Method and device for determining vehicle driving road type

Abstract

This article belongs to the field of autonomous driving technology, specifically to a method and device for determining the type of road a vehicle is traveling on. The method includes: when a vehicle enters a road type identified by a visual perception model, continuously collecting image data of the road type the vehicle is traveling on; uploading the collected image data of the road type to a model server corresponding to the visual perception model, so that the model server updates the model parameters of the visual perception model based on the received image data of the road type, and then sends the updated model parameters to the vehicle equipped with the visual perception model; receiving the updated model parameters returned by the model server, and updating the visual perception model based on the updated model parameters to obtain an updated visual perception model; receiving the collected current image data; and inputting the current image data into the updated visual perception model to determine the road type the vehicle is currently on. In practical applications, this method can ensure that the model can recognize a wider range of scenarios and is more accurate.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a method and device for determining the type of road a vehicle is traveling on. Background Art

[0002] With technological advancements, autonomous driving technology has developed rapidly. Currently, autonomous vehicles are equipped with a variety of sensors, such as forward-looking cameras, surround-view cameras, perimeter cameras, LiDAR, millimeter-wave radar, GPS, IMU, and RTK. These sensors provide perception data to determine the vehicle itself and its surroundings. The decision-making system controls the vehicle's speed and direction based on perception and control algorithms. AI perception models and high-precision maps are core technologies in autonomous driving systems and require large amounts of data for updating and improvement. While data acquisition for common scenarios is simple and low-cost, for some specialized scenarios, the cost of acquiring data is high and the data volume is relatively small. This inevitably results in the constructed models being unable to accurately identify specific scenarios, resulting in autonomous driving scenarios failing to meet user needs. Summary of the Invention

[0003] In order to solve the above technical problems, the specific technical solutions of this article are as follows:

[0004] In one aspect, a method for determining a type of road on which a vehicle is traveling is provided, the method comprising:

[0005] Get the current vehicle location information;

[0006] Determining whether the current vehicle is entering a road type identified by the visual perception model based on the position information;

[0007] When the current vehicle enters a road type identified by the visual perception model, continuously collecting image data of the current vehicle passing through the road type;

[0008] Uploading the collected image data of the road type to a model server corresponding to the visual perception model, so that the model server updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model;

[0009] receiving updated model parameters returned by the model server, and updating the visual perception model according to the updated model parameters to obtain an updated visual perception model;

[0010] Receive and collect current image data;

[0011] The current image data is input into the updated visual perception model to obtain the road type on which the vehicle is currently located.

[0012] Furthermore, the method further comprises:

[0013] Collecting trajectory information of the current vehicle passing through the road type;

[0014] Uploading the image data of the road type and the trajectory information to a map server for constructing a map, so that the map server updates the map image data corresponding to the location information according to the received image data of the road type and the trajectory information, and sends the updated map image data to a navigation device in the vehicle;

[0015] receiving the map image data sent by the map server, and updating the map in the current vehicle based on the map image data to obtain an updated vehicle map;

[0016] The current position information of the current vehicle is obtained, and the type of road on which the current vehicle is located is determined based on the current position information and the vehicle updated map.

[0017] Furthermore, the method further comprises:

[0018] Obtaining distance information of fixed obstacles in the image data;

[0019] Accordingly, uploading the collected image data of the road type and the trajectory information to a map server for constructing a map includes:

[0020] The collected image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data are uploaded to the map server for constructing the map, so that the map server updates the map corresponding to the location information based on the received image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data.

[0021] Furthermore, uploading the collected image data of the road type to the model server corresponding to the visual perception model includes:

[0022] Encrypting and compressing the collected image data of the road type based on a pre-configured encryption compression model to obtain encrypted image data;

[0023] The encrypted image data is uploaded to the model server corresponding to the visual perception model.

