Vehicle data processing method and device, computer device and storage medium

Abstract

The vehicle data processing method based on the traffic field comprises the following steps: obtaining first sample driving data of a first sample vehicle, second sample driving data of a second sample vehicle which has a driving sequence with the first sample vehicle on a sample road; obtaining sample road surface data of each sample sensor related to the first sample vehicle; determining a first correlation relationship according to the first sample driving data and the second sample driving data; determining a second correlation relationship according to the sample road surface data; based on a first initial weight parameter of the second sample vehicle and a second initial weight parameter of each sample sensor, combining the first sample driving data and the second sample driving data, the first correlation relationship and the second correlation relationship, and the sample road surface data, performing cross-iteration weight parameter adjustment to obtain a first target weight parameter and each second target weight parameter; the first target weight parameter and the second target weight parameter are used to determine the distance between the vehicles which have the driving sequence. The method can improve the accuracy of the distance prediction between the vehicles.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a vehicle data processing method, apparatus, computer equipment, storage medium, and computer program product. Background Technology

[0002] Vehicle-road cooperation is a safe, efficient and environmentally friendly road traffic system that uses advanced wireless communication and next-generation Internet technologies to implement dynamic real-time information interaction between vehicles and roads in all aspects. Based on the collection and fusion of dynamic traffic information in all time and space, it carries out active safety control of vehicles and cooperative road management, fully realizes effective cooperation between people, vehicles and roads, ensures traffic safety, improves traffic efficiency and thus forms a safe, efficient and environmentally friendly road traffic system.

[0003] In traditional vehicle-road cooperative solutions, the distance between vehicles is typically predicted using feature data collected by the vehicles themselves. However, feature data collected by different vehicles may contain overlapping parts. Repeatedly aggregating the same feature data during distance prediction leads to inaccurate distance predictions. Summary of the Invention

[0004] Therefore, it is necessary to provide a vehicle data processing method, apparatus, computer equipment, computer-readable storage medium, and computer program product that can improve accuracy in addressing the aforementioned technical problems.

[0005] This application provides a vehicle data processing method, the method comprising:

[0006] Obtain the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road;

[0007] Acquire sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road;

[0008] Based on the first sample driving data and the second sample driving data, a first correlation relationship is determined between the first sample vehicle and the second sample vehicle; based on the multiple sample road surface data, a second correlation relationship is determined between the multiple sample sensors.

[0009] Obtain the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor;

[0010] Based on the first initial weight parameter and each of the second initial weight parameters, and in combination with the first sample driving data, the second sample driving data, each of the sample road surface data, the first correlation relationship, and the second correlation relationship, the weight parameters are adjusted through cross-iteration to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each of the sample sensors; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

[0011] In one embodiment, the iterative processing of the weight parameters based on each of the second initial weight parameters, the forward initial prediction distance, and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, each of the sample road surface data, and the second correlation relationship, to obtain the second target weight parameters corresponding to each sample sensor, includes:

[0012] Based on the forward expected distance and the backward expected distance, the global expected distance is determined; based on the forward initial prediction distance and the backward initial prediction distance, the global initial prediction distance is determined; based on each second initial weight parameter and the global initial prediction distance, combined with the global expected distance, each of the sample road surface data and the second correlation relationship, the weight parameters are iteratively processed to obtain the second target weight parameter corresponding to each sample sensor.

[0013] This application also provides a vehicle data processing apparatus, the apparatus comprising:

[0014] The sample driving data acquisition module is used to acquire the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road;

[0015] The sample road surface data acquisition module is used to acquire sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road.

[0016] The correlation determination module is used to determine a first correlation between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data; and to determine a second correlation between the multiple sample sensors based on the multiple sample road surface data.

[0017] The weight parameter acquisition module is used to acquire the first initial weight parameter of the second sample vehicle and the second initial weight parameter of each sample sensor;

[0018] The weight parameter adjustment module is used to perform cross-iterative weight parameter adjustment based on the first initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the second sample driving data, each of the sample road surface data, the first correlation relationship and the second correlation relationship, to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each of the sample sensors; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

[0019] In one embodiment, the association determination module is further configured to perform feature extraction on multiple first dimensions based on the first sample driving data to obtain first sample features of the first sample driving data on each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension; perform feature extraction on the multiple first dimensions based on the second sample driving data to obtain second sample features of the second sample driving data on each first dimension; and determine a first association between the first sample vehicle and the second sample vehicle based on each first sample feature and each second sample feature.

[0020] In one embodiment, the association determination module is further configured to determine sub-associations between each of the first sample features and each of the second sample features; and to determine a first association between the first sample vehicle and the second sample vehicle based on each of the sub-associations.

[0021] In one embodiment, the correlation determination module is further configured to perform feature extraction of multiple second dimensions based on each of the sample road surface data to obtain sample road surface features corresponding to each of the sample road surface data in each second dimension; the multiple second dimensions include road surface driving dimension and road surface attribute dimension; and determine the second correlation between the multiple sample sensors according to the multiple sample road surface features corresponding to each of the sample road surface data.

[0022] In one embodiment, the weight parameter adjustment module is further configured to: determine an initial predicted distance between the first sample vehicle and the second sample vehicle based on the first initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the second sample driving data, and each of the sample road surface data; obtain the expected distance between the first sample vehicle and the second sample vehicle; perform iterative processing of the weight parameters based on the first initial weight parameter and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship, to obtain a first target weight parameter corresponding to the second sample vehicle; and perform iterative processing of the weight parameters based on each of the second initial weight parameters and the initial predicted distance, combined with the expected distance, each of the sample road surface data, and the second correlation relationship, to obtain a second target weight parameter corresponding to each of the sample sensors.

