Vehicle data processing method, vehicle-mounted equipment, device, vehicle and storage medium

Abstract

The invention discloses a vehicle data processing method, vehicle-mounted equipment, a device, a vehicle and a storage medium. The method comprises the following steps: acquiring sensor data collected by a vehicle-mounted sensor of a target vehicle; acquiring first road condition information, detected by external equipment, of a target vehicle; according to the sensor data and the first road condition information, a training sample of a first road condition determination model is generated, and the first road condition determination model is used for determining the road condition information of the target vehicle according to the sensor data collected by the vehicle-mounted sensor. According to the embodiment of the invention, the sample label is generated based on the first road condition information, and the training sample of the first road condition determination model is generated according to the sensor data and the sample label, so that the sensor data is labeled by means of the road condition information detected by the external equipment, and the generation efficiency of the training sample is improved; and the training speed of the first road condition determination model can be improved.

Description

Technical Field

[0001] The present application relates to the field of electronic technology, and in particular to a vehicle data processing method, vehicle-mounted equipment, a device, a vehicle and a storage medium, wherein the storage medium is a computer-readable storage medium. Background Art

[0002] Data is the "fuel" that drives the development of automobile driving. With the connectivity of data, the car is no longer an information island. In the existing scheme of determining vehicle condition information through models, it is necessary to manually label the training data of the model and then train the model with the labeled data. The manual labeling method is too labor-intensive, resulting in inefficient training sample generation. Summary of the invention

[0003] Embodiments of the present application provide a vehicle data processing method, device, vehicle, and computer-readable storage medium, which can improve the efficiency of generating training samples.

[0004] In order to achieve the above object, according to a first aspect of the present application, a vehicle data processing method is provided, comprising:

[0005] Obtaining sensor data collected by onboard sensors of the target vehicle;

[0006] Acquire first road condition information of the target vehicle detected by an external device;

[0007] A training sample of a first road condition determination model is generated according to the sensor data and the first road condition information. The first road condition information is used to generate a sample label of the training sample. The first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data.

[0008] Optionally, a second road condition determination model is deployed on the target vehicle, and the first road condition determination model is used to update the second road condition determination model. Before generating a training sample of the first road condition determination model according to the sensor data and the first road condition information, the method further includes:

[0009] Performing data analysis and processing on the sensor data by using the second road condition determination model to obtain second road condition information;

[0010] Obtaining comparison information between the first road condition information and the second road condition information;

[0011] The step of generating a training sample of a first road condition determination model according to the sensor data and the first road condition information includes:

[0012] The training sample is generated according to the comparison information, the sensor data and the first road condition information.

[0013] Optionally, the comparison information includes one of a comparison result and a first collection time of the first road condition information, and the obtaining the comparison information of the first road condition information and the second road condition information includes:

[0014] Comparing the first road condition information and the second road condition information to obtain the comparison result;

[0015] or,

[0016] Sending the second traffic condition information to the external device, triggering the external device to compare the first traffic condition information and the second traffic condition information, and returning the first collection time of the first traffic condition information according to the comparison result obtained by the comparison process;

[0017] Receive the first collection time sent by the external device.

[0018] Optionally, when a comparison result between the first road condition information and the second road condition information is a target comparison result, the external device sends the first acquisition time to the target vehicle, and the generating the training sample according to the comparison information, the sensor data and the first road condition information includes:

[0019] If the comparison result is a target result, determining a target time period corresponding to the second collection time of the sensor data;

[0020] Acquire multiple sensor data collected by the vehicle-mounted device within a target time period, and acquire first road condition information corresponding to the multiple sensor data;

[0021] The training sample is generated according to the plurality of sensor data and the corresponding first road condition information.

[0022] Optionally, when a comparison result between the first road condition information and the second road condition information is a target comparison result, the external device sends the first acquisition time to the target vehicle, the first road condition determination model is deployed on a server, and the generating of the training sample according to the comparison information, the sensor data and the first road condition information includes:

[0023] If the comparison result is a target result, the first road condition information and the sensor data are sent to a server, so that the server uses the sensor data as sample content and generates sample labels based on the first road condition information to obtain training samples;

[0024] The method further comprises:

[0025] Receiving a trained first road condition determination model sent by the server, where the trained first road condition determination model is obtained by the server training the first road condition determination model based on the training sample;

[0026] The second road condition determination model is updated according to the trained first road condition determination model.

[0027] Optionally, the first road condition information includes first obstacle information of an obstacle detected by the target vehicle based on the on-board sensor, and the second road condition information includes second obstacle information of an obstacle detected by an external device.

[0028] The comparing the first road condition information and the second road condition information to obtain the comparison result includes:

[0029] Performing a consistency comparison based on the first obstacle information and the second obstacle information to determine a matching degree between the obstacle detected by the vehicle and the obstacle detected by the external device;

[0030] A comparison result between the first road condition information and the second road condition information is determined according to the matching degree.

[0031] According to a third aspect of the present application, a vehicle-mounted device is provided, the vehicle-mounted device is integrated in a target vehicle, the vehicle-mounted device includes a vehicle-mounted sensor, a communication module and a domain controller, wherein:

[0032] The vehicle-mounted sensor is used to collect sensor data of the target vehicle;

[0033] The communication module is used to obtain the first road condition information of the target vehicle detected by an external device;

[0034] The domain controller is used to generate training samples of a first road condition determination model based on the sensor data and the first road condition information, the first road condition information is used to generate sample labels of the training samples, and the first road condition determination model is used to determine the road condition information of the vehicle based on the sensor data.

