Radar calibration data adjustment method and system, vehicle and storage medium

Abstract

The invention discloses a radar calibration data adjustment method and system, a vehicle and a storage medium. The method can be at least applied to the field of vehicles and comprises the steps of collecting image information of a target vehicle on a current driving road; identifying the image information to obtain an identification result; current calibration data corresponding to the recognition result are searched from a target database, the target database comprises mapping relations between attribute information of different driving roads and different calibration data, and the different calibration data comprise the current calibration data; the current calibration data is used for assisting the target radar to detect whether an obstacle exists in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; the historical calibration data of the target radar is adjusted to be the current calibration data, and the historical calibration data is used for assisting the target radar to detect whether the target vehicle has an obstacle in the historical driving process or not. According to the invention, the technical problem of high frequency of radar misinformation is solved.

Description

Technical Field

[0001] The present invention relates to the field of vehicles, and in particular to a radar calibration data adjustment method, system, vehicle and storage medium. Background Art

[0002] At present, during the driving process of a vehicle, when the vehicle drives on a bumpy road or a gravel road, due to the echo of ultrasonic waves irradiating the ground, some parking radars often falsely report obstacles, resulting in a technical problem of a high frequency of radar false alarms.

[0003] With respect to the technical problem of the high frequency of false alarms caused by the above-mentioned radar, no effective solution has been proposed so far. Summary of the invention

[0004] Embodiments of the present invention provide a radar calibration data adjustment method, system, vehicle and storage medium to at least solve the technical problem of high frequency of radar false alarms.

[0005] According to one aspect of an embodiment of the present invention, a method for adjusting calibration data of a radar is provided. The method may include: collecting image information of a target vehicle on a current driving road; identifying the image information to obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; searching for current calibration data corresponding to the recognition result from a target database, wherein the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, and the different calibration data include current calibration data, and the current calibration data is used to assist the target radar in detecting whether there are obstacles in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; adjusting the historical calibration data of the target radar to the current calibration data, wherein the historical calibration data is used to assist the target radar in detecting whether there are obstacles in the historical driving process of the target vehicle.

[0006] Optionally, the image information is recognized to obtain a recognition result, including: according to a transmission protocol, the image information is input into a target recognition model, wherein the target recognition model is obtained by training an initial recognition model based on image information samples of vehicle samples on the driving road; using the target recognition model, image features are extracted from the image information, and the image features are recognized to obtain a recognition result.

[0007] Optionally, the method also includes: in response to an adjustment instruction, adjusting current parameters of the target recognition model to target parameters, wherein the target parameters are determined based on attribute information of the current driving road and attribute information of different driving roads, and the target parameters are used to make the accuracy corresponding to the recognition result output by the adjusted target recognition model greater than the accuracy corresponding to the recognition result output by the target recognition model before adjustment; using the target recognition model, extracting image features from image information, and identifying the image features to obtain a recognition result, including: inputting the image information into the adjusted target recognition model according to a transmission protocol; using the adjusted target recognition model, extracting image features from the image information, and identifying the image features to obtain a recognition result.

[0008] Optionally, the method also includes: determining historical calibration data, including: collecting historical image information of the target vehicle on the historical driving road, wherein the time period of the target vehicle on the historical driving road is earlier than the time period of the target vehicle on the current driving road; inputting the historical image information into the target recognition model according to the transmission protocol; using the target recognition model, extracting historical image features from the historical image information, and identifying the historical image features to obtain a historical recognition result, wherein the historical recognition result is used to at least represent the attribute information of the historical driving road where the target vehicle is located; and searching the target database for historical calibration data corresponding to the historical recognition result.

[0009] Optionally, searching the target database for current calibration data corresponding to the recognition result includes: in response to the attribute information of different driving roads including the attribute information of the current driving road, searching the target database for current calibration data whose attribute information of the current driving road satisfies the mapping relationship.

[0010] Optionally, adjusting the historical calibration data of the target radar to the current calibration data includes: in response to the current calibration data being different from the historical calibration data, adjusting the historical calibration data to the current calibration data.

[0011] According to another aspect of an embodiment of the present invention, a method for adjusting calibration data of a radar is also provided. The method may include: obtaining calibration data of a target radar, wherein the target radar is deployed on a target vehicle, the calibration data of the target radar is historical calibration data or current calibration data, the historical calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during historical driving, the current calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during current driving, the current calibration data is calibration data corresponding to the recognition result found from a target database, the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the recognition result is obtained by recognizing image information of the target vehicle on the current driving road, and the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; based on the calibration data, detecting whether there are obstacles in the driving process of the target vehicle.

[0012] According to another aspect of an embodiment of the present invention, a radar calibration data adjustment system is also provided. The system may include: a control end and a target radar, wherein the control end is used to collect image information of a target vehicle on a current driving road; recognize the image information to obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; search the current calibration data corresponding to the recognition result from the target database, wherein the target database includes a mapping relationship between the attribute information of different driving roads and different calibration data, the calibration data includes current calibration data, and the current calibration data is used to assist the target radar in detecting whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; the target radar is used to use historical calibration data to detect whether there is an obstacle in the historical driving process of the target vehicle, and use current calibration data to detect whether there is an obstacle in the current driving process of the target vehicle.

[0013] According to another aspect of an embodiment of the present invention, a calibration data adjustment device for a radar is also provided. The device may include: a collection unit for collecting image information of a target vehicle on a current driving road; a recognition unit for recognizing the image information to obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; a search unit for searching the current calibration data corresponding to the recognition result from a target database, wherein the target database includes a mapping relationship between the attribute information of different driving roads and different calibration data, and the different calibration data include current calibration data, and the current calibration data is used to assist the target radar in detecting whether there are obstacles in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; an adjustment unit for adjusting the historical calibration data of the target radar to the current calibration data, wherein the historical calibration data is used to assist the target radar in detecting whether there are obstacles in the historical driving process of the target vehicle.

[0014] Optionally, the recognition unit includes: an input module, used to input the image information into a target recognition model according to a transmission protocol, wherein the target recognition model is obtained by training an initial recognition model based on image information samples of vehicle samples on the driving road; a recognition module, used to use the target recognition model to extract image features from the image information, and recognize the image features to obtain a recognition result.

[0015] Optionally, the device may also include: a response unit, used to adjust the current parameters of the target recognition model to target parameters in response to an adjustment instruction, wherein the target parameters are determined based on the attribute information of the current driving road and the attribute information of different driving roads, and the target parameters are used to make the accuracy corresponding to the recognition result output by the adjusted target recognition model greater than the accuracy corresponding to the recognition result output by the target recognition model before the adjustment; the recognition module may include: an input submodule, used to input the image information into the adjusted target recognition model according to the transmission protocol; and a recognition submodule, used to use the adjusted target recognition model to extract image features from the image information, and recognize the image features to obtain a recognition result.

[0016] Optionally, the device may also include: a determination unit, used to determine historical calibration data, including: a first acquisition module, used to collect historical image information of the target vehicle on the historical driving road, wherein the time period of the target vehicle on the historical driving road is earlier than the time period of the target vehicle on the current driving road; a first input module, used to input the historical image information into the target recognition model according to the transmission protocol; a first recognition module, used to use the target recognition model to extract historical image features from the historical image information, and recognize the historical image features to obtain a historical recognition result, wherein the historical recognition result is used to at least represent the attribute information of the historical driving road where the target vehicle is located; a first search module, used to search for historical calibration data corresponding to the historical recognition result from the target database.

[0017] Optionally, the search unit includes: a second search module for searching, from a target database, current calibration data whose attribute information of the current driving road satisfies a mapping relationship in response to the attribute information of different driving roads including the attribute information of the current driving road.