[0023] In one embodiment, the second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle; the forward sample vehicle is positioned before the first sample vehicle on the sample road, and the backward sample vehicle is positioned after the first sample vehicle on the sample road; the second sample driving data includes forward sample driving data of the forward sample vehicle and backward sample driving data of the backward sample vehicle; the initial prediction distance includes forward initial prediction distance and backward initial prediction distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the expected distance includes forward expected distance and backward expected distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the first association relationship includes forward association relationship and backward association relationship between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the first initial weight parameter includes forward initial weight parameter of the forward sample vehicle and backward initial weight parameter of the backward sample vehicle; the first target weight parameter includes forward target weight parameter of the forward sample vehicle and backward target weight parameter of the backward sample vehicle.

[0024] In one embodiment, the weight parameter adjustment module is further configured to determine the forward initial predicted distance between the first sample vehicle and the forward sample vehicle based on the forward initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the forward sample driving data, and each of the sample road surface data; and to determine the backward initial predicted distance between the first sample vehicle and the backward sample vehicle based on the backward initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the backward sample driving data, and each of the sample road surface data.

[0025] In one embodiment, the weight parameter adjustment module is further configured to perform iterative processing of the weight parameters based on the forward initial weight parameters and the forward initial predicted distance, combined with the forward expected distance, the first sample driving data, the forward sample driving data, and the forward correlation, to obtain the forward target weight parameters corresponding to the forward sample vehicle; and to perform iterative processing of the weight parameters based on the backward initial weight parameters and the backward initial predicted distance, combined with the backward expected distance, the first sample driving data, the backward sample driving data, and the backward correlation, to obtain the backward target weight parameters corresponding to the backward sample vehicle.

[0026] In one embodiment, the weight parameter adjustment module is further configured to perform iterative processing of the weight parameters based on each of the second initial weight parameters, the forward initial prediction distance and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, each of the sample road surface data and the second correlation relationship, to obtain the second target weight parameter corresponding to each of the sample sensors.

[0027] In one embodiment, the weight parameter adjustment module is further configured to: determine a global expected distance based on the forward expected distance and the backward expected distance; determine a global initial predicted distance based on the forward initial predicted distance and the backward initial predicted distance; and perform iterative processing of the weight parameters based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, each of the sample road surface data and the second correlation relationship, to obtain a second target weight parameter corresponding to each sample sensor.

[0028] This application also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0029] The process involves: acquiring first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle; determining a driving order between the first and second sample vehicles on a sample road; acquiring sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road; determining a first correlation between the first and second sample vehicles based on the first and second sample driving data; determining a second correlation between the multiple sample sensors based on the multiple sample road surface data; acquiring a first initial weight parameter for the second sample vehicle and a second initial weight parameter for each of the sample sensors; performing cross-iterative weight parameter adjustments based on the first initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the second sample driving data, each of the sample road surface data, the first correlation, and the second correlation, to obtain a first target weight parameter corresponding to the second sample vehicle and a second target weight parameter corresponding to each of the sample sensors; and using the first target weight parameter and the second target weight parameter to determine the distance between at least two vehicles with a driving order.

[0030] This application also provides a computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0031] The process involves: acquiring first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle; determining a driving order between the first and second sample vehicles on a sample road; acquiring sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road; determining a first correlation between the first and second sample vehicles based on the first and second sample driving data; determining a second correlation between the multiple sample sensors based on the multiple sample road surface data; acquiring a first initial weight parameter for the second sample vehicle and a second initial weight parameter for each of the sample sensors; performing cross-iterative weight parameter adjustments based on the first initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the second sample driving data, each of the sample road surface data, the first correlation, and the second correlation, to obtain a first target weight parameter corresponding to the second sample vehicle and a second target weight parameter corresponding to each of the sample sensors; and using the first target weight parameter and the second target weight parameter to determine the distance between at least two vehicles with a driving order.

[0032] This application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0033] The process involves: acquiring first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle; determining a driving order between the first and second sample vehicles on a sample road; acquiring sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road; determining a first correlation between the first and second sample vehicles based on the first and second sample driving data; determining a second correlation between the multiple sample sensors based on the multiple sample road surface data; acquiring a first initial weight parameter for the second sample vehicle and a second initial weight parameter for each of the sample sensors; performing cross-iterative weight parameter adjustments based on the first initial weight parameter and each of the second initial weight parameters, combined with the first sample driving data, the second sample driving data, each of the sample road surface data, the first correlation, and the second correlation, to obtain a first target weight parameter corresponding to the second sample vehicle and a second target weight parameter corresponding to each of the sample sensors; and using the first target weight parameter and the second target weight parameter to determine the distance between at least two vehicles with a driving order.

[0034] The aforementioned vehicle data processing method, apparatus, computer equipment, storage medium, and computer program product acquire first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle, wherein the first and second sample vehicles have a driving sequence on a sample road. Based on the first and second sample driving data, a first correlation relationship is determined between the first and second sample vehicles, thereby determining the correlation between the driving data collected by each of the first and second sample vehicles. Furthermore, sample road surface data collected by multiple sample sensors associated with the first sample vehicle on the sample road are acquired, and based on the multiple sample road surface data, a second correlation relationship is determined between the multiple sample sensors, thereby determining the correlation between the multiple sample sensors on the road surface data.