[0024] In another aspect, a method for determining the type of road a vehicle is traveling on is provided. The method comprises:

[0025] receiving image data corresponding to the same road type;

[0026] When the number of received image data corresponding to the same road type reaches a preset number requirement for the visual perception model input, the image data is used as the input of a pre-built visual perception model, and the road type corresponding to the image data is used as the output of the visual perception model. The visual perception model is trained until the model parameters in the visual perception model reach the preset requirements, thereby obtaining updated model parameters.

[0027] The updated model parameters are sent to a vehicle equipped with the visual perception model.

[0028] Furthermore, the training of the visual perception model includes:

[0029] identifying image features in the image data corresponding to the road type;

[0030] Filtering out the image data having the same image features to obtain target image data;

[0031] All image features in the target image data are marked according to feature categories of different image features.

[0032] Furthermore, the receiving of image data corresponding to the same road type includes:

[0033] receiving encrypted image data corresponding to the same road type;

[0034] The encrypted image data corresponding to the same road type is decrypted and decompressed based on a pre-configured decryption and decompression model to obtain image data corresponding to the same road type.

[0035] In another aspect, a method for determining a type of road on which a vehicle is traveling is provided, the method comprising:

[0036] receiving image data and trajectory information of a road type, wherein the trajectory information carries position information;

[0037] calibrating and splicing the image data based on the trajectory information to obtain position image data corresponding to the position information;

[0038] comparing the position image data with map image data in a pre-stored map, where the position information of the map image data is the same as the position information of the position image data;

[0039] When the accuracy of the position image data is higher than the accuracy of the map image data, updating the map image data in the pre-stored map based on the position image data to obtain updated map image data;

[0040] The updated map image data is transmitted to a navigation device in a vehicle, so that the navigation device updates a map in the navigation device based on the updated map image data.

[0041] Furthermore, the method further comprises:

[0042] receiving distance information of fixed obstacles in the image data;

[0043] The calibrating and stitching the image data based on the trajectory information to obtain position image data corresponding to the position information includes:

[0044] The image data is calibrated and spliced ​​based on the trajectory information and the distance information of the fixed obstacles in the image data to obtain position image data corresponding to the position information.

[0045] In another aspect, the present invention provides a device for determining the type of a road on which a vehicle is traveling, the device comprising:

[0046] Position information acquisition module, used to obtain the current vehicle's position information;

[0047] a judgment module, configured to judge, based on the position information, whether the current vehicle is entering a road type identified by the visual perception model;

[0048] an image acquisition module, configured to continuously acquire image data of the current vehicle passing through the road type identified by the visual perception model when the current vehicle enters the road type;

[0049] a data sending module, configured to upload the collected image data of the road type to a model server corresponding to the visual perception model, so that the model server updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model;

[0050] an updating module, configured to receive updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model;

[0051] The image acquisition module is also used to receive the current image data;

[0052] The road type determination module is used to input the current image data into the updated visual perception model to obtain the road type on which the vehicle is currently located.

[0053] On the other hand, this document provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for determining the type of vehicle driving road as described above.

[0054] On the other hand, this document provides an electronic device comprising at least one processor and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the at least one processor implements the method for determining the type of road on which the vehicle is traveling as described above by executing the instructions stored in the memory.

[0055] By adopting the above-mentioned technical solution, a method, device, equipment and medium for vehicle driving under unprotected left-turn conditions in this article are provided. When the vehicle drives to the road type identified by the visual perception model, the collected image data of the road type is uploaded to the model server corresponding to the visual perception model, so that the model server can update the model parameters of the visual perception model according to the received image data of the road type, and send the updated model parameters to the vehicle equipped with the visual perception model; receive the updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model; in actual application, the collected current image data can be input into the updated visual perception model, which can accurately identify the road type on which the current vehicle is located, ensure that the data volume of the model input sample in a specific scenario is sufficient, optimize the parameters in the corresponding model, and ensure that the model recognition scenarios are more extensive and more accurate in actual application.