[0018] Optionally, the adjustment unit includes: an adjustment module, configured to adjust the historical calibration data to the current calibration data in response to the current calibration data being different from the historical calibration data.

[0019] According to another aspect of an embodiment of the present invention, a calibration data adjustment device for a radar is also provided. The device may include: an acquisition unit, used to acquire calibration data of a target radar, wherein the target radar is deployed on a target vehicle, and the calibration data of the target radar is historical calibration data or current calibration data, the historical calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during historical driving, the current calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during current driving, the current calibration data is calibration data corresponding to the recognition result found from a target database, the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the recognition result is obtained by recognizing image information of the target vehicle on the current driving road, and the recognition result is used to at least indicate the attribute information of the current driving road where the target vehicle is located; a detection unit, used to acquire based on the calibration data, and detect whether there are obstacles in the driving process of the target vehicle.

[0020] According to another aspect of an embodiment of the present invention, a vehicle is provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes any one of the above methods when running.

[0021] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein when the executable program runs, the device where the storage medium is located is controlled to execute any one of the above methods.

[0022] According to another aspect of an embodiment of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, any one of the above methods is implemented.

[0023] In an embodiment of the present invention, when adjusting the calibration data of the radar, image information of the target vehicle on the current driving road can be collected. The collected image information is identified to obtain a recognition result, for example, the attribute information of the current driving road where the target vehicle is located can be obtained. From the target database, the current calibration data corresponding to the above recognition result can be found, and the historical calibration data of the target radar is adjusted to the current calibration data corresponding to the above recognition result. Since in the embodiment of the present application, the collected image information of the target vehicle on the current driving road is identified to obtain a recognition result, and the historical calibration data of the target radar is adjusted to the current calibration data corresponding to the above recognition result found from the target database, the purpose of avoiding some parking radars from falsely reporting obstacles is achieved, thereby solving the technical problem of the high frequency of radar false alarms, and further achieving the technical effect of increasing the frequency of radar false alarms. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0025] FIG1( a ) is a schematic diagram of an application scenario of a radar calibration data adjustment method according to an embodiment of the present invention;

[0026] FIG1( b ) is a flow chart of a radar calibration data adjustment method according to an embodiment of the present invention;

[0027] FIG2( a ) is a schematic diagram of an application scenario of another radar calibration data adjustment method according to an embodiment of the present invention;

[0028] FIG2( b ) is a flow chart of another radar calibration data adjustment method according to an embodiment of the present invention;

[0029] FIG3( a ) is a schematic diagram of a detection area of ​​an ultrasonic probe according to the related art;

[0030] FIG3( b ) is a schematic diagram of a real-time switching system for ultrasonic thresholds according to an embodiment of the present invention;

[0031] FIG3( c ) is a schematic diagram of data communication between an ultrasonic threshold calibration database and an ultrasonic threshold switching unit according to an embodiment of the present invention;

[0032] FIG3( d ) is a schematic diagram of a road surface condition of a driving road according to an embodiment of the present invention;

[0033] FIG3( e ) is a flow chart of a method for switching an ultrasonic threshold according to an embodiment of the present invention;

[0034] FIG3( f ) is a schematic diagram of road surface information and threshold setting according to an embodiment of the present invention;

[0035] FIG3( g) is a schematic diagram of an installation position of a radar according to an embodiment of the present invention;

[0036] FIG4( a ) is a structural block diagram of a radar calibration data adjustment system according to an embodiment of the present invention;

[0037] FIG4( b ) is a structural block diagram of a radar calibration data adjustment device according to an embodiment of the present invention;

[0038] FIG4( c ) is a structural block diagram of another radar calibration data adjustment device according to an embodiment of the present invention;

[0039] Figure 5 is a structural block diagram of a vehicle according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged 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" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] According to an embodiment of the present invention, a method for adjusting calibration data of a radar is provided.

[0043] As an optional implementation, the calibration data adjustment method of the radar can be applied to, but not limited to, the application scenario shown in FIG1 (a). FIG1 (a) is a schematic diagram of an application scenario of a calibration data adjustment method of a radar according to an embodiment of the present invention. As shown in FIG1 (a), in the application scenario, the mobile terminal 10 can communicate with the server 13 through the network 11, but not limited to, and the server 13 can perform operations on the database, such as writing data operations or reading data operations. The mobile terminal 10 can be a terminal device, and the terminal device can include, but not limited to, a human-computer interaction screen, a processor, and a memory. The human-computer interaction screen can be used to display a virtual machine on the mobile terminal 10, but not limited to. The vehicle 12 can be used to respond to the human-computer interaction operation, perform corresponding operations, or generate corresponding instructions, and send the generated instructions to the server 13. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be performed in a different order than that here.

[0044] FIG1( b ) is a flow chart of a radar calibration data adjustment method according to an embodiment of the present invention. As shown in FIG1( b ), the method may include the following steps:

[0045] Step S112, collecting image information of the target vehicle on the current driving road.

[0046] In the technical solution provided in the above step S112 of the present invention, the above image information can be used to display the driving environment of the current driving road where the target vehicle is located. The driving environment can be characterized by at least one of the following information: the congestion level of the current driving road, the road surface information of the current driving road, and the indication information of the current driving road, etc., which are only given as examples and are not specifically limited.

[0047] In this embodiment, the congestion level of the current driving road may include: the density of vehicles and pedestrians in the current driving road. The road surface information of the current driving road may include: the road surface flatness of the current driving road, the road surface friction of the current driving road, and the road surface material characteristics of the current driving road. The indication information of the current driving road may include: the traffic signal light status of the current driving road and the traffic police command status of the current driving road, which are only given as examples and are not specifically limited.

[0048] In this embodiment, image information of the target vehicle on the current driving road is collected. Optionally, this embodiment uses a surround view device to take surround shots of the current driving road where the target vehicle is located, thereby collecting image information of the target vehicle on the current driving road. In other words, the driving environment of the current driving road where the target vehicle is located can be collected. Among them, the surround view device can be but is not limited to: at least one surround view camera (Around View Monitor, referred to as AVM), which is only used as an example here and is not specifically limited.

[0049] Step S114, identifying the image information to obtain a recognition result.

[0050] In the technical solution provided in the above step S114 of the present invention, the above recognition result can be used to at least represent the attribute information of the current driving road where the target vehicle is located. For example, the attribute information of the current driving road can be the road surface information of the current driving road, and the road surface information of the current driving road can include: the road surface flatness of the current driving road, the road surface friction of the current driving road, and the road surface material characteristics of the current driving road, etc., which are only given as examples and are not specifically limited.

[0051] In this embodiment, after collecting image information of the target vehicle on the current driving road, the image information is recognized to obtain a recognition result. Optionally, based on the collected image information, this embodiment extracts features from the collected image information to obtain image features, and then recognizes the image features to obtain a recognition result. In other words, at least the attribute information of the current driving road where the target vehicle is located can be obtained, which is only an example and not specifically limited here.

[0052] Optionally, the above-mentioned road surface flatness can be represented by 0x0, 0x1, and 0x2. If the number corresponding to the road surface flatness is smaller, it means that the road surface of the current driving road is smoother. For example, if the road surface flatness of the first current driving road is recorded as 0x1, and the road surface flatness of the second current driving road is recorded as 0x0, then the road surface flatness of the second current driving road is better than the road surface flatness of the first current driving road. Among them, a road surface flatness of 0x0 can indicate a flat road surface, a road surface flatness of 0x1 can indicate a slightly uneven road surface, and a road surface flatness of 0x2 can indicate a severely uneven road surface. This is only an example for illustration and is not specifically limited.