[0035] According to a third aspect of the present application, a vehicle data processing device is provided, comprising:

[0036] A data acquisition unit, used to acquire sensor data collected by an onboard sensor of a target vehicle;

[0037] An information acquisition unit, used to acquire first road condition information of the target vehicle detected by an external device;

[0038] A generating unit is used to generate training samples of a first road condition determination model according to the sensor data and the first road condition information, wherein the first road condition information is used to generate sample labels of the training samples, and the first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data.

[0039] According to a fourth aspect of the present application, a vehicle is provided, comprising a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute any vehicle data processing method provided in the embodiments of the present application.

[0040] According to a fifth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium is used to store a computer program, and the computer program is loaded by a processor to execute any vehicle data processing method provided in the embodiments of the present application.

[0041] The embodiment of the present application obtains sensor data collected by an on-board sensor; obtains first road condition information of a target vehicle detected by an external device; and generates training samples of a first road condition determination model based on the sensor data and the first road condition information. The first road condition determination model is used to determine the road condition information of the target vehicle based on the sensor data collected by the on-board sensor.

[0042] In the embodiment of the present application, sample labels are generated based on the first road condition information, and training samples of the first road condition determination model are generated according to the sensor data and the sample labels, so that the sensor data can be labeled with the help of the road condition information detected by the external device, thereby improving the efficiency of generating training samples, and further improving the training speed of the first road condition determination model. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative work.

[0044] Figure 1 is a flow chart of a vehicle data processing method provided by an embodiment of the present application;

[0045] Figure 2 is a schematic diagram of a vehicle-mounted device provided in an embodiment of the present application;

[0046] Figure 3 is a schematic diagram of another vehicle-mounted device provided in an embodiment of the present application;

[0047] Figure 4 is a schematic diagram of a vehicle data processing device provided in an embodiment of the present application;

[0048] Figure 5 It is a schematic diagram of the structure of a vehicle provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0050] The embodiments of the present application provide a vehicle data processing method, device, vehicle and computer-readable storage medium. The vehicle data processing device can be integrated in the vehicle.

[0051] It should be noted that the description order of the following embodiments is not intended to limit the preferred order of the embodiments.

[0052] This embodiment will be described from the perspective of a data processing device of a vehicle, and the data processing device of the vehicle may specifically be integrated in the vehicle.

[0053] The present application provides a vehicle data processing method, such as Figure 1 As shown, the specific process of the vehicle data processing method can be as follows:

[0054] 101. Obtain sensor data collected by an onboard sensor of a target vehicle.

[0055] Among them, the on-board sensor is a sensor configured for the target vehicle. The on-board sensor can be used to collect data about obstacles around the target vehicle, that is, sensor data. The target vehicle or server processes the sensor data to determine the road condition information of the target vehicle.

[0056] The vehicle-mounted sensors may include cameras, radars, etc. The radars may be millimeter-wave radars, ultrasonic radars, etc. One or more sensors may be configured on the target vehicle, for example, the target vehicle may be configured with cameras, millimeter-wave radars, and ultrasonic radars.

[0057] The target vehicle is any vehicle that executes the data processing method provided in the embodiments of the present application.

[0058] 102. Obtain first road condition information of a target vehicle detected by an external device.

[0059] Among them, the external device is a device different from the target vehicle, for example, it can be a roadside device or other vehicles other than the target vehicle. The other vehicle can be at least one vehicle around the target vehicle. The other vehicle can determine the first road condition information of the target vehicle through data collected by its configured on-board sensors.

[0060] Roadside equipment is equipment that can be deployed along the road. It can receive and send wireless signals and communicate with passing vehicles in real time. Through roadside equipment, vehicles can be remotely monitored and managed to improve road safety and efficiency.

[0061] Roadside equipment may include roadside units (RSUs), as well as perception units and computing and decision-making units. Roadside perception units may be composed of perception devices and processing devices to achieve perception of local traffic environment and status, including signal light information, traffic participant information, traffic event information, positioning information, etc. The roadside computing and decision-making unit is responsible for processing and storing local or regional data, as well as calculating and publishing applications and services. RSU can communicate with vehicles and receive information from them, such as location, speed, direction of travel, etc. RSU can also receive traffic light status, traffic control information, etc., and pass this information to the vehicle. RSU helps the traffic management center monitor and control traffic flow through real-time communication with vehicles.

[0062] Since the roadside equipment can collect information about traffic participants, the roadside information can determine the target vehicle traveling on the road, so the roadside equipment can determine the first road condition information of the target vehicle based on the collected data through image processing, etc.

[0063] The target vehicle may include a communication module, which can be used to communicate with roadside equipment and servers. The communication module can support multiple communication methods, such as cellular vehicle-to-everything (C-V2X), WIFI, mobile network (e.g., 5G), etc., and support vehicle-to-vehicle interconnection, vehicle-to-Internet interconnection, vehicle-to-Global Navigation Satellite System (GNSS) and other system interconnection functions.

[0064] 103. Generate training samples of a first road condition determination model according to the sensor data and the first road condition information, the first road condition information is used to generate sample labels of the training samples, and the first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data.