[0056] In order to make the above and other purposes, features and advantages of this article more obvious and easy to understand, the following specifically lists preferred embodiments and provides detailed descriptions in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of this article or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of this article. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0058] Figure 1 A schematic diagram showing the steps of a method for determining the type of road a vehicle is traveling on provided in an embodiment of this document;

[0059] Figure 2 A schematic diagram showing the steps of another method for determining the type of road a vehicle is traveling on provided in an embodiment of this document;

[0060] Figure 3 A schematic diagram showing the steps of another method for determining the type of road on which a vehicle is traveling provided in an embodiment of this invention;

[0061] Figure 4 A schematic diagram showing the steps of another method for determining the type of road on which a vehicle is traveling provided in an embodiment of this invention;

[0062] Figure 5 A schematic diagram showing the steps of yet another method for determining the type of road on which a vehicle is traveling provided in an embodiment of this invention;

[0063] Figure 6 A schematic structural diagram of a device for determining the type of road a vehicle is traveling on according to an embodiment of the present invention is shown;

[0064] Figure 7 A schematic diagram of the structure of an electronic device of a device for determining the type of road a vehicle is traveling on provided in an embodiment of this document is shown. DETAILED DESCRIPTION

[0065] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of this document. Obviously, the embodiments described are only part of the embodiments of this document, not all of the embodiments. Based on the embodiments of this document, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this document.

[0066] It should be noted that the terms "first," "second," and the like in the specification and claims herein and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0067] This article mainly focuses on visual perception algorithms, which can identify the road type on which a vehicle is traveling. The road type can be relatively rare road types such as on- and off-ramps of highways, culvert sections under highways, tunnels, and river / sea bridges.

[0068] In order to solve the above problems, the embodiment of this article provides a method for determining the type of road a vehicle is traveling on. Figure 1FIG. 1 is a schematic diagram showing the steps of a method for determining the type of road a vehicle is traveling on provided in an embodiment of the present invention. Figure 1 As shown, the method can be applied to a vehicle provided with a visual perception model, and the visual perception model can be used to identify the type of road the vehicle is currently traveling on. The method may include the following steps:

[0069] S102: Obtain the current vehicle location information.

[0070] Specifically, location information can be determined using a navigation map or a GPS device. This information can be used to identify the vehicle's current location, such as the type of road the vehicle is traveling on, including on- and off-ramps, culverts under highways, tunnels, and river / sea bridges.

[0071] S104. Determine, based on the position information, whether the current vehicle is entering a road type identified by the visual perception model.

[0072] Specifically, the visual perception model may be a pre-trained model, the input of the visual perception model may be image data of the road ahead captured by the current road, and the output of the visual perception model may be the road type corresponding to the image data.

[0073] S106. When the current vehicle enters the road type identified by the visual perception model, continuously collect image data of the current vehicle passing through the road type.

[0074] Specifically, when it is determined through position information that the current vehicle is about to enter a road type identified by the visual perception model, image data of the current vehicle passing through the road type can be continuously captured by an image capture device installed on the vehicle.

[0075] It is understood that the image acquisition device can capture the front image data of the current vehicle in real time, where the front image data can be stored in the current vehicle in the form of a stack. When the current vehicle is about to enter a road type identified by the visual perception model, a corresponding upload tag can be added to the captured front image data. When the current vehicle is about to exit the road type identified by the visual perception model, the image data captured while passing through the lane of the road type can be uploaded to the corresponding model server storing the visual perception model. The visual perception model in the model server is the same as the visual perception model in the current vehicle.

[0076] In some possible embodiments, the type of road the vehicle is currently on may also be determined by using position information and a visual perception model to collect image data.

[0077] S108, uploading the collected image data of the road type to the model server corresponding to the visual perception model, so that the model server updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model.

[0078] Specifically, the visual perception model in the server may include multiple model parameters. The model parameters may represent constraints and may be set based on expert experience, such as: what road type the image data corresponds to, and the rules for determining the road type based on image features in the image data. The visual perception model can be trained using historical image data. The historical image data may include image features, traffic signs, and road features corresponding to the road type (traffic signs can be determined using methods such as expert experience and mathematical fitting). The image features, traffic signs, and road features corresponding to the road type in the historical image data are used as input to the visual perception model, and the road type corresponding to the historical image data is used as output. The model parameters in the visual perception model are continuously adjusted until the visual perception model meets preset requirements, such as meeting a preset accuracy and the number of model parameter adjustments meets a preset number of requirements. Model training is then completed. Specifically, machine learning algorithms such as the GBDT (Gradient Boosting Decision Tree) algorithm can be used for model training.