[0053] Optionally, the road friction force can be represented by 0x0, 0x1, 0x2, 0x3, 0x4 and 0x5. If the road friction force is smaller, it means that the road friction force of the current driving road is greater. For example, if the road friction force of the first current driving road is recorded as 0x0, and the road friction force of the second current driving road is recorded as 0x5, the road friction force of the second current driving road is less than the road friction force of the first current driving road. Among them, a road friction of 0x0 can represent a dry road surface, that is, the road friction is large; a road friction of 0x1 can indicate a wet road surface, that is, the road friction is less than the friction of a dry road surface; a road friction of 0x2 can indicate a puddle of water on the road surface, that is, the road friction is less than the friction of a wet road surface; a road friction of 0x3 can indicate new snow on the road surface, that is, the road friction is less than the friction of a road surface with puddles of water; a road friction of 0x4 can indicate melted snow on the road surface, that is, the road friction is less than the friction of a road surface with new snow; a road friction of 0x5 can indicate ice on the road surface, that is, the road friction is less than the friction of a road surface with melted snow. These are only examples for illustration and are not specifically limited.

[0054] Optionally, the above-mentioned pavement material characteristics can be represented by 0x0, 0x1, and 0x2. If the number corresponding to the pavement material characteristics is smaller, it means that the pavement of the current driving road is easier for vehicles to drive on. For example, if the pavement material characteristics of the first current driving road are recorded as 0x1, and the pavement material characteristics of the second current driving road are recorded as 0x0, then the pavement of the second current driving road is easier for vehicles to drive on than the pavement of the first current driving road. Among them, the pavement material characteristics of 0x0 can represent an asphalt road, the pavement material characteristics of 0x1 can represent an ecological brick pavement, and the pavement material characteristics of 0x2 can represent a gravel road. This is only an example for illustration and is not specifically limited.

[0055] Step S116, searching the target database for current calibration data corresponding to the recognition result.

[0056] In the technical solution provided in step S116 of the present invention, the target database may include mapping relationships between attribute information of different driving roads and different calibration data. For example, the target database may be an ultrasonic threshold calibration database, which is only used as an example and is not specifically limited.

[0057] In this embodiment, the above-mentioned different calibration data may include current calibration data, and the current calibration data may be used to assist the target radar in detecting whether there is an obstacle in the current driving process of the target vehicle. The target radar may be deployed on the target vehicle.

[0058] In this embodiment, the above-mentioned current calibration data may be a detection threshold for a target radar to detect whether there is an obstacle during the current driving process of the target vehicle. The above-mentioned target radar may be implemented based on an ultrasonic sensor, which is only used as an example here without specific limitation.

[0059] In this embodiment, after the image information is recognized and the recognition result is obtained, the current calibration data corresponding to the recognition result is searched from the target database. Optionally, based on the recognition result, according to the mapping relationship between the attribute information of different driving roads and different calibration data stored in the target database, the current calibration data whose attribute information of the current driving road where the target vehicle is located satisfies the above mapping relationship can be searched.

[0060] It should be noted that the above method of searching for the current calibration data corresponding to the recognition result from the target database is only an example and is not specifically limited here. As long as the process and method can search for the current calibration data corresponding to the recognition result from the target database, it is within the protection scope of the embodiments of the present application and will not be described one by one here.

[0061] Step S118, adjusting the historical calibration data of the target radar to the current calibration data.

[0062] In the technical solution provided in the above step S118 of the present invention, the above historical calibration data can be used to assist the target radar in detecting whether there are obstacles in the historical driving process of the target vehicle.

[0063] In this embodiment, after searching the target database for the current calibration data corresponding to the recognition result, the historical calibration data of the target radar is adjusted to the current calibration data. Optionally, this embodiment judges the relationship between the historical calibration data of the target radar and the current calibration data found on the basis of finding the current calibration data corresponding to the recognition result. According to the judgment result obtained that the current calibration data found is different from the historical calibration data of the target radar, the historical calibration data of the target radar can be adjusted to the current calibration data found. Among them, the above judgment result can be that the current calibration data found is the same as the historical calibration data of the target radar, or it can be that the current calibration data found is different from the historical calibration data of the target radar. This is only used as an example and is not specifically limited.

[0064] It should be noted that the calibration data adjustment method of the radar in the present application can also be applied to at least the following scenarios: autonomous driving scenario, assisted driving scenario and passive driving scenario (which can be referred to as human driving scenario). Among them, the autonomous driving scenario can be used to indicate that the vehicle control system controls the driving of the vehicle during the driving process, and the driving object does not need to maintain control and supervision of the driving. The assisted driving scenario can be used to indicate that the vehicle control system provides assistance during the driving process, but the driving object still needs to maintain control and supervision of the driving. The passive driving scenario can be used to indicate the scenario where the driving object controls the vehicle to complete the driving operation.

[0065] In the artificial intelligence AI driving of autonomous vehicles, commuting mode usually refers to a driving mode designed for commuting or daily commuting. Among them, the AI ​​driving can collect data by learning the user's driving route and driving behavior. When all the learning is completed, the collected data is screened and checked, and then the full segment of data after screening and checking is automatically transmitted back to the cloud server to execute the construction of the cloud map. The next time you drive and choose the same route, the AI ​​driving will be activated, and the vehicle will be driven to the destination of the route according to the constructed cloud map. Among them, matching and complementing the offline map with the constructed cloud map can improve the accuracy of the map.

[0066] In the above steps S112 to S118 of the present application, when the same route as the last time you drove is chosen for this driving, the AI ​​driver can be activated to drive the vehicle according to the acquired navigation information. During the driving process of the vehicle, the AI ​​driver can collect image information of the target vehicle on the current road, and recognize the collected image information to obtain a recognition result, for example, the attribute information of the current road where the target vehicle is located can be obtained. The AI ​​driver can search the target database for the current calibration data corresponding to the above recognition result, thereby using the AI ​​driver to adjust the historical calibration data of the target radar to the current calibration data corresponding to the above recognition result.

[0067] In the above steps S112 to S118 of the present application, when adjusting the calibration data of the radar, the image information of the target vehicle on the current driving road can be collected. The above image information collected can be identified to obtain a recognition result, for example, the attribute information of the current driving road where the target vehicle is located can be obtained. From the target database, the current calibration data corresponding to the above recognition result can be found, and the historical calibration data of the target radar can be adjusted to the current calibration data corresponding to the above recognition result. Because in the embodiment of the present application, when the target vehicle travels to a pothole road or a gravel road, the image information of the target vehicle collected on the current driving road can be identified, and the recognition result can be obtained, and the historical calibration data of the target radar can be adjusted to the current calibration data corresponding to the above recognition result found from the target database, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, thereby solving the technical problem of the high frequency of radar false alarms, and then achieving the technical effect of increasing the frequency of radar false alarms.

[0068] The above method of this embodiment is further introduced below.

[0069] As an optional implementation method, step S114, identifying the image information to obtain a recognition result, includes: inputting the image information into a target recognition model according to a transmission protocol; using the target recognition model to extract image features from the image information, and identifying the image features to obtain a recognition result.

[0070] In this embodiment, the target recognition model may be obtained by training an initial recognition model based on image information samples of vehicle samples on the driving road.

[0071] In this embodiment, the vehicle samples may include vehicles of different models and styles. For example, the vehicle samples may include sedans, sports cars, and sports utility vehicles (SUVs), which are only used as examples and are not specifically limited.

[0072] In this embodiment, the above-mentioned image information samples may include road surface datasets of different sizes. For example, the above-mentioned image information samples may include: a road surface clustering dataset (Road Surface Classification Dataset, RSCD for short), and the RSCD dataset may include: a training set, a test set, and a validation set. The training set may include 960,000 road surface images, the test set may include 50,000 road surface images, and the validation set may include 20,000 road surface images. This is only an example and is not specifically limited.