[0035] Based on the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor, and combining the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship, and the second correlation relationship, cross-iterative weight parameter adjustments are performed. This allows for weight calculation by combining the correlation between the driving data of each vehicle and the correlation between the road surface data collected by each sensor, resulting in the first target weight parameters of the second sample vehicle and the second target weight parameters of each sample sensor. This avoids the problem of inaccurate weight parameters caused by the repeated superposition and calculation of the same data, effectively improving the accuracy of weight parameter calculation. Using the first target weight parameters and the second target weight parameters to determine the distance between at least two vehicles with a driving sequence can effectively improve the accuracy of vehicle distance prediction.

[0036] This application also provides a vehicle data processing method, the method comprising:

[0037] Acquire the first driving data of the first vehicle and the second driving data of the second vehicle; the first vehicle and the second vehicle have a driving order on the target road;

[0038] Acquire target road surface data collected by multiple sensors associated with the first vehicle for the target road;

[0039] According to the driving sequence, obtain the first target weight parameters obtained by pre-training for the second vehicle and the second target weight parameters obtained by pre-training for each of the sensors;

[0040] Regression analysis is performed based on the first driving data, the second driving data, and the first target weight parameter to obtain the first regression data;

[0041] Regression analysis is performed based on the road surface data and each of the second target weight parameters to obtain the second regression data.

[0042] Based on the first regression data and the second regression data, the vehicle distance between the first vehicle and the second vehicle is determined.

[0043] In one embodiment, the second vehicle includes a vehicle traveling in front of the first vehicle and a vehicle traveling behind the first vehicle. The vehicle traveling in front is positioned ahead of the first vehicle on the target road, and the vehicle traveling behind the first vehicle is positioned behind the first vehicle on the target road. The second driving data includes forward driving data of the vehicle traveling in front and backward driving data of the vehicle traveling behind. The first target weight parameter includes forward target weight parameters of the vehicle traveling in front and backward target weight parameters of the vehicle traveling behind. The first regression data includes forward regression data of the vehicle traveling in front and backward regression data of the vehicle traveling behind. The vehicle distance includes the forward distance between the first vehicle and the vehicle traveling in front, and the backward distance between the first vehicle and the vehicle traveling behind.

[0044] In one embodiment, the regression analysis based on the first driving data, the second driving data, and the first target weight parameter to obtain the first regression data includes:

[0045] Regression analysis is performed based on the first driving data, the forward driving data, and the forward target weight parameter to obtain forward regression data; regression analysis is performed based on the first driving data, the backward driving data, and the backward target weight parameter to obtain backward regression data.

[0046] Determining the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data includes:

[0047] Based on the forward regression data and the second regression data, the forward distance between the first vehicle and the vehicle in front is determined; based on the backward regression data and the second regression data, the backward distance between the first vehicle and the vehicle in back is determined.

[0048] This application also provides a vehicle data processing apparatus, the apparatus comprising:

[0049] The driving data acquisition module is used to acquire first driving data of the first vehicle and second driving data of the second vehicle; the first vehicle and the second vehicle have a driving sequence on the target road;

[0050] The road surface data acquisition module is used to acquire target road surface data collected by multiple sensors related to the first vehicle for the target road.

[0051] The target weight acquisition module is used to acquire, according to the driving sequence, the first target weight parameter obtained by pre-training corresponding to the second vehicle and the second target weight parameter obtained by pre-training corresponding to each of the sensors;

[0052] The processing module is used to perform regression analysis processing based on the first driving data, the second driving data, and the first target weight parameter to obtain first regression data; and to perform regression analysis processing based on each of the road surface data and each of the second target weight parameters to obtain second regression data.

[0053] The distance determination module is used to determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0054] In one embodiment, the second vehicle includes a vehicle traveling in front of the first vehicle and a vehicle traveling behind the first vehicle. The vehicle traveling in front is positioned ahead of the first vehicle on the target road, and the vehicle traveling behind the first vehicle is positioned behind the first vehicle on the target road. The second driving data includes forward driving data of the vehicle traveling in front and backward driving data of the vehicle traveling behind. The first target weight parameter includes forward target weight parameters of the vehicle traveling in front and backward target weight parameters of the vehicle traveling behind. The first regression data includes forward regression data of the vehicle traveling in front and backward regression data of the vehicle traveling behind. The vehicle distance includes the forward distance between the first vehicle and the vehicle traveling in front, and the backward distance between the first vehicle and the vehicle traveling behind.

[0055] In one embodiment, the processing module is further configured to perform regression analysis processing based on the first driving data, the forward driving data, and the forward target weight parameter to obtain forward regression data; and to perform regression analysis processing based on the first driving data, the backward driving data, and the backward target weight parameter to obtain backward regression data.

[0056] The distance determination module is further configured to determine the forward distance between the first vehicle and the forward vehicle based on the forward regression data and the second regression data; and to determine the backward distance between the first vehicle and the backward vehicle based on the backward regression data and the second regression data.