[0079] It is understandable that the model parameters in the visual perception model corresponding to different road types are different. Different image data and corresponding road types are used for training. The specific training process can be found in the above-mentioned visual perception model training process.

[0080] When the model server receives a certain amount of image data of road types uploaded by different vehicles, it can train the visual perception model based on the received image data of the road type and the corresponding road model. After the training is completed, the model parameters of the visual perception model will be updated, and the model server can push the updated model parameters to the vehicle equipped with the visual perception model.

[0081] S110. Receive updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model.

[0082] S112: Receive and collect current image data.

[0083] S114: Input the current image data into the updated visual perception model to obtain the road type on which the vehicle is currently located.

[0084] Specifically, in actual applications, the collected current image data can be input into the updated visual perception model, which can accurately identify the type of road the current vehicle is on, ensure that the data volume of the model input samples in specific scenarios is sufficient, optimize the parameters in the corresponding model, and ensure that the model recognition scenarios are wider and more accurate in actual applications.

[0085] Based on the above embodiment, in one embodiment of this specification, Figure 2 FIG. 1 is a schematic diagram showing the steps of another method for determining the type of road a vehicle is traveling on provided in an embodiment of the present invention. Figure 2 As shown, the method further includes:

[0086] S202: Collect trajectory information of the current vehicle passing through the road type.

[0087] Specifically, the trajectory information can be determined based on the navigation map in the vehicle or the vehicle's posture, orientation, etc. The specific collection equipment can be the output power / torque of the engine / motor, the throttle / electric switch force, the steering wheel direction, and the driving speed. Sensors corresponding to the collected information can be used.

[0088] S204. Upload the image data of the road type and the trajectory information to the map server that constructs the map, so that the map server updates the map image data corresponding to the location information according to the received image data of the road type and the trajectory information, and sends the updated map image data to the navigation device in the vehicle.

[0089] Specifically, the map server stores historically constructed maps, which may contain corresponding road information. The road information may include corresponding road type annotations, such as ramps, highways, bridges, etc. When the received image data and trajectory information of a road type are not marked in the current map or the marked road is unclear, the map server may update the map image data in the current map based on the image data of the road type and the trajectory information.

[0090] S206: Receive the map image data sent by the map server, and update the map in the current vehicle based on the map image data to obtain a vehicle updated map.

[0091] S208: Acquire current location information of the current vehicle, and determine the type of road the current vehicle is on based on the current location information and the vehicle updated map.

[0092] Specifically, in actual application, when a vehicle travels to a certain location in the vehicle update map, it can be determined that the road type currently traveled by the vehicle is the road type marked in the vehicle update map, which can ensure the timely update of the navigation map. There is no need to deliberately collect corresponding road information in areas that are not marked or updated in the navigation map, thereby improving user favorability.

[0093] Based on the above embodiment, in one embodiment of this specification, the method further includes:

[0094] Obtain distance information of fixed obstacles in the image data.

[0095] Specifically, the distance information of fixed obstacles can be collected by devices such as lidar, millimeter-wave radar, GPS, IMU, RTK, CAN, etc., and the collected distance information of fixed obstacles can be annotated into the image data to represent the actual distance of the fixed obstacles in the image data.

[0096] Accordingly, uploading the collected image data of the road type and the trajectory information to a map server for constructing a map includes:

[0097] The collected image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data are uploaded to the map server for constructing the map, so that the map server updates the map corresponding to the location information based on the received image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data.

[0098] Specifically, when updating the map, the map server can identify the width of the corresponding road and the number of lanes on the road through the distance information of the fixed obstacles in the received image data, and mark the identified information such as the width of the corresponding road and the number of lanes on the road in the updated map to achieve accurate positioning of the vehicle.