[0065] Among them, the first road condition determination model can be a neural network model that determines the road condition of the target vehicle based on sensor data. Optionally, the first road condition determination model can be a part of the automatic driving model, and the automatic driving model can be used to output driving instructions for the vehicle based on sensor data, etc.

[0066] The first road condition information of the external device can guide the road condition determination model to learn the characteristics of the sensor data and accurately determine the road condition of the target vehicle. The training samples are used to train the first road condition determination model.

[0067] The first road condition determination model can be deployed on the target vehicle or on the server. If the first road condition determination model is deployed on the target vehicle, the target vehicle trains the first road condition determination model based on the training samples. If the first road condition determination model is deployed on the server, the server can train the first road condition determination model.

[0068] The vehicle may generate a training sample based on the sensor data and the first road condition information, or may send the sensor data and the first road condition information to a server, and the server may generate a training sample based on the sensor data and the first road condition information.

[0069] In one embodiment, a second road condition determination model may be deployed on the target vehicle, and the second road condition determination model may determine the road condition information of the target vehicle according to the sensor data. The model structure of the second road condition determination model may be the same as or different from that of the first road condition determination model, that is, the second road condition determination model is deployed on the target vehicle, and the first road condition determination model is used to update the second road condition determination model. Before the step of "generating a training sample of the first road condition determination model according to the sensor data and the first road condition information", the vehicle data processing method provided in the embodiment of the present application may also include:

[0070] Performing data analysis and processing on the sensor data by using a second road condition determination model to obtain second road condition information;

[0071] Obtaining comparison information between the first traffic condition information and the second traffic condition information;

[0072] The step of “generating a training sample of a first road condition determination model according to the sensor data and the first road condition information” may include:

[0073] A training sample is generated according to the comparison information, the sensor data and the first road condition information.

[0074] For example, the sensor data can be input into a second road condition determination model, and the sensor data can be analyzed and processed by the second road condition determination model to obtain second road condition information. The data analysis and processing can be the processing of processing the sensor data to determine the second road condition information. The data analysis and processing can include image processing, signal processing, target detection and other processing.

[0075] If the first road condition information and the second road condition information match, it can be considered that the result output by the second road condition determination model based on the sensor data is accurate; otherwise, it is considered that the result output by the second road condition determination model based on the sensor data is wrong, and it is necessary to learn the characteristic distribution characteristics of the sensor data. The first road condition determination model can be updated based on the sensor data and the first road condition information.

[0076] Among them, the comparison information may be a comparison result, which may indicate whether the first road condition information and the second road condition information match. The comparison information may also be that when the first road condition information and the second road condition information do not match, the external device determines the first collection time of the data based on which the first road condition information is based. The first collection time may also be the collection time of the sensor data by the vehicle-mounted sensor.

[0077] In one embodiment, the comparison information is a comparison result. The target vehicle can compare the first road condition information and the second road condition information to obtain a comparison result. If the comparison result is a target result, and the target result indicates that the first road condition information and the second road condition information do not match, a training sample can be generated based on the first road condition information and the sensor data.

[0078] In another embodiment, the comparison information is the first collection time of the first road condition information that does not match the second road condition information. The vehicle can send the second road condition information to an external device and trigger the external device to compare the first road condition information and the second road condition information to obtain a comparison result. If the comparison result is the target result, the first collection time corresponding to the first road condition information is sent to the target vehicle.

[0079] After receiving the first acquisition time sent by the external device, the target vehicle can obtain the first road condition information and sensor data corresponding to the first acquisition time according to the first acquisition time, and generate a training sample. Therefore, the comparison information may include one of the comparison result and the first acquisition time corresponding to the first road condition information. The step of "obtaining comparison information of the first road condition information and the second road condition information" includes:

[0080] Compare the first traffic condition information and the second traffic condition information to obtain a comparison result;

[0081] Or, sending the second traffic condition information to an external device, and triggering the external device to compare the first traffic condition information and the second traffic condition information, and returning the first collection time of the first traffic condition information according to a comparison result obtained by the comparison process;

[0082] Receive the first collection time sent by the external device.

[0083] The comparison process of the first road condition information and the second road condition information can be performed by the target vehicle or by an external device.

[0084] If executed by the target vehicle, the target vehicle may obtain the first road condition information from an external device for comparison with the second road condition information.

[0085] If executed by an external device, the target vehicle can send the second road condition information and the collection time of the sensor data to the external device, so that the external device can obtain the first road condition information according to the collection time. The second road condition information is used to compare with the first road condition information to obtain a comparison result.

[0086] If the comparison result indicates that the first road condition information and the second road condition information do not match, the first road condition information and the first collection time corresponding to the first road condition information can be sent to the target vehicle. The first collection time corresponding to the first road condition information can be the collection time of the data based on which the external device determines the first road condition information, or it can be the collection time when the vehicle-mounted sensor collects the sensor data.

[0087] The target vehicle can query the corresponding sensor data based on the first collection time, and generate a training sample based on the sensor data and the first road condition information.

[0088] Optionally, the first road condition information and the second road condition information do not match, indicating that the second road condition determination model on the target vehicle has a weak ability to analyze the current environment of the target vehicle, and the sensor data collected by the on-board sensor in the time period including the time point corresponding to the second road condition information and the corresponding first road condition information can be used to generate a training sample. That is, in one embodiment, when the comparison result of the first road condition information and the second road condition information is the target comparison result, the external device sends the first collection time to the target vehicle, and the step of "generating a training sample according to the comparison information, the sensor data and the first road condition information" includes:

[0089] If the comparison result is the target result, then determining the target time period corresponding to the second collection time;

[0090] Acquire multiple sensor data collected by the vehicle-mounted device within a target time period, and acquire first road condition information corresponding to the multiple sensor data;

[0091] A training sample is generated according to the plurality of sensor data and the corresponding first road condition information.