[0057] This application also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0058] Acquire first driving data of a first vehicle and second driving data of a second vehicle; the first vehicle and the second vehicle have a driving order on the target road; acquire target road surface data collected by multiple sensors related to the first vehicle for the target road; according to the driving order, acquire the first target weight parameter pre-trained for the second vehicle and the second target weight parameter pre-trained for each of the sensors; perform regression analysis based on the first driving data, the second driving data, and the first target weight parameter to obtain first regression data; perform regression analysis based on each road surface data and the second target weight parameter to obtain second regression data; determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0059] This application also provides a computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0060] Acquire first driving data of a first vehicle and second driving data of a second vehicle; the first vehicle and the second vehicle have a driving order on the target road; acquire target road surface data collected by multiple sensors related to the first vehicle for the target road; according to the driving order, acquire the first target weight parameter pre-trained for the second vehicle and the second target weight parameter pre-trained for each of the sensors; perform regression analysis based on the first driving data, the second driving data, and the first target weight parameter to obtain first regression data; perform regression analysis based on each road surface data and the second target weight parameter to obtain second regression data; determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0061] This application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0062] Acquire first driving data of a first vehicle and second driving data of a second vehicle; the first vehicle and the second vehicle have a driving order on the target road; acquire target road surface data collected by multiple sensors related to the first vehicle for the target road; according to the driving order, acquire the first target weight parameter pre-trained for the second vehicle and the second target weight parameter pre-trained for each of the sensors; perform regression analysis based on the first driving data, the second driving data, and the first target weight parameter to obtain first regression data; perform regression analysis based on each road surface data and the second target weight parameter to obtain second regression data; determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0063] The aforementioned vehicle data processing method, apparatus, computer equipment, storage medium, and computer program product acquire second driving data for a first vehicle and a second vehicle with a driving sequence on a target road, as well as a pre-trained first target weight parameter corresponding to the second vehicle. Regression analysis is performed based on the first driving data, second driving data, and first target weight parameter, using the first target weight parameter as a regression coefficient to regress the first and second driving data, obtaining the correlation between the vehicle distance to be solved and the driving data of each vehicle, i.e., the first regression data. Target road surface data collected by multiple sensors related to the first vehicle on the target road are acquired, along with pre-trained second target weight parameters corresponding to each sensor. Regression analysis is performed based on the road surface data and second target weight parameters, using the second target weight parameter as a regression coefficient to regress the road surface data of each sensor, obtaining the correlation between the vehicle distance to be solved and the data collected by each sensor, i.e., the second regression data. Based on the correlation between the vehicle distance to be solved and the driving data of each vehicle, and the correlation with the data collected by each sensor, the vehicle distance between the first and second vehicles can be calculated more accurately. Attached Figure Description

[0064] Figure 1 This is an application environment diagram of a vehicle data processing method in one embodiment;

[0065] Figure 2 This is a flowchart illustrating a vehicle data processing method in one embodiment;

[0066] Figure 3 This is a schematic diagram of the interface between a sample vehicle and a sample sensor on a sample road in one embodiment.

[0067] Figure 4This is a schematic diagram illustrating the construction of a second correlation covariance matrix based on road surface data from various samples in one embodiment.

[0068] Figure 5 This is a flowchart illustrating the weight parameter adjustment process during cross-iteration in one embodiment.

[0069] Figure 6 This is a flowchart illustrating a vehicle data processing method in another embodiment;

[0070] Figure 7 This is a schematic diagram of the interface for determining forward and backward distances in one embodiment;

[0071] Figure 8 This is a flowchart illustrating a vehicle data processing method in another embodiment;

[0072] Figure 9 This is a structural block diagram of a vehicle data processing device in one embodiment;

[0073] Figure 10 This is a structural block diagram of the vehicle data processing device in another embodiment;

[0074] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0075] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0076] The embodiments of this application can be applied to various scenarios, including but not limited to cloud technology, artificial intelligence, smart transportation, assisted driving, autonomous driving, and data mining. For example, the Intelligent Transportation System (ITS) applied in the field of smart transportation is also known as the Intelligent Transportation System, and the Intelligent Vehicle Infrastructure Cooperative System (IVICS) is also known as the vehicle-road cooperative system.

[0077] The solutions provided in this application relate to vehicle data processing methods in the transportation field, which are specifically illustrated through the following embodiments.

[0078] The vehicle data processing method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102, deployed on the first sample vehicle, communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or placed in the cloud or on another server. Both terminal 102 and server 104 can independently execute the vehicle data processing method provided in this embodiment. Terminal 102 and server 104 can also work together to execute the vehicle data processing method provided in this embodiment. When terminal 102 and server 104 work together to execute the vehicle data processing method provided in this embodiment, terminal 102 acquires first sample driving data of the first sample vehicle and second sample driving data of the second sample vehicle; the first and second sample vehicles have a driving sequence on the sample road; terminal 102 acquires sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road. Terminal 102 sends the first sample driving data, the second sample driving data, and the sample road surface data to server 104. Server 104 determines a first correlation between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data, and determines a second correlation between multiple sample sensors based on multiple sample road surface data. Server 104 obtains the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor. Based on the first initial weight parameters and each second initial weight parameter, combined with the first sample driving data, the second sample driving data, each sample road surface data, the first correlation, and the second correlation, it performs cross-iterative weight parameter adjustments to obtain the first target weight parameters corresponding to the second sample vehicle and the second target weight parameters corresponding to each sample sensor. The first target weight parameters and the second target weight parameters are used to determine the distance between at least two vehicles with a driving sequence. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, smart voice interaction devices, smart home appliances, in-vehicle terminals, aircraft, portable wearable devices, etc. The in-vehicle terminal, also known as an in-vehicle intelligent terminal or in-vehicle terminal, refers to the in-vehicle infotainment product installed in the vehicle. Functionally, the in-vehicle terminal enables information communication between people and vehicles, and between vehicles and the outside world (vehicle-to-vehicle communication). Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. Terminal 102 and server 104 can be directly or indirectly connected via wired or wireless communication, which is not limited herein.