[0099] Based on the above embodiment, in one embodiment of this specification, Figure 3 FIG. 1 shows a schematic diagram of the steps of another method for determining the type of road a vehicle is traveling on provided in an embodiment of the present invention. Figure 3 As shown, uploading the collected image data of the road type to the model server corresponding to the visual perception model includes:

[0100] S402: Encrypt and compress the collected image data of the road type based on a pre-configured encryption compression model to obtain encrypted image data.

[0101] Specifically, the encryption compression model can be used to encrypt and compress the collected image data of the road type. The specific encryption method and compression method are not specifically limited in the embodiments of this specification and can be set according to actual needs. The encryption compression model can ensure the safe upload of vehicle-collected data and improve the security and efficiency of data transmission.

[0102] S404: Upload the encrypted image data to a model server corresponding to the visual perception model.

[0103] Specifically, encrypted image data can be transmitted to the model server via the Telematics BOX, or in-vehicle T-BOX for short. The Internet of Vehicles system consists of four parts: the host, the in-vehicle T-BOX, the mobile phone APP, and the backend system.

[0104] It is understandable that all data uploaded by the vehicle can be uploaded to the corresponding server via the on-board T-BOX.

[0105] On the other hand, this paper provides a method for determining the type of road a vehicle is traveling on. Figure 4 FIG. 1 shows a schematic diagram of the steps of another method for determining the type of road a vehicle is traveling on provided in an embodiment of the present invention. Figure 4 As shown, the method for determining the type of road on which a vehicle is traveling can be applied to the above-mentioned model server, and the method includes:

[0106] S502, receiving image data corresponding to the same road type;

[0107] S504: When the number of received image data corresponding to the same road type reaches a preset number requirement for the visual perception model input, the image data is used as the input of a pre-built visual perception model, and the road type corresponding to the image data is used as the output of the visual perception model. The visual perception model is trained until the model parameters in the visual perception model reach the preset requirements, thereby obtaining updated model parameters.

[0108] Specifically, the preset quantity requirement may be the minimum number of samples that the visual perception model can perform training.

[0109] S506: Send the updated model parameters to the vehicle equipped with the visual perception model.

[0110] Specifically, the model server can send the updated model parameters to all vehicles equipped with the visual perception model to ensure that all historical versions of the visual perception model are updated.

[0111] Based on the above embodiment, in one embodiment of this specification, the training of the visual perception model includes:

[0112] Image features in the image data corresponding to the road type are identified.

[0113] Specifically, the image features may be road shoulders, lane lines, lane indicator lines, road signs, display signs, and other feature categories that can represent the road type.

[0114] Filtering out the image data having the same image features to obtain target image data;

[0115] Specifically, since the collected adjacent image data have the same image features, when the adjacent image data have the same image features, the relatively unclear image data can be filtered out, and only the clear image data is retained, and the clear image data is used as the target image data.

[0116] All image features in the target image data are marked according to feature categories of different image features.

[0117] When training the visual perception model, all the image features in the target image data can be marked as the input of the visual perception model, and the corresponding road type can be used as the output of the visual perception model. The specific training method can be referred to the description above and will not be repeated here.

[0118] Based on the above embodiment, in one embodiment of this specification, the receiving of image data corresponding to the same road type includes:

[0119] receiving encrypted image data corresponding to the same road type;

[0120] The encrypted image data corresponding to the same road type is decrypted and decompressed based on a pre-configured decryption and decompression model to obtain image data corresponding to the same road type.

[0121] Specifically, the pre-configured decryption and decompression model may correspond to the above-mentioned encryption and compression model.

[0122] The method for determining the type of road a vehicle is traveling on provided in the embodiments of this specification provides a variety of image data with marked road types by different vehicles, thereby achieving a sufficient number of training samples for the visual perception model, improving the convergence of the visual perception model, and improving the application scenarios of the visual perception model and the accuracy of the identified road types.