[0073] In this embodiment, the transmission protocol can be used to transmit the collected image information to the target recognition model. For example, the transmission protocol can be a communication transmission protocol (such as a DSI3 protocol), which is only used as an example and is not specifically limited.

[0074] In this embodiment, after collecting image information of the target vehicle on the current driving road, the image information is input into the target recognition model according to the transmission protocol. Optionally, this embodiment, based on the collected image information, inputs the collected image information into the target recognition model obtained by training the initial recognition model based on the image information samples of the vehicle samples on the driving road according to the transmission protocol, thereby achieving the purpose of using the target recognition model and realizing the technical effect of improving the accuracy of determining the attribute information of the current driving road.

[0075] In this embodiment, after the image information is input into the target recognition model according to the transmission protocol, the target recognition model is used to extract image features from the image information, and the image features are recognized to obtain a recognition result. Optionally, this embodiment uses the extraction layer of the trained target recognition model to extract features from the collected image information on the basis of inputting the image information into the target recognition model to obtain image features, and uses the recognition layer of the trained target recognition model to recognize the obtained image features to obtain a recognition result. In other words, at least the attribute information of the current driving road where the target vehicle is located can be obtained, thereby achieving the purpose of using the target recognition model for image recognition and realizing the technical effect of improving the accuracy of determining the attribute information of the current driving road.

[0076] The method for recognizing image information and obtaining a recognition result in this embodiment is further described below.

[0077] As an optional implementation mode, the method also includes: in response to an adjustment instruction, adjusting the current parameters of the target recognition model to the target parameters; using the target recognition model, extracting image features from the image information, and identifying the image features to obtain a recognition result, including: according to the transmission protocol, inputting the image information into the adjusted target recognition model; using the adjusted target recognition model, extracting image features from the image information, and identifying the image features to obtain a recognition result.

[0078] In this embodiment, the above current parameters may include at least one of the following hyperparameters: learning rate, batch size, number of iterations, etc., which are only used as examples here without specific limitation.

[0079] In this embodiment, the target parameters can be determined based on the attribute information of the current driving road and the attribute information of different driving roads. Optionally, if the attribute information of different driving roads does not include the attribute information of the current driving road, it means that the attribute information of the current driving road represented by the recognition result has errors, that is, the recognition result output by the target recognition model before adjustment has errors. Therefore, it is necessary to determine the target parameters based on the attribute information of different driving roads and the errors in the attribute information of the current driving road.

[0080] In this embodiment, the above target parameters can be used to make the accuracy corresponding to the recognition result output by the adjusted target recognition model greater than the accuracy corresponding to the recognition result output by the target recognition model before the adjustment.

[0081] In this embodiment, the target recognition model after adjustment may be the target recognition model after adjusting parameters, and the target recognition model before adjustment may be the target recognition model before adjusting parameters.

[0082] In this embodiment, in response to the adjustment instruction, the current parameters of the target recognition model are adjusted to the target parameters. Optionally, based on the determination of the current parameters of the target recognition model, in response to the adjustment instruction to adjust the current parameters, the embodiment can adjust the determined current parameters of the target recognition model to the target parameters, thereby achieving the purpose of optimizing the target recognition model and realizing the technical effect of improving the accuracy of determining the attribute information of the current driving road.

[0083] In this embodiment, after adjusting the current parameters of the target recognition model to the target parameters in response to the adjustment instruction, the image information is input into the adjusted target recognition model according to the transmission protocol. Optionally, this embodiment inputs the collected image information into the target recognition model after the parameters are adjusted according to the transmission protocol on the basis of adjusting the current parameters of the target recognition model to the target parameters, thereby achieving the purpose of using the optimized target recognition model and realizing the technical effect of improving the accuracy of determining the attribute information of the current driving road.

[0084] In this embodiment, after the image information is input into the adjusted target recognition model according to the transmission protocol, the image features are extracted from the image information using the adjusted target recognition model, and the image features are recognized to obtain a recognition result. Optionally, this embodiment, based on inputting the collected image information into the target recognition model after adjusting the parameters, uses the extraction layer of the target recognition model after adjusting the parameters to extract features from the collected image information to obtain image features, and uses the recognition layer of the target recognition model after adjusting the parameters to recognize the obtained image features to obtain a recognition result. In other words, at least the attribute information of the current driving road where the target vehicle is located can be obtained, thereby achieving the purpose of using the optimized target recognition model for image recognition and realizing the technical effect of improving the accuracy of determining the attribute information of the current driving road.

[0085] The following further describes the method for adjusting the calibration data of the radar in this embodiment.

[0086] As an optional implementation method, the method also includes: determining historical calibration data, including: collecting historical image information of the target vehicle on the historical driving road; inputting the historical image information into the target recognition model according to the transmission protocol; using the target recognition model, extracting historical image features from the historical image information, and identifying the historical image features to obtain historical recognition results; searching for historical calibration data corresponding to the historical recognition results from the target database.

[0087] In this embodiment, the time period of the target vehicle on the historical driving road may be earlier than the time period of the target vehicle on the current driving road.

[0088] In this embodiment, the above historical recognition result can be used to at least represent the attribute information of the historical driving road where the target vehicle is located. For example, the attribute information of the historical driving road can be the road surface information of the historical driving road, and the road surface information of the historical driving road can include: the road surface flatness of the historical driving road, the road surface friction of the historical driving road, and the road surface material characteristics of the historical driving road, etc., which are only given as examples and are not specifically limited.

[0089] In this embodiment, historical image information of the target vehicle on the historical driving road is collected. Optionally, this embodiment uses a surround view device to take surround shots of the historical driving road where the target vehicle is located, thereby collecting historical image information of the target vehicle on the historical driving road. In other words, the driving environment of the historical driving road where the target vehicle is located can be collected, thereby achieving the purpose of determining the driving environment of the historical driving road, and realizing the technical effect of improving the accuracy of collecting image information of the target vehicle on the driving road.

[0090] In this embodiment, after collecting historical image information of the target vehicle on the historical driving road, the historical image information is input into the target recognition model according to the transmission protocol. Optionally, this embodiment, based on the historical image information collected, inputs the collected historical image information into the target recognition model obtained by training the initial recognition model based on the image information samples of the vehicle samples on the driving road according to the transmission protocol, thereby achieving the purpose of using the target recognition model and realizing the technical effect of improving the accuracy of determining the attribute information of the historical driving road.

[0091] In this embodiment, after the historical image information is input into the target recognition model according to the transmission protocol, the target recognition model is used to extract historical image features from the historical image information, and the historical image features are identified to obtain historical recognition results. Optionally, this embodiment, based on the input of the historical image information into the target recognition model, uses the extraction layer of the trained target recognition model to extract features from the collected historical image information to obtain historical image features, and uses the recognition layer of the trained target recognition model to identify the obtained historical image features to obtain historical recognition results. In other words, at least the attribute information of the historical driving road where the target vehicle is located can be obtained, thereby achieving the purpose of using the target recognition model for image recognition and realizing the technical effect of improving the accuracy of determining the attribute information of the historical driving road.

[0092] In this embodiment, after extracting historical image features from historical image information using a target recognition model and identifying the historical image features to obtain historical recognition results, historical calibration data corresponding to the historical recognition results are searched from the target database. Optionally, based on the historical recognition results, this embodiment can search for historical calibration data whose attribute information of the historical driving road where the target vehicle is located satisfies the above mapping relationship according to the mapping relationship between the attribute information of different driving roads and different calibration data stored in the target database, thereby achieving the purpose of determining the attribute information of the historical driving road where the target vehicle is located, and realizing the technical effect of improving the efficiency of determining the attribute information of the historical driving road where the target vehicle is located.