[0092] If the target vehicle compares the first road condition information and the second road condition information, the comparison result in the step "if the comparison result is the target result, determine the second collection time when the on-board sensor collects the sensor data" is the result obtained by the target vehicle performing the comparison processing; if the first road condition information and the second road condition information are compared by an external device, the comparison result in the step "if the comparison result is the target result, determine the second collection time when the on-board sensor collects the sensor data" is the result obtained by the external device performing the comparison processing; determining the second collection time when the on-board sensor collects the sensor data is equivalent to determining the second collection time based on the first collection time sent by the external device.

[0093] For example, if the first road condition information and the second road condition information do not match, the target time period corresponding to the second collection time when the vehicle-mounted sensor collects the sensor data may be determined, such as one minute before and one minute after the second collection time.

[0094] Acquire multiple sensor data collected by the vehicle-mounted sensors within the target time period, and acquire first road condition information corresponding to the multiple sensors. The first road condition information corresponding to the sensor data may refer to the collection time of the data based on which the external device determines the road condition information, and the road condition information close to the second collection time.

[0095] A training sample is generated according to a plurality of sensor data within a target time period and first road condition information corresponding to each sensor data within the target time period.

[0096] According to the comparison results, some sensor data are selected as training samples for the first road condition determination model, which can reduce the amount of data to be processed. According to the comparison results, problems with the accuracy of the road condition determination model can be discovered in time, and the road condition determination model can be updated according to the training samples.

[0097] The first road condition determination model is trained based on the training samples, which can be executed by the target vehicle or by the server. If the first road condition determination model is trained by the target vehicle, the second road condition determination model deployed on the target vehicle is the first road condition determination model.

[0098] If the first road condition determination model is trained by an external device, the second road condition determination model deployed on the target vehicle is a different model from the first road condition determination model, the first road condition determination model and the second road condition determination model have the same model structure and model task, and the first road condition determination model and the second road condition determination model have different model parameters, that is, in one embodiment, the first road condition determination model is deployed on a server, and the step of "generating a training sample according to the comparison information, the sensor data and the first road condition information" includes:

[0099] If the comparison result is the target result, the first road condition information and the sensor data are sent to the server, so that the server uses the sensor data as sample content and generates a sample label based on the first road condition information to obtain a training sample;

[0100] The method also includes:

[0101] Receiving a trained first road condition determination model sent by a server, where the trained first road condition determination model is obtained by the server training the first road condition determination model based on training samples;

[0102] The second road condition determination model is updated according to the trained first road condition determination model.

[0103] For example, if the comparison result is a target result, the first road condition information and the sensor data are sent to the server.

[0104] The server uses the sensor data as samples and the first road condition information as sample labels to generate training samples. The server trains the first road condition determination model based on the training samples to obtain the trained first road condition determination model, and sends the trained first road condition determination model to the target vehicle.

[0105] The target vehicle may update the second road condition determination model according to the model parameters of the trained first road condition determination model, or may replace the second road condition determination model with the trained first road condition determination model.

[0106] The road condition information can indicate the current road condition level of the vehicle, such as good road conditions and bad road conditions. It can be further subdivided, such as good road conditions, average road conditions, and bad road conditions, etc. The first road condition information and the second road condition information are compared to see whether they are the same.

[0107] Optionally, the road condition information may also include obstacle information around the target vehicle, such as cones, triangular warning signs, basement locks, cartons, tires, water barriers, pillars, stone piers, etc. The first road condition information and the second road condition information are compared and processed, and the obstacle detected by the external device is compared with whether the obstacle detected by the target vehicle matches. That is, in one embodiment, the first road condition information includes the first obstacle information of the obstacle detected by the vehicle based on the sensor, and the second road condition information includes the second obstacle information of the obstacle detected by the external device. The step of "comparing the first road condition information and the second road condition information to obtain a comparison result" may include:

[0108] Performing a consistency comparison based on the first obstacle information and the second obstacle information to determine a matching degree between the obstacle detected by the target vehicle and the obstacle detected by the external device;

[0109] A comparison result between the first traffic condition information and the second traffic condition information is determined according to the matching degree.

[0110] The matching degree may be used to measure the similarity between the first traffic condition information and the second traffic condition information. For example, a completely identical matching degree may be 1, a completely different matching degree may be 0, and a partially identical matching degree may be between 0 and 1.

[0111] For example, a consistency comparison is performed based on the first obstacle information and the second obstacle information to determine the matching degree between the obstacle detected by the target vehicle and the obstacle detected by the external device.

[0112] Exemplarily, if the obstacle detected by the target vehicle is exactly the same as the obstacle detected by the external device, the matching degree is 1. If the obstacle detected by the target vehicle is different from the obstacle detected by the external device, the ratio of the number of obstacles detected by the target vehicle that are the same as the obstacles detected by the external device to the sum of the obstacles detected by the target vehicle and the obstacles detected by the external device can be used as the matching degree.