[0079] It should be noted that the quantities of "multiple" mentioned in the embodiments of this application all refer to the quantity of "at least two".

[0080] In one embodiment, such as Figure 2 As shown, a vehicle data processing method is provided, which is applied to a computer device (the computer device may be...) Figure 1 The instructions, which involve the terminal or server in the process, include the following steps:

[0081] Step S202: Obtain the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road.

[0082] Here, the first sample vehicle and the second sample vehicle refer to the vehicles for which distance prediction needs to be performed. The second sample vehicle includes at least one of the forward sample vehicles or the backward sample vehicles of the first sample vehicle.

[0083] The first sample driving data refers to the driving data related to the first sample vehicle, and may include data in at least one dimension. The data in at least one dimension includes at least one of the following: data of the first sample vehicle in the vehicle driving dimension, or vehicle attribute data of the first sample vehicle in the vehicle attribute dimension.

[0084] The second sample driving data refers to driving data related to the second sample vehicle, and may include data in at least one dimension. The data in at least one dimension includes at least one of the following: data of the second sample vehicle in the vehicle driving dimension, or vehicle attribute data of the second sample vehicle in the vehicle attribute dimension.

[0085] Among them, the vehicle driving dimension includes at least one of the following: vehicle speed dimension, vehicle distance dimension, or vehicle POI (Point of Interest) dimension.

[0086] Data in the vehicle driving dimension includes at least one of the following: vehicle speed, speeds of vehicles in front and behind, road surface images collected by the vehicle, distances to vehicles in front and behind, distances to vehicles in the left and right, or vehicle POI data. Vehicle attribute data in the vehicle attribute dimension includes at least one of the following: vehicle type, vehicle length, height, width, vehicle capacity, actual number of passengers, maximum speed, or engine displacement.

[0087] Specifically, data is collected from the first sample vehicle in the vehicle driving dimension to obtain the corresponding first sample driving data. Further, data is collected from the first sample vehicle in the vehicle attribute dimension, that is, the vehicle's own attribute data is obtained to obtain vehicle attribute data. The relevant data collected in the vehicle driving dimension and the vehicle attribute data are used as the first sample driving data of the first sample vehicle.

[0088] Data was collected from the second sample vehicle along the vehicle driving dimension to obtain corresponding second sample driving data. Further, data was collected from the second sample vehicle along the vehicle attribute dimension, i.e., acquiring the vehicle's own attribute data to obtain vehicle attribute data. The relevant data collected along the vehicle driving dimension and the vehicle attribute data will be used as the second sample driving data for the second sample vehicle.

[0089] The computer equipment can determine that there are first sample vehicles and second sample vehicles traveling in a driving order on the sample road, and obtain the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle.

[0090] Step S204: Obtain sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road.

[0091] Among them, the sample sensor refers to the sensor installed in the sample road, which is used to collect data on the vehicles in the sample road.

[0092] Multiple sample sensors associated with the first sample vehicle include a target sensor closest to the driving position of the first sample vehicle, and at least one of a forward sensor or a backward sensor corresponding to the target sensor. The forward sensor is positioned before the target sensor on the sample road, and the backward sensor is positioned after the target sensor on the sample road.

[0093] Alternatively, multiple sample sensors associated with the first sample vehicle, including a target sensor within a preset range of the first sample vehicle, and at least one of a forward sensor or a backward sensor corresponding to the target sensor.

[0094] For example, multiple sample sensors associated with the first sample vehicle, including a target sensor that the first sample vehicle is passing by, and at least one of a forward sensor or a backward sensor corresponding to the sample sensor that the first sample vehicle is passing by.

[0095] Sample road surface data refers to the data obtained by sample sensors collecting data on sample roads and vehicles within those roads. Sample road surface data may include data in at least one dimension, such as road driving data in the road driving dimension of vehicles traveling on the sample road, or road attribute data in the road attribute dimension of the sample road.

[0096] Road driving data in the road driving dimension includes at least one of the following: vehicle speed, vehicle speed in front and behind, distance between vehicles in front and behind, distance between vehicles in the left and right, or vehicle POI (Point of Interest) data.

[0097] Road surface attribute data in the road surface attribute dimension includes at least one of the following: road surface images, average vehicle speed, speed limit data, speeding data, traffic accident images, traffic flow data, waiting time, or road segment POI information.

[0098] Specifically, sensors in the sample road collect vehicle driving data in the road surface driving dimension, obtaining road surface driving data in this dimension. Sensors in the sample road can also collect data in the road surface attribute dimension, obtaining road surface attribute data.

[0099] For example, a sample sensor collects vehicle driving data from a first sample vehicle, obtaining road surface driving data corresponding to the first sample vehicle, and also collects road surface attribute data of the sample road as the first sample vehicle passes. At least one of the road surface driving data or road surface attribute data is used as the sample road surface data corresponding to the sample sensor. Each sample sensor performs the same data acquisition process on each passing vehicle, thus obtaining the sample road surface data corresponding to each sample sensor.

[0100] The computer equipment identifies the target sensor closest to the driving position of the first sample vehicle on the sample road, and at least one of the forward or backward sensors corresponding to the target sensor. The target sensor and the forward sensor are used as multiple sample sensors associated with the first sample vehicle, or the target sensor and the backward sensor are used as multiple sample sensors associated with the first sample vehicle, or the target sensor, the forward sensor, and the backward sensor are used as multiple sample sensors associated with the first sample vehicle.