[0093] The method of searching the target database for current calibration data corresponding to the recognition result in this embodiment is further described below.

[0094] As an optional implementation method, step S116, searching the target database for current calibration data corresponding to the recognition result, includes: in response to the attribute information of different driving roads including the attribute information of the current driving road, searching the target database for current calibration data whose attribute information of the current driving road satisfies the mapping relationship.

[0095] In this embodiment, after the image information is recognized and the recognition result is obtained, in response to the fact that the attribute information of different driving roads includes the attribute information of the current driving road, the current calibration data whose attribute information of the current driving road satisfies the mapping relationship is searched from the target database. Optionally, on the basis of obtaining the recognition result, this embodiment judges the relationship between the attribute information of the current driving road represented by the recognition result and the attribute information of different driving roads. If it is judged that the attribute information of different driving roads includes the attribute information of the current driving road, the current calibration data whose attribute information of the current driving road satisfies the above mapping relationship can be searched according to the mapping relationship between the attribute information of different driving roads and different calibration data stored in the target database, thereby achieving the purpose of determining the attribute information of the historical driving road where the target vehicle is located, and realizing the technical effect of improving the efficiency of determining the attribute information of the historical driving road where the target vehicle is located.

[0096] The following further describes the method for adjusting the historical calibration data of the target radar to the current calibration data in this embodiment.

[0097] As an optional implementation manner, step S118, adjusting the historical calibration data of the target radar to the current calibration data, includes: in response to the current calibration data being different from the historical calibration data, adjusting the historical calibration data to the current calibration data.

[0098] In this embodiment, after searching the target database for the current calibration data corresponding to the recognition result, in response to the current calibration data being different from the historical calibration data, the historical calibration data is adjusted to the current calibration data. Optionally, this embodiment determines the relationship between the current calibration data and the historical calibration data on the basis of finding the current calibration data corresponding to the recognition result. If the judgment result is that the current calibration data is different from the historical calibration data, the historical calibration data can be adjusted to the current calibration data, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, and further achieving the technical effect of increasing the frequency of false alarms of the radar.

[0099] Optionally, if the current calibration data found is 5.2, and the historical calibration data of the target radar is 4.2, the relationship between the current calibration data found and the historical calibration data is judged, and the judgment result is that the current calibration data found is different from the historical calibration data. Therefore, the historical calibration data of the target radar can be adjusted from 4.2 to the current calibration data of 5.2. This is only an example for illustration and is not specifically limited.

[0100] According to an embodiment of the present invention, another radar calibration data adjustment method is provided.

[0101] As an optional implementation, the calibration data adjustment method of the radar can be applied to, but not limited to, the application scenario shown in FIG2. FIG2 (a) is a schematic diagram of an application scenario of another calibration data adjustment method of a radar according to an embodiment of the present invention. As shown in FIG2 (a), in the application scenario, the mobile terminal 20 can communicate with the server 23 through the network 21, but not limited to, and the server 23 can perform operations on the database, such as writing data operations or reading data operations. The mobile terminal 20 can be a terminal device, and the terminal device can include, but is not limited to, a human-computer interaction screen, a processor, and a memory. The human-computer interaction screen can be used to display a virtual machine on the mobile terminal 20, but not limited to. The vehicle 22 can be used to respond to the human-computer interaction operation, perform a corresponding operation, or generate a corresponding instruction, and send the generated instruction to the server 23.

[0102] FIG2( b ) is a flow chart of another radar calibration data adjustment method according to an embodiment of the present invention. As shown in FIG2( b ), the method may include the following steps:

[0103] Step S212, obtaining calibration data of the target radar, wherein the target radar is deployed on the target vehicle, and the calibration data of the target radar is historical calibration data or current calibration data. The historical calibration data is used to assist the target radar in detecting whether there are obstacles in the historical driving process of the target vehicle, and the current calibration data is used to assist the target radar in detecting whether there are obstacles in the current driving process of the target vehicle. The current calibration data is calibration data corresponding to the recognition result found from the target database, and the target database includes a mapping relationship between attribute information of different driving roads and different calibration data. The recognition result is obtained by recognizing image information of the target vehicle on the current driving road, and the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located.

[0104] In the technical solution provided in the above step S212 of the present invention, the above current calibration data may be calibration data corresponding to the recognition result found from the target database. The target database may include a mapping relationship between the attribute information of different driving roads and different calibration data. The above recognition result may be obtained by recognizing the image information of the target vehicle on the current driving road, and the above recognition result may be used to at least represent the attribute information of the current driving road where the target vehicle is located. For example, the attribute information of the current driving road may be the road surface information of the current driving road, and the road surface information of the current driving road may include: the road surface flatness of the current driving road, the road surface friction of the current driving road, and the road surface material characteristics of the current driving road, etc., which are only given as examples here and are not specifically limited.

[0105] In this embodiment, the calibration data of the target radar is obtained. Optionally, this embodiment can obtain the calibration data corresponding to the recognition result from the cloud, that is, the historical calibration data of the target radar or the current calibration data of the target radar can be obtained.

[0106] Step S214: based on the calibration data, detect whether there is an obstacle while the target vehicle is driving.

[0107] In the technical solution provided in the above step S214 of the present invention, after the calibration data of the target radar is obtained, it is detected whether there is an obstacle in the driving process of the target vehicle based on the calibration data. Optionally, if the calibration data of the target radar obtained in this embodiment is historical calibration data, the obtained historical calibration data can be used to detect whether there is an obstacle in the historical driving process of the target vehicle. Alternatively, if the calibration data of the target radar obtained is current calibration data, the obtained current calibration data can be used to detect whether there is an obstacle in the current driving process of the target vehicle.

[0108] In the above steps S212 to S214 of the present application, when adjusting the calibration data of the radar, the calibration data of the target radar can be obtained, and according to the obtained calibration data, whether there is an obstacle in the driving process of the target vehicle can be detected. Because in the embodiment of the present application, when the target vehicle is driving on a pothole road or a gravel road, the obtained calibration data can be used to detect whether there is an obstacle in the driving process of the target vehicle, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, thereby solving the technical problem of high frequency of radar false alarms, and further achieving the technical effect of increasing the frequency of radar false alarms.

[0109] The technical solution of the embodiment of the present invention is illustrated below in conjunction with preferred implementation modes.

[0110] At present, during the driving process of a vehicle, when the vehicle drives on a pothole road or a gravel road, due to the echo of the ultrasonic wave irradiating the ground, some parking radars often generate false alarms of obstacles, thereby causing a high frequency of false alarms of the radar. For example, FIG3 (a) is a schematic diagram of a detection area of ​​an ultrasonic probe according to the related art. As shown in FIG3 (a), when an ultrasonic probe is used to detect an obstacle, within the field of view (FOV) of the ultrasonic probe, the echo generated by the ground to the ultrasonic probe will affect the echo generated by the obstacle to the ultrasonic probe, thereby causing a high frequency of false alarms of the radar.

[0111] However, an embodiment of the present invention proposes a radar calibration data adjustment method, which can distinguish the surrounding environment by collecting image information, thereby switching the calibration data of the target radar according to different surrounding environments, thereby achieving the purpose of avoiding the phenomenon of some parking radars falsely reporting obstacles, thereby solving the technical problem of high frequency of radar false alarms, and achieving the technical effect of reducing the frequency of radar false alarms.

[0112] In this embodiment, through the ultrasonic threshold real-time switching system, when the target vehicle travels to a pothole road or a gravel road, the collected image information of the target vehicle on the current driving road can be identified, and the identification result can be obtained, and the historical calibration data of the target radar can be adjusted to the current calibration data corresponding to the above identification result found from the target database. For example, FIG3 (b) is a schematic diagram of an ultrasonic threshold real-time switching system according to an embodiment of the present invention. As shown in FIG3 (b), the system may include: a surround view camera 311, an automatic driving controller 312, a threshold switching unit 313, an ultrasonic data processing unit 314 and an ultrasonic sensor 315.