[0113] The comparison result of the first road condition information and the second road condition information is determined based on the size relationship between the matching degree and the preset matching degree threshold. For example, if the matching degree is greater than the preset matching degree threshold, the comparison result may be that the first road condition information and the second road condition information match; if the matching degree is not greater than the preset matching degree threshold, the comparison result may be that the first road condition information and the second road condition information do not match.

[0114] As can be seen from the above, the embodiment of the present application obtains sensor data collected by the on-board sensor of the target vehicle; obtains first road condition information of the target vehicle detected by an external device; and generates training samples of a first road condition determination model based on the sensor data and the first road condition information. The first road condition determination model is used to determine the road condition information of the target vehicle based on the sensor data collected by the on-board sensor.

[0115] In the embodiment of the present application, sample labels are generated based on the first road condition information, and training samples of the first road condition determination model are generated according to the sensor data and the sample labels, so that the sensor data can be labeled with the help of the road condition information detected by the external device, thereby improving the efficiency of generating training samples, and further improving the training speed of the first road condition determination model.

[0116] In order to facilitate better implementation of the vehicle data processing method provided in the embodiment of the present application, an on-board device is also provided in one embodiment, and the vehicle data processing method provided in the embodiment of the present application can be implemented by the on-board device. The meanings of the terms are the same as those in the above-mentioned vehicle data processing method, and the specific implementation details can refer to the description in the method embodiment.

[0117] The vehicle-mounted device can be specifically integrated in a vehicle, such as Figure 2 As shown, the vehicle-mounted device may include a vehicle-mounted sensor, a communication module and a domain controller, as follows:

[0118] First, refer to Figure 2 A vehicle-mounted device according to an embodiment of the present application is described. The vehicle-mounted device is integrated in a target vehicle. The vehicle-mounted device 100 includes a vehicle-mounted sensor 110, a domain controller 120, and a communication module 130. Figure 2 A schematic block diagram of a vehicle-mounted device 100 according to an embodiment of the present application is shown.

[0119] On-board sensors are used to collect sensor data from target vehicles.

[0120] The communication module is used to obtain the first road condition information of the target vehicle detected by an external device.

[0121] The domain controller is used to generate training samples of a first road condition determination model based on the sensor data and the first road condition information. The first road condition determination model is used to determine the road condition information of the vehicle based on the sensor data collected by the vehicle-mounted sensor.

[0122] In one embodiment, the vehicle-mounted sensor includes at least one of a camera, a millimeter-wave radar, and an ultrasonic radar.

[0123] In one embodiment, a second road condition determination model is deployed on the target vehicle, and specifically may be deployed on a domain controller of the target vehicle. The domain controller is further used to perform data analysis and processing on the sensor data through the second road condition determination model to obtain second road condition information, and to compare the first road condition information with the second road condition information to obtain the comparison result.

[0124] If the above comparison result is the target result, the above sensor data and the above first road condition information are sent to the server through the above communication module, so that the above server uses the above sensor data as sample content, generates sample labels based on the above first road condition information to obtain training samples, and trains the above first road condition determination model based on the above training samples to obtain the trained first road condition determination model, and the trained first road condition determination model is used to update the above second road condition determination model.

[0125] The server uses the above sensor data as samples, generates sample labels based on the above first road condition information to obtain training samples, and trains the above first road condition determination model based on the above training samples to obtain the trained first road condition determination model, and sends the trained first road condition determination model to the target vehicle.

[0126] The communication module can receive the trained first road condition determination model sent by the server, and replace the second road condition determination model on the target vehicle with the trained first road condition determination model.

[0127] Exemplarily, the vehicle-mounted sensor includes at least one of a camera, a millimeter-wave radar, and an ultrasonic radar. In one example, the camera in the vehicle-mounted sensor 110 includes a front-view camera (e.g., a monocular / binocular / tri-camera camera), a side-view auxiliary camera (e.g., a wide-angle camera), and a night-vision auxiliary camera; the radar includes a millimeter-wave radar and / or an ultrasonic radar.

[0128] In one example, the vehicle-mounted device 100 measures the speed and distance of targets in the vehicle's surrounding environment by using the frequency difference between the transmitted signal and the received signal of the millimeter wave radar. The vehicle-mounted device 100 releases ultrasonic waves through ultrasonic radar. When encountering an obstacle, the ultrasonic waves are reflected, thereby obtaining the specific distance of the obstacle. The vehicle-mounted sensor 110 data includes images obtained by the camera in the vehicle-mounted sensor 110 and radar data obtained by the radar.

[0129] The domain controller 120 is used to obtain the second road condition information of the target vehicle based on the on-board sensor data, and send the sensor data and the first road condition information to the server based on the comparison result of the first road condition information and the second road condition information.

[0130] The communication module 130 is used to receive the first road condition information sent by the external device. In one example, the communication module 130 can be a communication module using the 3GPP communication protocol, the IEEE protocol, or other communication protocols, supporting functions such as vehicle-to-vehicle interconnection, vehicle-to-Internet interconnection, vehicle-to-other systems (such as the Global Navigation Satellite System (GNSS)) interconnection, and supporting multiple communication protocols (such as WIFI, Ethernet, etc.).

[0131] The domain controller 120 is further configured to receive first road condition information from the first device via the communication module 130 .