[0123] On the other hand, this paper provides a method for determining the type of road a vehicle is traveling on. Figure 5 FIG. 1 shows a schematic diagram of steps of yet another method for determining the type of road on which a vehicle is traveling provided in an embodiment of the present invention. Figure 5 As shown, the execution subject of the method may be a map server, and the method includes:

[0124] S802, receiving image data and trajectory information of a road type, wherein the trajectory information carries position information;

[0125] S804: Calibrate and stitch the image data based on the trajectory information to obtain position image data corresponding to the position information.

[0126] Based on the above embodiment, in one embodiment of this specification, the method further includes:

[0127] receiving distance information of fixed obstacles in the image data;

[0128] The calibrating and stitching the image data based on the trajectory information to obtain position image data corresponding to the position information includes:

[0129] The image data is calibrated and spliced ​​based on the trajectory information and the distance information of the fixed obstacles in the image data to obtain position image data corresponding to the position information.

[0130] Specifically, the map server can calibrate the received image data according to the trajectory information, and splice the calibrated image data into a whole to obtain position image data corresponding to the position information. The position image data may include feature categories such as shoulders, lane lines, lane indicator lines, road signs, display signs, etc. in the image data that can represent the road type, and identify the relative distance corresponding to the feature category in the image to improve the accuracy of the map.

[0131] S806: Compare the position image data with map image data in a pre-stored map, and the position information of the map image data is the same as the position information of the position image data.

[0132] Specifically, the map server stores historically constructed maps, which may contain corresponding road information. The road information may include corresponding road type annotations, such as ramps, highways, bridges, etc. When the received image data and trajectory information of a road type are not marked in the current map or the marked road is unclear, the map server may update the map image data in the current map based on the image data of the road type and the trajectory information.

[0133] S808: When the accuracy of the position image data is higher than the accuracy of the map image data, update the map image data in the pre-stored map based on the position image data to obtain updated map image data.

[0134] S810: Send the updated map image data to a navigation device in a vehicle, so that the navigation device updates a map in the navigation device based on the updated map image data.

[0135] Specifically, the map server can send the updated map image data to the navigation device in the vehicle. The actual application scenario of the updated map can be: when the vehicle travels to a certain location in the vehicle's updated map, it can be determined that the road type currently traveled by the vehicle is the road type marked in the vehicle's updated map, which can ensure the timely update of the navigation map. There is no need to deliberately collect corresponding road information for areas that are not marked or updated in the navigation map, thereby improving user favorability.

[0136] It is understandable that the map server and the model server can be in the same server, and the map server and the model server are only used to distinguish different types of data. The specific data has been explained above and will not be repeated here.

[0137] On the other hand, this invention provides a device for determining the type of road a vehicle is traveling on. Figure 6 FIG. 1 shows a schematic diagram of a structure of a device for determining the type of road a vehicle is traveling on in an embodiment of the present invention. Figure 6 As shown, the device includes:

[0138] Position information acquisition module 11, used to obtain the current vehicle position information;

[0139] A judgment module 12 is used to judge whether the current vehicle enters a road type identified by the visual perception model based on the position information;

[0140] An image acquisition module 13 is configured to continuously acquire image data of the current vehicle passing through the road type identified by the visual perception model when the current vehicle enters the road type;

[0141] a data sending module 14, configured to upload the collected image data of the road type to a model server corresponding to the visual perception model, so that the model server updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model;

[0142] An updating module 15 is configured to receive updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model;

[0143] The image acquisition module 13 is further configured to receive and acquire current image data;

[0144] The road type determination module 16 is configured to input the current image data into the updated visual perception model to obtain the road type on which the vehicle is currently located.

[0145] On the other hand, an embodiment of the present specification provides a computer-readable storage medium, wherein the storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the method for determining the type of vehicle driving road as described above.