[0113] In this embodiment, the surround view camera 311 may include multiple fisheye cameras, and the surround view camera 311 may be used to send the captured images to the automatic driving controller 312. For example, the surround view camera 311 may be 4 fisheye cameras, and the 4 fisheye cameras may be arranged at the front side, the rear side, the left side and the right side of the vehicle respectively, and the 4 fisheye cameras may be used for 360-degree surround view of the vehicle to assist parking, etc., thereby achieving the purpose of assisting the ultrasonic radar to avoid obstacles.

[0114] In this embodiment, the autonomous driving controller 312 may be integrated into the domain controller unit of the vehicle. A trained target recognition model may be integrated into the autonomous driving controller 312. The trained target recognition model may be used to receive pictures taken by the AVM, recognize the pictures, obtain the road surface information of the current driving road where the target vehicle is located, and send the obtained road surface information to the threshold switching unit 313.

[0115] In this embodiment, the threshold switching unit 313 may be integrated into a domain controller unit of the vehicle. The threshold switching unit 313 may be used to receive an input signal of road surface information from the automatic driving controller 312, thereby determining whether to send a threshold switching instruction to different ultrasonic data processing units 314.

[0116] In this embodiment, if it is determined to send an instruction for switching the threshold to a different ultrasound data processing unit 314 , one of the 54 sets of threshold calibration data stored internally is sent to the ultrasound data processing unit 314 at the same time.

[0117] In this embodiment, the ultrasonic data processing unit 314 can be integrated into the ultrasonic sensor 315. The ultrasonic data processing unit 314 can be used to process the ultrasonic raw data and communicate with the vehicle. When the threshold is set, the ultrasonic data processing unit 314 can be used to receive the threshold from the threshold switching unit 313 and write the threshold into the ultrasonic internal software. When the radar works next time, the latest threshold written is used for distance detection warning.

[0118] In this embodiment, the ultrasonic sensor 315 can be installed on the front bumper, rear bumper or wheel arch of the vehicle to provide obstacle distance for the automatic parking radar. The ultrasonic sensor 315 can be installed on the automatic door to provide obstacle information around the door. In the ultrasonic sensor, thresholds are set for different echo intensities, that is, the intensity of the echo at different times is set to perform effective obstacle detection.

[0119] In this embodiment, the calibration data of the target radar can be switched through data communication between the ultrasonic threshold calibration database and the ultrasonic threshold switching unit. For example, FIG3 (c) is a schematic diagram of data communication between an ultrasonic threshold calibration database and an ultrasonic threshold switching unit according to an embodiment of the present invention. As shown in FIG3 (c), through the control area network (Controller Area Network, referred to as CAN), according to the mapping relationship between the road surface information of different driving roads and different threshold calibration data stored in the ultrasonic threshold calibration database 321, the road surface information of the current driving road where the target vehicle is located can be searched for the threshold calibration data that satisfies the above mapping relationship. In other words, the road surface roughness signal of the current driving road, the road surface friction signal of the current driving road, and the road surface material signal of the current driving road can be searched for the threshold calibration data that jointly satisfies the above mapping relationship. The threshold calibration data found can be sent to the threshold switching unit 322 through the local interconnect network (LocalInterconnect Network, referred to as LIN).

[0120] In this embodiment, the road condition of the driving road on which the target vehicle is located may include the road condition shown in FIG3(d), which is a schematic diagram of a road condition of a driving road according to an embodiment of the present invention. The road condition corresponding to FIG3(d)-(a) is a dry asphalt-smooth road condition, the road condition corresponding to FIG3(d)-(b) is a dry asphalt-mild road condition, the road condition corresponding to FIG3(d)-(c) is a dry asphalt-severe road condition, the road condition corresponding to FIG3(d)-(d) is a water asphalt-severe road condition, and the road condition corresponding to FIG3(d)-(e) is a ) corresponds to wet asphalt - mild pavement, Figure 3 (d) - (f) corresponds to wet concrete - smooth pavement, Figure 3 (d) - (g) corresponds to wet concrete - severe pavement, Figure 3 (d) - (h) corresponds to water concrete - mild pavement, Figure 3 (d) - (i) corresponds to cement slurry pavement, Figure 3 (d) - (j) corresponds to dry gravel pavement, Figure 3 (d) - (k) corresponds to a pavement with melted snow, and Figure 3 (d) - (l) corresponds to an icy pavement.

[0121] In this embodiment, through the ultrasonic threshold switching method, when the target vehicle travels to a pothole road or a gravel road, the collected image information of the target vehicle on the current driving road can be identified, and the identification result can be obtained, and the historical calibration data of the target radar can be adjusted to the current calibration data corresponding to the above identification result found in the target database. For example, Figure 3 (e) is a flow chart of a method for switching ultrasonic thresholds according to an embodiment of the present invention, and the method may include the following steps:

[0122] Step S332, obtaining the RSCD data set.

[0123] In the technical solution provided in the above step S332 of the present invention, the RSCD data set may include: a training set, a test set and a validation set. The training set contains 960,000 road images, the test set contains 50,000 road images, and the validation set contains 20,000 road images. This is only used as an example and is not specifically limited.

[0124] After the RSCD data set is obtained, the process proceeds to step S334, where the initial recognition model is trained using the training set in the obtained RSCD data set to obtain a trained initial recognition model.

[0125] After the trained initial recognition model is obtained, the process proceeds to step S336 to verify the initial recognition result output by the trained initial recognition model using the verification set in the RSCD data set.

[0126] In the technical solution provided in the above step S336 of the present invention, while verifying the initial recognition result output by the trained initial recognition model, data related to the scene can be supplemented for the partial scene where the vehicle is located.

[0127] After verifying the initial recognition result output by the trained initial recognition model, the process proceeds to step S338 to determine the verified initial recognition model as the target recognition model, thereby distinguishing different road surface information.

[0128] After verifying the initial recognition result output by the trained initial recognition model, proceed to step S340 to collect pictures using AVM.

[0129] In the technical solution provided in step S340 of the present invention, the AVM can be used to collect environmental information around the vehicle. The AVM can be used to transmit the collected environmental information (for example, road surface information) in the form of pictures to the automatic driving controller in real time through protocols such as DSI3.

[0130] After collecting road surface information using AVM, the process proceeds to step S342, where the collected image is input into the target recognition model for recognition, thereby obtaining road surface information of the current driving road where the target vehicle is located.

[0131] In the technical solution provided in the above step S342 of the present invention, the above automatic driving controller can output the above road surface information to the threshold switching unit through model recognition.

[0132] After obtaining the road surface information of the current driving road where the target vehicle is located, the process proceeds to step S344 to search for threshold calibration data corresponding to the road surface information from the target database.

[0133] In the technical solution provided in the above step S344 of the present invention, the mature site of the car factory is used to calibrate and collect the threshold value. Generally, at least 300 sets of data are collected in the same site, thereby achieving the purpose of ensuring the authenticity of the calibration data. At the same time, it is necessary to perform secondary functional verification on the threshold value. After the calibrated threshold value is input into the ultrasonic radar, different types of obstacles are actually detected, such as water barriers, ground locks and polyvinyl chloride pipes (PVC for short). For example, FIG3 (f) is a schematic diagram of road surface information and threshold setting according to an embodiment of the present invention. As shown in FIG3 (f), different threshold values ​​can be set according to different road surface information actually collected. For example, according to a gravel road, a threshold value corresponding to a gravel road can be set, according to an asphalt road, a threshold value corresponding to an asphalt road can be set, and according to an ecological brick road, a threshold value corresponding to an ecological brick road can be set.