[0132] In one embodiment, if Figure 3 As shown, the vehicle-mounted device 100 also includes a gateway and a vehicle controller, and the communication module is integrated in Figure 3 In the telematics control unit (TCU), the domain controller 120 is also used to transmit information (such as the second road condition information, or a control instruction determined based on the second road condition information) to the vehicle controller via the gateway, so that the vehicle controller controls the vehicle where the vehicle-mounted device 100 is located.

[0133] The gateway of the domain controller 120 is an interface connecting different functional domains of the vehicle, realizing data conversion, routing, filtering and security processing, and supporting multiple communication protocols (such as CAN, CAN-FD, Ethernet, LIN, etc.). The domain controller 120 transmits information to the vehicle controller via the gateway.

[0134] In one embodiment, if Figure 3 As shown, the vehicle controller includes at least one of the following: a vehicle controller, a body controller, and a chassis controller.

[0135] The vehicle controller receives the second road condition information, vehicle information (such as vehicle positioning information) and driver operation instructions (such as via Figure 3 The human-machine interaction interface HMI shown in the figure sends control instructions to the control unit after processing according to the control strategy, and monitors the vehicle operation status in real time; the body controller controls the body electrical system based on the second road condition information sent by the domain controller 120, including controlling electric windows, electric rearview mirrors, air conditioning, headlights, turn signals, anti-theft locking system, central locking, defrosting device, etc.; the chassis controller controls the automobile chassis system based on the second road condition information sent by the domain controller 120. The chassis controller controls chassis subsystems including the engine, gearbox, electronic body stability system (ESP), electronic hand brake (EPB), electric power steering system (EPS), etc., to achieve lateral, longitudinal and vertical motion control of the vehicle, and adjusts the working status of the chassis system according to the second road condition information sent by the domain controller 120, the vehicle's own speed and road conditions, and the driver's intention.

[0136] As can be seen from the above, the vehicle-mounted device in the embodiment of the present application includes a vehicle-mounted sensor, a communication module and a domain controller, wherein: the vehicle-mounted sensor is used to collect sensor data; the communication module is used to obtain the first road condition information of the target vehicle detected by the external device; the domain controller is used to generate training samples of the first road condition determination model based on the sensor data and the first road condition information, the first road condition information is used to generate sample labels for the training samples, and the first road condition determination model is used to determine the road condition information of the vehicle based on the sensor data.

[0137] In the embodiment of the present application, sample labels are generated based on the first road condition information, and training samples of the first road condition determination model are generated according to the sensor data and the sample labels, so that the sensor data can be labeled with the help of the road condition information detected by the external device, thereby improving the efficiency of generating training samples, and further improving the training speed of the first road condition determination model.

[0138] In order to facilitate better implementation of the vehicle data processing method provided in the embodiment of the present application, a vehicle data processing device is also provided in one embodiment. The meanings of the terms are the same as those in the above vehicle data processing method, and the specific implementation details can refer to the description in the method embodiment.

[0139] The data processing device of the vehicle may be integrated in the vehicle, such as Figure 4 As shown, the data processing device of the vehicle may include: a data acquisition unit 301, an information acquisition unit 302 and a generation unit 303, which are specifically as follows:

[0140] (1) a data acquisition unit 301, used to acquire sensor data collected by an onboard sensor of a target vehicle;

[0141] (2) an information acquisition unit 302, used to acquire first road condition information of the target vehicle detected by an external device;

[0142] (3) A generating unit 303, configured to generate training samples of a first road condition determination model based on the sensor data and the first road condition information, wherein the first road condition determination model is configured to determine the road condition information of the target vehicle based on the sensor data collected by the vehicle-mounted sensor.

[0143] In one embodiment, a second road condition determination model is deployed on the target vehicle, the first road condition determination model is used to update the second road condition determination model, and the data processing device of the vehicle further includes:

[0144] An analysis unit, configured to perform data analysis and processing on the sensor data by using a second road condition determination model to obtain second road condition information;

[0145] A comparison unit, used to obtain comparison information between the first road condition information and the second road condition information;

[0146] The generating unit 303 may also be used for:

[0147] The training sample is generated according to the comparison information, the sensor data and the first road condition information.

[0148] In one embodiment, the first road condition information includes first obstacle information of an obstacle detected by a sensor of the vehicle, and the second road condition information includes second obstacle information of an obstacle detected by an external device.

[0149] The above-mentioned comparison unit comprises:

[0150] a consistency comparison subunit, configured to perform consistency comparison based on the first obstacle information and the second obstacle information to determine a matching degree between the obstacle detected by the vehicle and the obstacle detected by the external device;

[0151] The result determination subunit is used to determine the comparison result of the first road condition information and the second road condition information according to the matching degree.

[0152] In one embodiment, the comparison information includes one of a comparison result and a first collection time of the first road condition information, and the comparison unit may also be used for:

[0153] Comparing the first road condition information with the second road condition information to obtain the comparison result;

[0154] Or comparison units may include:

[0155] a sending subunit, configured to send the second road condition information to the external device, trigger the external device to compare the first road condition information with the second road condition information, and return the first collection time of the first road condition information according to a comparison result obtained by the comparison;

[0156] The receiving subunit is used to receive the first acquisition time sent by the external device.