[0101] In this embodiment, the computer device determines the target sensor within a preset range of the first sample vehicle on the sample road, and determines at least one of the forward sensor or the backward sensor corresponding to the target sensor.

[0102] The computer equipment can acquire the sample road surface data collected by each sample sensor, thus obtaining multiple sample road surface data.

[0103] like Figure 3As shown, vehicle-mounted terminals are installed on the first and second sample vehicles. On the sample road, the vehicle-mounted terminal of the first sample vehicle collects data to obtain the first sample driving data. The vehicle-mounted terminal of the second sample vehicle collects data to obtain the second sample driving data. Sensors on the sample road collect data to obtain sample road surface data. When the first sample vehicle passes sensor B, sensor B, its forward sensor (i.e., sensor C), and its rearward sensor (i.e., sensor A) are used as multiple sample sensors corresponding to the first sample vehicle.

[0104] Step S206: Determine the first correlation between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data; determine the second correlation between multiple sample sensors based on multiple sample road surface data.

[0105] The first correlation represents the data correlation or feature correlation between the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle, that is, the correlation between the first sample vehicle and the second sample vehicle in driving data, or the correlation in the features of driving data. The second correlation represents the data correlation or feature correlation between the sample road surface data of multiple sample sensors, that is, the correlation between the road surface data collected by multiple sample sensors, or the correlation in the features of road surface data.

[0106] The first and second correlations can be represented by a correlation matrix, which can specifically be represented by a covariance matrix. The first correlation can be represented by the covariance matrix between the first and second sample driving data. The second correlation can be represented by the covariance matrix between multiple sample road surface data.

[0107] Specifically, the computer equipment can determine the data correlation between the first sample driving data and the second sample driving data based on the first sample driving data and the second sample driving data, and use this data correlation as the first association relationship between the first sample vehicle and the second sample vehicle. The computer equipment can also determine the data correlation between multiple sample road surface data based on multiple sample road surface data and use this data correlation as the second association relationship between multiple sample sensors.

[0108] In this embodiment, the computer device can construct a first correlation matrix based on the first sample driving data and the second sample driving data, and use the first correlation matrix to represent the first correlation relationship between the first sample vehicles and the second sample vehicles. The computer device can also construct a second correlation matrix based on multiple sample road surface data, and use the second correlation matrix to represent the second correlation relationship between multiple sample sensors.

[0109] For example, based on the first sample driving data and the second sample driving data, a first correlation covariance matrix is ​​constructed to represent the first correlation relationship between the first sample vehicles and the second sample vehicles. A second correlation covariance matrix is ​​constructed based on multiple sample road surface data to represent the second correlation relationship between multiple sample sensors.

[0110] Step S208: Obtain the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor.

[0111] Here, the first initial weight parameter refers to the initial value of the weight parameter corresponding to the second sample vehicle, which is adjusted during iteration. The second initial weight parameter refers to the initial value of the weight parameter corresponding to the sample sensor, which is also adjusted during iteration.

[0112] Specifically, the computer equipment can initialize the weight parameters of the second sample vehicle and the weight parameters of each sample sensor to obtain the first initial weight parameters corresponding to the second sample vehicle and the second initial weight parameters corresponding to each sample sensor.

[0113] Step S210: Based on the first initial weight parameter and each second initial weight parameter, and combining the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship and the second correlation relationship, the weight parameters are adjusted by cross-iteration to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

[0114] Here, the first target weight parameter refers to the target value of the weight parameter corresponding to the second sample vehicle, which is generated by adjusting the initial value during iteration. The second target weight parameter refers to the target value of the weight parameter corresponding to the sample sensor, which is also generated by adjusting the initial value during iteration.

[0115] Specifically, the computer device, based on the first initial weight parameter and each second initial weight parameter, combines the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship, and the second correlation relationship to perform cross-iterative weight parameter adjustments, thereby obtaining intermediate weight parameters for each iteration. These intermediate weight parameters include the first intermediate weight parameter corresponding to the second sample vehicle and the second intermediate weight parameter corresponding to each sample sensor.

[0116] In the first iteration, the computer device adjusts the first initial weight parameter and each second initial weight parameter based on the first initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship and the second correlation relationship, to obtain the first intermediate weight parameter of the second sample vehicle and the second intermediate weight parameter corresponding to each sample sensor.

[0117] Starting from the second iteration, the first intermediate weight parameter of the second sample vehicle obtained in the previous iteration is used as the first initial weight parameter of the second sample vehicle in the next iteration. The second intermediate weight parameter corresponding to each sample sensor obtained in the previous iteration is used as the second intermediate weight parameter corresponding to each sample sensor in the next iteration. The weight parameters are adjusted in the above manner until the iteration stopping condition is met, and then the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor are obtained.

[0118] The iteration stopping condition can be met if the number of iterations reaches a preset number, or if the difference between the predicted distance and the expected distance generated during the iteration is less than a difference threshold. The expected distance refers to the true distance between the first sample vehicle and the second sample vehicle, and is used as a distance label in the cross-iteration.

[0119] Cross-iteration weight parameter adjustment refers to adjusting both the first initial weight parameter and each of the second initial weight parameters in the same iteration. After completing the adjustment of the first initial weight parameter and each of the second initial weight parameters in the same iteration, the next iteration begins, and so on. That is, after obtaining the first intermediate weight parameter and each of the second intermediate weight parameters from the previous iteration, the next iteration begins, and the first intermediate weight parameter and each of the second intermediate weight parameters obtained in the previous iteration are adjusted in the next iteration until the iteration stopping condition is met, at which point the first target weight parameter and each of the second target weight parameters are obtained. The order of adjustment of the first and second initial weight parameters in the same iteration is not restricted.