[0134] In this embodiment, the target database may store the mapping relationship between the ground information of different driving roads and different calibration data in the form of Table 1 below.

[0135] Table 1 Mapping relationship table of target database

[0136]

[0137] In this embodiment, FIG3 (g) is a schematic diagram of the installation position of a radar according to an embodiment of the present invention. As shown in FIG3 (g), ultrasonic radars 351 to 354 are ultrasonic radars installed in the front direction of the vehicle, AVM35#1 can be a surround view camera installed in the front direction of the vehicle, parking ultrasonic radars 355 to 358 can be ultrasonic radars installed in the rear direction of the vehicle, AVM35#2 can be a surround view camera installed in the rear direction of the vehicle, ultrasonic radars 359 to 3510 can be ultrasonic radars installed in the left side of the vehicle, AVM35#3 can be a surround view camera installed in the left side of the vehicle, ultrasonic radars 3511 to 3512 can be ultrasonic radars installed in the right side of the vehicle, and AVM35#4 can be a surround view camera installed in the right side of the vehicle. Among them, the surround view camera can cover the 360 ​​environment of the vehicle, thereby ensuring the real-time effectiveness of radar threshold switching for different areas.

[0138] In this embodiment, ultrasonic radars 351 to 354 can use road surface information obtained by recognizing pictures taken by AVM35#1, parking ultrasonic radars 355 to 358 can use road surface information obtained by recognizing pictures taken by AVM35#2, ultrasonic radars 359 to 3510 can use road surface information obtained by recognizing pictures taken by AVM35#3, and ultrasonic radars 3511 to 3512 can use road surface information obtained by recognizing pictures taken by AVM35#4.

[0139] In this embodiment, after using AVM to collect road surface information, the collected images are input into the target recognition model for identification, so that the road surface information of the current driving road where the target vehicle is located can be obtained, and the threshold calibration data corresponding to the road surface information can be found from the target database, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, thereby solving the technical problem of high frequency of radar false alarms, and achieving the technical effect of increasing the frequency of radar false alarms.

[0140] According to another aspect of an embodiment of the present invention, corresponding to the embodiment of the above-mentioned radar calibration data adjustment method, the present specification also provides a radar calibration data adjustment system. Figure 4 (a) is a structural block diagram of a radar calibration data adjustment system according to an embodiment of the present invention. As shown in Figure 4 (a), the radar calibration data adjustment system 400 may include: a control terminal 402 and a target radar 404.

[0141] The control terminal 402 is used to collect image information of the target vehicle on the current driving road; identify the image information to obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; search for current calibration data corresponding to the recognition result from the target database, wherein the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the calibration data includes current calibration data, and the current calibration data is used to assist the target radar to detect whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle.

[0142] The target radar 404 is used to detect whether there are obstacles in the target vehicle's historical driving process using historical calibration data, and to detect whether there are obstacles in the target vehicle's current driving process using current calibration data.

[0143] In the technical solution provided by the calibration data adjustment system 400 of the above-mentioned radar of the present invention, the mobile terminal is used to send a trigger instruction to the vehicle to be controlled. The vehicle is used to obtain the guidance movement data in response to the trigger instruction; based on the guidance movement data and the surrounding environment data recognized by the vehicle, the vehicle movement data is generated; the vehicle movement is controlled according to the vehicle movement data, and during the movement of the vehicle, the vehicle is controlled to update the recognized surrounding environment data, and the vehicle movement data is adjusted according to the updated surrounding environment data.

[0144] According to another aspect of an embodiment of the present invention, corresponding to the embodiment of the calibration data adjustment method of the above-mentioned radar, the present specification also provides a calibration data adjustment device for a radar. FIG4( b) is a structural block diagram of a calibration data adjustment device for a radar according to an embodiment of the present invention. As shown in FIG4( b), the calibration data adjustment device 410 for the radar may include: an acquisition unit 412, an identification unit 414, a search unit 416 and an adjustment unit 418.

[0145] The acquisition unit 412 is used to acquire image information of the target vehicle on the current driving road.

[0146] The recognition unit 414 is used to recognize the image information and obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located.

[0147] The search unit 416 is used to search for current calibration data corresponding to the recognition result from the target database, wherein the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the different calibration data include current calibration data, and the current calibration data is used to assist the target radar in detecting whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle.

[0148] The adjustment unit 418 is used to adjust the historical calibration data of the target radar to the current calibration data, wherein the historical calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during its historical driving process.

[0149] Optionally, the recognition unit 414 includes: an input module, used to input the image information into a target recognition model according to a transmission protocol, wherein the target recognition model is obtained by training an initial recognition model based on image information samples of vehicle samples on the driving road; and a recognition module, used to use the target recognition model to extract image features from the image information, and recognize the image features to obtain a recognition result.

[0150] Optionally, the device 410 may also include: a response unit, used to adjust the current parameters of the target recognition model to target parameters in response to an adjustment instruction, wherein the target parameters are determined based on the attribute information of the current driving road and the attribute information of different driving roads, and the target parameters are used to make the accuracy corresponding to the recognition result output by the adjusted target recognition model greater than the accuracy corresponding to the recognition result output by the target recognition model before the adjustment; the recognition module may include: an input submodule, used to input the image information into the adjusted target recognition model according to the transmission protocol; and a recognition submodule, used to use the adjusted target recognition model to extract image features from the image information, and recognize the image features to obtain a recognition result.

[0151] Optionally, the device may also include: a determination unit, used to determine historical calibration data, including: a first acquisition module, used to collect historical image information of the target vehicle on the historical driving road, wherein the time period of the target vehicle on the historical driving road is earlier than the time period of the target vehicle on the current driving road; a first input module, used to input the historical image information into the target recognition model according to the transmission protocol; a first recognition module, used to use the target recognition model to extract historical image features from the historical image information, and recognize the historical image features to obtain a historical recognition result, wherein the historical recognition result is used to at least represent the attribute information of the historical driving road where the target vehicle is located; a first search module, used to search for historical calibration data corresponding to the historical recognition result from the target database.

[0152] Optionally, the search unit 416 includes: a second search module for searching the target database for current calibration data whose attribute information of the current driving road satisfies the mapping relationship in response to the attribute information of different driving roads including the attribute information of the current driving road.

[0153] Optionally, the adjustment unit 418 includes: an adjustment module, configured to adjust the historical calibration data to the current calibration data in response to the current calibration data being different from the historical calibration data.

[0154] In this embodiment, the following units are provided in the calibration data adjustment device of the radar: a collection unit, which is used to collect image information of the target vehicle on the current driving road; a recognition unit, which is used to recognize the image information and obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; a search unit, which is used to search for current calibration data corresponding to the recognition result from a target database, wherein the target database includes a mapping relationship between the attribute information of different driving roads and different calibration data, and the different calibration data include current calibration data, and the current calibration data is used to assist the target radar in detecting whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; an adjustment unit, which is used to adjust the historical calibration data of the target radar to the current calibration data, wherein the historical calibration data is used to assist the target radar in detecting whether there is an obstacle in the historical driving process of the target vehicle, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, thereby solving the technical problem of high frequency of radar false alarms, and further achieving the technical effect of increasing the frequency of radar false alarms.

[0155] According to another aspect of an embodiment of the present invention, corresponding to the embodiment of the above-mentioned radar calibration data adjustment method, the present specification also provides a radar calibration data adjustment device. Figure 4 (c) is a structural block diagram of another radar calibration data adjustment device in an embodiment of the present invention. As shown in Figure 4 (c), the radar calibration data adjustment device 420 may include: an acquisition unit 422 and a detection unit 424.

