 15 In claim 9, the method further comprises the step of adding the duration of the special road condition in the first road area to the map data. Claim 16 In claim 15, the method further comprises: a step of obtaining a planned route based on a start point and an end point, a duration of the special road condition in the first road area, and a scenario type of the special road condition in the first road area when a route planning request is received from a terminal device, wherein the route planning request includes the start point and the end point; and a step of pushing the planned route to the terminal device. Claim 17 A method according to any one of claims 9 to 13, wherein the vehicle parameters include attribute data and driving data of the vehicle. Claim 18 An apparatus comprising units configured to perform the steps of a method according to any one of claims 1 to 6. Claim 19 An apparatus comprising units configured to perform the steps of a method according to any one of claims 9 through 13. Claim 20 A computer-readable medium in which a program is stored, wherein when the program is executed by a processor, a method according to any one of claims 1 to 6 is performed. Claim 21 A computer-readable medium in which a program is stored, wherein when the program is executed by a processor, a method according to any one of claims 9 to 13 is performed.

Images