[0120] The first objective weight parameter and each second objective weight parameter are used to determine the distance between at least two vehicles that have a driving sequence.

[0121] In this embodiment, feature extraction is performed on the first sample driving data and the second sample driving data to obtain the first sample feature corresponding to the first sample vehicle and the second sample feature corresponding to the second sample vehicle; feature extraction is performed on each sample road surface data to obtain the sample road surface feature corresponding to each sample sensor; based on the first initial weight parameter and each second initial weight parameter, combined with the first sample feature, the second sample feature, each sample road surface feature, the first correlation relationship and the second correlation relationship, the weight parameter is adjusted by cross-iteration to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

[0122] In this embodiment, first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle are acquired. These two vehicles have a driving order on the sample road. Based on the first and second sample driving data, a first correlation relationship is determined between the first and second sample vehicles, thereby determining the correlation between the driving data collected by each vehicle. Sample road surface data collected by multiple sample sensors associated with the first sample vehicle on the sample road are acquired. Based on this sample road surface data, a second correlation relationship is determined between the multiple sample sensors, thereby determining the correlation between the multiple sample sensors on the road surface data.

[0123] Based on the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor, and combining the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship, and the second correlation relationship, cross-iterative weight parameter adjustments are performed. This allows for weight calculation by combining the correlation between the driving data of each vehicle and the correlation between the road surface data collected by each sensor, resulting in the first target weight parameters of the second sample vehicle and the second target weight parameters of each sample sensor. This avoids the problem of inaccurate weight parameters caused by the repeated superposition and calculation of the same data, effectively improving the accuracy of weight parameter calculation. Using the first target weight parameters and the second target weight parameters to determine the distance between at least two vehicles with a driving sequence can effectively improve the accuracy of vehicle distance prediction.

[0124] In one embodiment, the second sample vehicle includes at least one of the forward sample vehicle or the backward sample vehicle of the first sample vehicle; the second sample driving data includes at least one of the forward sample driving data of the forward sample vehicle or the backward sample driving data of the backward sample vehicle; the first association includes at least one of the forward association between the first sample vehicle and the forward sample vehicle, or the backward association between the first sample vehicle and the backward sample vehicle; the first initial weight parameter includes at least one of the forward initial weight parameter of the forward sample vehicle, or the backward initial weight parameter of the backward sample vehicle; the first target weight parameter includes at least one of the forward target weight parameter of the forward sample vehicle, or the backward target weight parameter of the backward sample vehicle.

[0125] In one embodiment, the second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle. The forward sample vehicle is positioned before the first sample vehicle on the sample road, and the backward sample vehicle is positioned after the first sample vehicle on the sample road. The second sample driving data includes forward sample driving data of the forward sample vehicle and backward sample driving data of the backward sample vehicle. The first association includes a forward association between the first sample vehicle and the forward sample vehicle, and a backward association between the first sample vehicle and the backward sample vehicle. The first initial weight parameter includes a forward initial weight parameter of the forward sample vehicle and a backward initial weight parameter of the backward sample vehicle. The first target weight parameter includes a forward target weight parameter of the forward sample vehicle and a backward target weight parameter of the backward sample vehicle.

[0126] Specifically, the process involves acquiring first sample driving data of the first sample vehicle, forward sample driving data of the sample vehicles preceding the first sample vehicle, and backward sample driving data of the sample vehicles following the first sample vehicle. The forward sample vehicles are positioned before the first sample vehicle on the sample road, and the backward sample vehicles are positioned after the first sample vehicle on the sample road. Sample road surface data collected by multiple sample sensors associated with the first sample vehicle are also acquired. Based on the first sample driving data and the forward sample driving data, a forward correlation is determined between the first sample vehicle and the forward sample vehicles. Based on the first sample driving data and the backward sample driving data, a backward correlation is determined between the first sample vehicle and the backward sample vehicles. Finally, a second correlation is determined among the multiple sample sensors based on the multiple sample road surface data.

[0127] The system obtains the forward initial weight parameters of the forward sample vehicles, the backward initial weight parameters of the backward sample vehicles, and the second initial weight parameters of each sample sensor. Based on the forward initial weight parameters, backward initial weight parameters, and each second initial weight parameter, and combined with the first sample driving data, forward sample driving data, backward sample driving data, road surface data of each sample, forward correlation, backward correlation, and second correlation, the system performs cross-iterative weight parameter adjustments to obtain the forward target weight parameters corresponding to the forward sample vehicles, the backward target weight parameters corresponding to the backward sample vehicles, and the second target weight parameters corresponding to each sample sensor. The forward target weight parameters, backward target weight parameters, and second target weight parameters are used to determine the distance between at least three vehicles with a driving sequence.

[0128] In one embodiment, determining a first association between the first sample vehicle and the second sample vehicle based on first sample driving data and second sample driving data includes:

[0129] Based on the first sample driving data, feature extraction is performed on multiple first dimensions to obtain the first sample feature of the first sample driving data on each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension; based on the second sample driving data, feature extraction is performed on multiple first dimensions to obtain the second sample feature of the second sample driving data on each first dimension; based on each first sample feature and each second sample feature, the first association relationship between the first sample vehicle and the second sample vehicle is determined.