[0146] On the other hand, an embodiment of the present specification provides an electronic device for determining the type of road a vehicle is traveling on. Figure 7 FIG. 1 is a schematic diagram showing the structure of an electronic device of a device for determining the type of road a vehicle is traveling on, as provided in an embodiment of the present invention. Figure 7 As shown, the device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the method for determining the type of vehicle driving road as described in any of the above.

[0147] In another aspect, the present application provides a vehicle, the vehicle being provided with an autonomous driving system, the autonomous driving system comprising a device for determining a type of road on which the vehicle is traveling;

[0148] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between the various embodiments can be referred to in conjunction with each other. The testing method provided in the embodiments of the present invention has the same implementation principles and technical effects as the aforementioned system embodiments. For the sake of brevity, any parts not mentioned in the method embodiments can be referred to the corresponding contents in the aforementioned system embodiments.

[0149] It should be understood that in the various embodiments of this document, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0150] It should also be understood that in the embodiments herein, the term "and / or" merely describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" could represent: A alone, A and B simultaneously, or B alone. Furthermore, the character " / " in this document generally indicates an "or" relationship between the associated objects.

[0151] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this document.

[0152] Those skilled in the art will clearly understand that, for the sake of convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0153] In the several embodiments provided herein, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices, or units, or can be an electrical, mechanical, or other form of connection.

[0154] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments herein.

[0155] In addition, the functional units in the various embodiments herein may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0156] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this article is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this article. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0157] This article uses specific embodiments to illustrate the principles and implementation methods of this article. The description of the above embodiments is only used to help understand the methods and core ideas of this article. At the same time, for those skilled in the art, based on the ideas of this article, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation to this article.

Claims

1. A method for determining the type of road a vehicle is traveling on, characterized in that: The method comprises: Get the current vehicle location information; Determining whether the current vehicle is entering a road type identified by the visual perception model based on the position information; When the current vehicle enters a road type identified by the visual perception model, continuously collecting image data of the current vehicle passing through the road type; Uploading the collected image data of the road type to a model server corresponding to the visual perception model, so that when the model server receives a preset number of image data of the same road type uploaded by different vehicles, it updates the model parameters of the visual perception model based on the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model; the model parameters of the visual perception model corresponding to different road types are different; receiving updated model parameters returned by the model server, and updating the visual perception model according to the updated model parameters to obtain an updated visual perception model; Receive and collect current image data; The current image data is input into the updated visual perception model to obtain the road type on which the vehicle is currently located.

2. The method for determining the type of road a vehicle is traveling on according to claim 1, characterized in that: The method further comprises: Collecting trajectory information of the current vehicle passing through the road type; Uploading the image data of the road type and the trajectory information to a map server for constructing a map, so that the map server updates the map image data corresponding to the location information according to the received image data of the road type and the trajectory information, and sends the updated map image data to a navigation device in the vehicle; receiving the map image data sent by the map server, and updating the map in the current vehicle based on the map image data to obtain an updated vehicle map; The current position information of the current vehicle is obtained, and the type of road on which the current vehicle is located is determined based on the current position information and the vehicle updated map.

3. The method for determining the type of road a vehicle is traveling on according to claim 2, characterized in that: The method further comprises: Obtaining distance information of fixed obstacles in the image data; Accordingly, uploading the image data of the road type and the trajectory information to a map server for constructing a map includes: The collected image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data are uploaded to the map server for constructing the map, so that the map server updates the map corresponding to the location information based on the received image data of the road type, the trajectory information, and the distance information of the fixed obstacles in the image data.

4. The method for determining the type of road a vehicle is traveling on according to claim 1, wherein: The uploading of the collected image data of the road type to the model server corresponding to the visual perception model includes: Encrypting and compressing the collected image data of the road type based on a pre-configured encryption compression model to obtain encrypted image data; The encrypted image data is uploaded to the model server corresponding to the visual perception model.