[0157] In one embodiment, when the comparison result between the first road condition information and the second road condition information is a target comparison result, the external device sends the first collection time to the target vehicle, and the generating unit 303 includes:

[0158] a time determination subunit, for determining a target time period corresponding to a second acquisition time when the vehicle-mounted sensor acquires the sensor data if the comparison result is a target result;

[0159] A road condition information acquisition subunit, used to acquire a plurality of sensor data collected by the vehicle-mounted device within a target time period, and to acquire first road condition information corresponding to the plurality of sensor data;

[0160] The sample generation subunit generates the training sample according to the plurality of sensor data and the corresponding first road condition information.

[0161] In one embodiment, when the comparison result between the first road condition information and the second road condition information is a target comparison result, the external device sends the first collection time to the target vehicle, the first road condition determination model is deployed on a server, and the generation unit 303 is further used to:

[0162] If the comparison result is the target result, the first road condition information and the sensor data are sent to a server, so that the server uses the sensor data as sample content and generates a sample label based on the first road condition information to obtain a training sample;

[0163] The data processing device of the vehicle may further include:

[0164] A model receiving unit, configured to receive the trained first road condition determination model sent by the server, wherein the trained first road condition determination model is obtained by the server training the first road condition determination model based on the training sample;

[0165] The model updating unit is used to update the second road condition determination model according to the trained first road condition determination model.

[0166] As can be seen from the above, the embodiment of the present application obtains sensor data collected by the on-board sensor of the target vehicle through the data acquisition unit 301; the information acquisition unit 302 obtains the first road condition information of the target vehicle detected by the external device; the generation unit 303 generates training samples of the first road condition determination model based on the sensor data and the first road condition information, and the first road condition determination model is used to determine the road condition information of the target vehicle based on the sensor data collected by the on-board sensor.

[0167] In the embodiment of the present application, sample labels are generated based on the first road condition information, and training samples of the first road condition determination model are generated according to the sensor data and the sample labels, so that the sensor data can be labeled with the help of the road condition information detected by the external device, thereby improving the efficiency of generating training samples, and further improving the training speed of the first road condition determination model.

[0168] The present application also provides a vehicle, such as Figure 5 As shown, it shows a schematic diagram of the structure of a vehicle involved in an embodiment of the present application, specifically:

[0169] The vehicle may include components such as a processor 1001 of one or more processing cores, a memory 1002 of one or more computer-readable storage media, a power supply 1003, and an input unit 1004. Those skilled in the art will appreciate that Figure 5 The vehicle structure shown in the figure does not constitute a limitation on the vehicle, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.

[0170] in:

[0171] The processor 1001 is the control center of the vehicle, and uses various interfaces and lines to connect various parts of the entire vehicle. By running or executing software programs and / or modules stored in the memory 1002, and calling data stored in the memory 1002, it performs various functions of the vehicle and processes data, thereby monitoring the vehicle as a whole. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly processes the operating system, user interface, and computer programs, etc., and the modem processor mainly processes wireless communications. It is understandable that the above-mentioned modem processor may not be integrated into the processor 1001.

[0172] The memory 1002 can be used to store software programs and modules. The processor 1001 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, a computer program required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the vehicle, etc. In addition, the memory 1002 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

[0173] The vehicle further includes a power supply 1003 for supplying power to various components. Preferably, the power supply 1003 can be logically connected to the processor 1001 through a power management system, so as to manage charging, discharging, and power consumption through the power management system. The power supply 1003 can also include any components such as one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators.

[0174] The vehicle may further include an input unit 1004, which may be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0175] Although not shown, the vehicle may also include a display unit, etc., which will not be described in detail here. Specifically in this embodiment, the processor 1001 in the vehicle will load the executable files corresponding to the processes of one or more computer programs into the memory 1002 according to the following instructions, and the processor 1001 will run the computer programs stored in the memory 1002 to achieve various functions, as follows:

[0176] Obtaining sensor data collected by onboard sensors of the target vehicle;

[0177] Acquire first road condition information of a target vehicle detected by an external device;

[0178] A training sample of a first road condition determination model is generated according to the sensor data and the first road condition information. The first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data collected by the vehicle-mounted sensor.

[0179] The specific implementation of the above operations can be found in the previous embodiments and will not be described in detail here.

[0180] As can be seen from the above, the embodiment of the present application obtains sensor data collected by the on-board sensor of the target vehicle; obtains first road condition information of the target vehicle detected by an external device; and generates training samples of a first road condition determination model based on the sensor data and the first road condition information. The first road condition determination model is used to determine the road condition information of the target vehicle based on the sensor data collected by the on-board sensor.

[0181] In the embodiment of the present application, sample labels are generated based on the first road condition information, and training samples of the first road condition determination model are generated according to the sensor data and the sample labels, so that the sensor data can be labeled with the help of the road condition information detected by the external device, thereby improving the efficiency of generating training samples, and further improving the training speed of the first road condition determination model.

[0182] According to one aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a vehicle reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the vehicle executes the methods provided in various optional implementations in the above-mentioned embodiments.

[0183] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be completed by a computer program, or by controlling related hardware through a computer program. The computer program may be stored in a computer-readable storage medium and loaded and executed by a processor.

[0184] To this end, an embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. The computer program can be loaded by a processor to execute any vehicle data processing method provided in the embodiment of the present application.

[0185] The specific implementation of the above operations can be found in the previous embodiments, which will not be described in detail here.