[0156] The acquisition unit 422 is used to acquire calibration data of the target radar, wherein the target radar is deployed on the target vehicle, and the calibration data of the target radar is historical calibration data or current calibration data. The historical calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during historical driving. The current calibration data is used to assist the target radar in detecting whether there are obstacles in the target vehicle during current driving. The current calibration data is calibration data corresponding to the recognition result found from the target database. The target database includes a mapping relationship between attribute information of different driving roads and different calibration data. The recognition result is obtained by recognizing image information of the target vehicle on the current driving road. The recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located.

[0157] The detection unit 424 is used to obtain calibration data and detect whether there is an obstacle during the driving of the target vehicle.

[0158] In this embodiment, the following units are provided in the calibration data adjustment device of the radar: an acquisition unit, used for acquiring the calibration data of the target radar, wherein the target radar is deployed on the target vehicle, and the calibration data of the target radar is historical calibration data or current calibration data, the historical calibration data is used to assist the target radar in detecting whether there is an obstacle in the target vehicle during the historical driving process, and the current calibration data is used to assist the target radar in detecting whether there is an obstacle in the target vehicle during the current driving process, and the current calibration data is calibration data corresponding to the recognition result found from the target database, the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the recognition result is obtained by recognizing the image information of the target vehicle on the current driving road, and the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; a detection unit, used for acquiring based on the calibration data, detecting whether there is an obstacle in the driving process of the target vehicle, thereby achieving the purpose of avoiding some parking radars from falsely reporting obstacles, thereby solving the technical problem of high frequency of radar false alarms, and further achieving the technical effect of increasing the frequency of radar false alarms.

[0159] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium including a stored executable program, wherein when the executable program runs, the device where the storage medium is located is controlled to execute any one of the above methods.

[0160] According to another aspect of an embodiment of the present invention, a computer program product is provided. The computer program product includes a computer program. When the computer program is executed by a processor, any one of the above methods is implemented.

[0161] According to another aspect of an embodiment of the present invention, a vehicle is provided, including: a memory storing an executable program; and a processor running the program, wherein the program executes any one of the above methods when running.

[0162] Figure 5 is a structural block diagram of a vehicle according to an embodiment of the present invention. Figure 5 As shown, the components of the vehicle 500 include but are not limited to a memory 510 and a processor 520. The processor 520 is connected to the memory 510 via a bus 530, and the database 550 is used to store data.

[0163] The vehicle 500 may also include an access device 540 that enables the vehicle 500 to communicate via one or more networks 560. Examples of these networks include a Public Switched Telephone Network (PSTN), a Local Area Network (LAN), a Wide Area Network (WAN), a Personal Area Network (PAN), or a combination of communication networks such as the Internet. The access device 540 may include one or more of any type of network interface (e.g., a network interface controller (NIC)) of wired or wireless type, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a Worldwide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a Universal Serial Bus (USB) interface, a cellular network interface, a Bluetooth interface, a Near Field Communication (NFC) interface, and the like.

[0164] In one embodiment of the present disclosure, the above components of the vehicle 500 and Figure 5 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 5 The vehicle structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present disclosure. Those skilled in the art may add or replace other components as needed.

[0165] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0166] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0167] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units can be a logical function division, and there can be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0168] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0169] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0170] 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 the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the various embodiments of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, referred to as Read-Only Memory), random access memory (RAM, referred to as Random Access Memory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0171] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A radar calibration data adjustment method, characterized in that: include: Collect image information of the target vehicle on the current road; Recognize the image information to obtain a recognition result, wherein the recognition result is used to at least represent attribute information of the current driving road where the target vehicle is located; Searching for current calibration data corresponding to the recognition result from a target database, wherein the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the different calibration data include the current calibration data, the current calibration data is used to assist a target radar in detecting whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; The historical calibration data of the target radar is adjusted to the current calibration data, wherein the historical calibration data is used to assist the target radar in detecting whether there is an obstacle in the historical driving process of the target vehicle.

2. The method according to claim 1, characterized in that Recognizing the image information to obtain a recognition result includes: According to the transmission protocol, the image information is input into a target recognition model, wherein the target recognition model is obtained by training an initial recognition model based on image information samples of vehicle samples on the driving road; The target recognition model is used to extract image features from the image information, and the image features are recognized to obtain the recognition result.

3. The method according to claim 2, characterized in that The method further includes: in response to the adjustment instruction, adjusting the current parameter of the target recognition model to a target parameter, wherein the target parameter is determined based on the attribute information of the current driving road and the attribute information of the different driving roads, and the target parameter is used to make the accuracy corresponding to the recognition result output by the adjusted target recognition model greater than the accuracy corresponding to the recognition result output by the target recognition model before the adjustment; Utilizing the target recognition model, extracting image features from the image information, and recognizing the image features to obtain the recognition result, including: inputting the image information into the adjusted target recognition model according to the transmission protocol; The adjusted target recognition model is used to extract the image features from the image information, and the image features are recognized to obtain the recognition result.

4. The method according to claim 2, characterized in that: The method further comprises: Determining the historical calibration data includes: collecting historical image information of the target vehicle on a historical driving road, wherein a period of time when the target vehicle was on the historical driving road is earlier than a period of time when the target vehicle was on the current driving road; According to the transmission protocol, the historical image information is input into the target recognition model; Utilizing the target recognition model, extracting historical image features from the historical image information, and recognizing the historical image features to obtain a historical recognition result, wherein the historical recognition result is used to at least represent attribute information of the historical driving road where the target vehicle is located; The historical calibration data corresponding to the historical recognition result is searched from the target database.

5. The method according to claim 1, characterized in that Searching the target database for current calibration data corresponding to the recognition result includes: In response to the attribute information of the different driving roads including the attribute information of the current driving road, current calibration data whose attribute information of the current driving road satisfies the mapping relationship is searched from the target database.

6. The method according to any one of claims 1 to 5, characterized in that Adjusting the historical calibration data of the target radar to the current calibration data includes: In response to the current calibration data being different from the historical calibration data, the historical calibration data is adjusted to be the current calibration data.

7. A radar calibration data adjustment method, characterized in that: include: Acquire calibration data of a target radar, wherein the target radar is deployed on a target vehicle, the calibration data of the target radar is historical calibration data or current calibration data, the historical calibration data is used to assist the target radar in detecting whether there is an obstacle in a historical driving process of the target vehicle, the current calibration data is used to assist the target radar in detecting whether there is an obstacle in a current driving process of the target vehicle, the current calibration data is calibration data corresponding to a recognition result found from a target database, the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the recognition result is obtained by recognizing image information of the target vehicle on a current driving road, and the recognition result is used to at least represent attribute information of the current driving road where the target vehicle is located; Based on the calibration data, it is detected whether there is an obstacle during the driving of the target vehicle.

8. A radar calibration data adjustment system, characterized in that: include: Control terminal and target radar, among which, The control end is used to collect image information of the target vehicle on the current driving road; identify the image information to obtain a recognition result, wherein the recognition result is used to at least represent the attribute information of the current driving road where the target vehicle is located; search for current calibration data corresponding to the recognition result from a target database, wherein the target database includes a mapping relationship between attribute information of different driving roads and different calibration data, the calibration data includes the current calibration data, and the current calibration data is used to assist the target radar in detecting whether there is an obstacle in the current driving process of the target vehicle, and the target radar is deployed on the target vehicle; The target radar is used to detect whether there are obstacles in the target vehicle's historical driving process using historical calibration data, and to detect whether there are obstacles in the target vehicle's current driving process using current calibration data.

9. A vehicle, characterized in that: include: A memory storing a program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 7 when running.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is executed, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images