5. A method for determining the type of road a vehicle is traveling on, characterized in that: The method comprises: receiving image data corresponding to the same road type; the image data being image data continuously collected by the vehicle as the vehicle enters the road type identified by the visual perception model; and determining, based on the acquired position information, whether the vehicle has entered the road type identified by the visual perception model; When the number of image data corresponding to the same road type uploaded by different vehicles reaches a preset number requirement for the visual perception model input, the image data is used as the input of a pre-built visual perception model, and the road type corresponding to the image data is used as the output of the visual perception model. The visual perception model is trained until the model parameters in the visual perception model reach the preset requirements, thereby obtaining updated model parameters; the model parameters in the visual perception model corresponding to different road types are different; The updated model parameters are sent to a vehicle equipped with the visual perception model; the vehicle receives the updated model parameters returned by the model server, and updates the visual perception model according to the updated model parameters to obtain an updated visual perception model; the current image data is received and collected; the current image data is input into the updated visual perception model to obtain the type of road on which the vehicle is currently located.

6. The method for determining the type of road a vehicle is traveling on according to claim 5, characterized in that: The training of the visual perception model includes: identifying image features in the image data corresponding to the road type; Filtering out the image data having the same image features to obtain target image data; All image features in the target image data are marked according to feature categories of different image features.

7. The method for determining the type of road a vehicle is traveling on according to claim 5, characterized in that: The receiving image data corresponding to the same road type includes: receiving encrypted image data corresponding to the same road type; The encrypted image data corresponding to the same road type is decrypted and decompressed based on a pre-configured decryption and decompression model to obtain image data corresponding to the same road type.

8. A method for determining the type of road a vehicle is traveling on, characterized in that: The method comprises: Receive image data and trajectory information of a road type, wherein the trajectory information carries location information; send the location information to a vehicle so that the vehicle obtains the location information of the current vehicle; determine whether the current vehicle enters a road type identified by a visual perception model based on the location information; when the current vehicle enters a road type identified by the visual perception model, continuously collect image data of the current vehicle passing through the road type; upload the collected image data of the road type to a model server corresponding to the visual perception model, so that when the model server receives a preset number of image data of the same road type uploaded by different vehicles, it updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model; the model parameters of the visual perception model corresponding to different road types are different; receive the updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model; receive the collected current image data; input the current image data into the updated visual perception model to obtain the road type on which the current vehicle is located; calibrating and splicing the image data based on the trajectory information to obtain position image data corresponding to the position information; comparing the position image data with map image data in a pre-stored map, where the position information of the map image data is the same as the position information of the position image data; When the accuracy of the position image data is higher than the accuracy of the map image data, updating the map image data in the pre-stored map based on the position image data to obtain updated map image data; The updated map image data is transmitted to a navigation device in a vehicle, so that the navigation device updates a map in the navigation device based on the updated map image data.

9. The method for determining the type of road a vehicle is traveling on according to claim 8, characterized in that: The method further comprises: receiving distance information of fixed obstacles in the image data; The calibrating and stitching the image data based on the trajectory information to obtain position image data corresponding to the position information includes: The image data is calibrated and spliced ​​based on the trajectory information and the distance information of the fixed obstacles in the image data to obtain position image data corresponding to the position information.

10. A device for determining the type of road a vehicle is traveling on, characterized in that: The device comprises: Position information acquisition module, used to obtain the current vehicle's position information; a judgment module, configured to judge, based on the position information, whether the current vehicle is entering a road type identified by the visual perception model; an image acquisition module, configured to continuously acquire image data of the current vehicle passing through the road type identified by the visual perception model when the current vehicle enters the road type; a data sending module, configured to upload the collected image data of the road type to a model server corresponding to the visual perception model, so that when the model server receives a preset number of image data of the same road type uploaded by different vehicles, it updates the model parameters of the visual perception model according to the received image data of the road type, and sends the updated model parameters to the vehicle equipped with the visual perception model; an updating module, configured to receive updated model parameters returned by the model server, and update the visual perception model according to the updated model parameters to obtain an updated visual perception model; different road types correspond to different model parameters in the visual perception model; The image acquisition module is also used to receive the current image data; The road type determination module is used to input the current image data into the updated visual perception model to obtain the road type on which the vehicle is currently located.

Images