[0186] The computer-readable storage medium may include: a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0187] Since the computer program stored in the computer-readable storage medium can execute any vehicle data processing method provided in the embodiments of the present application, the beneficial effects that can be achieved by any vehicle data processing method provided in the embodiments of the present application can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0188] The above is a detailed introduction to a vehicle data processing method, device, vehicle and computer-readable storage medium provided in an embodiment of the present application. Specific examples are used herein to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A vehicle data processing method, characterized in that: include: Obtaining sensor data collected by onboard sensors of the target vehicle; Acquire first road condition information of the target vehicle detected by an external device; A training sample of a first road condition determination model is generated according to the sensor data and the first road condition information. The first road condition information is used to generate a sample label of the training sample. The first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data.

2. The method according to claim 1, characterized in that The target vehicle is equipped with a second road condition determination model, the first road condition determination model is used to update the second road condition determination model, and before generating a training sample for the first road condition determination model according to the sensor data and the first road condition information, the method further includes: Performing data analysis and processing on the sensor data by using the second road condition determination model to obtain second road condition information; Obtaining comparison information between the first road condition information and the second road condition information; The step of generating a training sample of a first road condition determination model according to the sensor data and the first road condition information includes: The training sample is generated according to the comparison information, the sensor data and the first road condition information.

3. The method according to claim 2, characterized in that The comparison information includes one of a comparison result and a first collection time of the first road condition information, and the obtaining the comparison information of the first road condition information and the second road condition information includes: Comparing the first road condition information and the second road condition information to obtain the comparison result; or, Sending the second traffic condition information to the external device, triggering the external device to compare the first traffic condition information and the second traffic condition information, and returning the first collection time of the first traffic condition information according to the comparison result obtained by the comparison process; Receive the first collection time sent by the external device.

4. The method according to claim 3, characterized in that The external device sends the first acquisition time to the target vehicle when the comparison result between the first road condition information and the second road condition information is a target comparison result, and generates the training sample according to the comparison information, the sensor data and the first road condition information, including: If the comparison result is a target result, determining a target time period corresponding to the second collection time of the sensor data; Acquire multiple sensor data collected by the vehicle-mounted device within a target time period, and acquire first road condition information corresponding to the multiple sensor data; The training sample is generated according to the plurality of sensor data and the corresponding first road condition information.

5. The method according to claim 3, characterized in that: The external device sends the first acquisition time to the target vehicle when the comparison result between the first road condition information and the second road condition information is a target comparison result, the first road condition determination model is deployed on a server, and the training sample is generated according to the comparison information, the sensor data and the first road condition information, including: If the comparison result is a target result, the first road condition information and the sensor data are sent to a server, so that the server uses the sensor data as sample content and generates sample labels based on the first road condition information to obtain training samples; The method further comprises: Receiving a trained first road condition determination model sent by the server, where the trained first road condition determination model is obtained by the server training the first road condition determination model based on the training sample; The second road condition determination model is updated according to the trained first road condition determination model.

6. The method according to claim 3, characterized in that: The first road condition information includes first obstacle information of an obstacle detected by the target vehicle based on the on-board sensor, and the second road condition information includes second obstacle information of an obstacle detected by an external device. The comparing the first road condition information and the second road condition information to obtain the comparison result includes: Performing a consistency comparison based on the first obstacle information and the second obstacle information to determine a matching degree between the obstacle detected by the target vehicle and the obstacle detected by the external device; A comparison result between the first road condition information and the second road condition information is determined according to the matching degree.

7. A vehicle-mounted device, characterized in that: The vehicle-mounted device is integrated into the target vehicle, and the vehicle-mounted device includes a vehicle-mounted sensor, a communication module and a domain controller, wherein: The vehicle-mounted sensor is used to collect sensor data of the target vehicle; The communication module is used to obtain the first road condition information of the target vehicle detected by an external device; The domain controller is used to generate training samples of a first road condition determination model based on the sensor data and the first road condition information, the first road condition information is used to generate sample labels of the training samples, and the first road condition determination model is used to determine the road condition information of the vehicle based on the sensor data.

8. The vehicle-mounted device according to claim 7, characterized in that: The target vehicle is equipped with a second road condition determination model, and the domain controller is further used for: Performing data analysis and processing on the sensor data by using the second road condition determination model to obtain second road condition information; Comparing the first road condition information and the second road condition information to obtain a comparison result; If the comparison result is the target result, the sensor data and the first road condition information are sent to the server through the communication module, so that the server uses the sensor data as sample content, generates sample labels based on the first road condition information to obtain training samples, and trains the first road condition determination model based on the training samples to obtain the trained first road condition determination model, and the trained first road condition determination model is used to update the second road condition determination model.

9. The vehicle-mounted device according to claim 7, characterized in that: The vehicle-mounted sensor includes at least one of a camera, a millimeter-wave radar and an ultrasonic radar.

10. A vehicle data processing device, characterized in that: include: A data acquisition unit, used to acquire sensor data collected by an onboard sensor of a target vehicle; An information acquisition unit, used to acquire first road condition information of the target vehicle detected by an external device; A generating unit is used to generate training samples of a first road condition determination model according to the sensor data and the first road condition information, wherein the first road condition information is used to generate sample labels of the training samples, and the first road condition determination model is used to determine the road condition information of the target vehicle according to the sensor data.

11. A vehicle, characterized in that: It comprises a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute the vehicle data processing method according to any one of claims 1 to 6.

12. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store a computer program, and the computer program is loaded by a processor to execute the vehicle data processing method according to any one of claims 1 to 6.

Images