[0130] The vehicle driving dimension includes at least one of the following: vehicle speed, speeds of vehicles in front and behind, road surface images collected by the vehicle, distances to vehicles in front and behind, distances to vehicles to the left and right, or vehicle POI. The vehicle attribute dimension includes at least one of the following: vehicle type, vehicle length, height, width, vehicle capacity, actual number of passengers, maximum speed, or engine displacement.

[0131] Specifically, the first sample driving data includes data in both the vehicle driving dimension and the vehicle attribute dimension. The computer equipment extracts features from the first sample driving data in both dimensions to obtain the vehicle driving features corresponding to the first sample driving data in the vehicle driving dimension and the vehicle attribute features corresponding to the vehicle attribute dimension. The computer equipment uses the vehicle driving features and vehicle attribute features corresponding to the first sample driving data as the first sample features of the first sample driving data in the corresponding first dimension, that is, as the first sample features of the first sample vehicle in the corresponding first dimension.

[0132] The second sample driving data includes data in both the vehicle driving dimension and the vehicle attribute dimension. The computer equipment extracts features from the second sample driving data in both dimensions to obtain the vehicle driving features corresponding to the vehicle driving dimension and the vehicle attribute features corresponding to the vehicle attribute dimension. The computer equipment uses these vehicle driving features and vehicle attribute features as the second sample features of the second sample driving data in the corresponding second dimension, i.e., as the second sample features of the second sample vehicle in the corresponding second dimension.

[0133] The computer equipment can construct a first correlation matrix based on the features of each first sample and each second sample, and use the first correlation matrix to characterize the first association relationship between the first sample vehicles and the second sample vehicles.

[0134] In this embodiment, the computer device can determine the sub-association relationships between a first sample feature and each second sample feature, and iterate through each first sample feature to obtain the sub-association relationships between each first sample feature and each second sample feature. Based on these sub-association relationships, a first association relationship between the first sample vehicles and the second sample vehicles is constructed. The sub-association relationships between the first sample features and the second sample features can be characterized by covariance, so the first correlation matrix can be a correlation covariance matrix.

[0135] In this embodiment, vehicle driving characteristics include at least one of the following: vehicle speed characteristics, forward vehicle speed characteristics, backward vehicle speed characteristics, road surface image characteristics collected by the vehicle, forward and backward vehicle distance characteristics, left and right vehicle distance characteristics, or vehicle POI (Point of Interest) characteristics. Vehicle attribute characteristics include at least one of the following: vehicle type characteristics, vehicle volume characteristics, vehicle passenger capacity characteristics, vehicle actual passenger capacity characteristics, vehicle maximum speed characteristics, or vehicle engine displacement characteristics. Vehicle volume characteristics include features such as length, height, and width.

[0136] In this embodiment, the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension. Feature extraction is performed on the first sample driving data for multiple first dimensions to extract the first sample features of the first sample driving data in the vehicle driving dimension and vehicle attribute dimension, respectively. Feature extraction is then performed on the second sample driving data for multiple first dimensions to obtain the second sample features of the second sample driving data in the vehicle driving dimension and vehicle attribute dimension, respectively. This allows for the accurate construction and determination of the first association relationship between the first sample vehicle and the second sample vehicle based on their features in the vehicle driving dimension and vehicle attribute dimension. Thus, the association relationship accurately characterizes the feature correlation between the first sample vehicle and the second sample vehicle in different dimensions.

[0137] In one embodiment, determining a first association relationship between the first sample vehicles and the second sample vehicles based on each first sample feature and each second sample feature includes:

[0138] Determine the sub-association relationships between each first sample feature and each second sample feature; based on each sub-association relationship, determine the first association relationship between the first sample vehicles and the second sample vehicles.

[0139] Among them, the sub-association relationship represents the feature correlation between the features of the first sample and the features of the second sample.

[0140] Specifically, the computer device can determine the feature correlation between a first sample feature and each second sample feature, and characterize the sub-association relationship between the first sample feature and the second sample feature through the feature correlation.

[0141] The computer device iterates through each first sample feature to obtain the feature correlation between each first sample feature and each second sample feature, thus obtaining the sub-association relationships. A first correlation matrix is ​​constructed based on these correlations.

[0142] In this embodiment, the sub-association relationship between the first sample feature and the second sample feature can be characterized by the first covariance, and the first correlation matrix can be the first correlation covariance matrix. The computer device calculates the first covariance between each first sample feature and each second sample feature, and constructs the first correlation covariance matrix between the first sample vehicles and the second sample vehicles based on each first covariance.

[0143] In this embodiment, the sub-association relationship between each first sample feature and each second sample feature is determined, so as to characterize the correlation between any two sample features through the sub-association relationship, thereby accurately reflecting the feature correlation between the first sample vehicle and the second sample vehicle on each feature based on the first association relationship determined by each sub-association relationship.

[0144] In one embodiment, determining a second correlation between multiple sample sensors based on multiple sample road surface data includes:

[0145] Based on each sample road surface data, multiple second-dimensional features are extracted to obtain the sample road surface features corresponding to each sample road surface data in each second dimension; the multiple second dimensions include road surface driving dimension and road surface attribute dimension; based on the multiple sample road surface features corresponding to each sample road surface data, the second correlation relationship between multiple sample sensors is determined.

[0146] The road travel dimension includes at least one of the following: vehicle speed dimension, vehicle distance dimension, or vehicle POI dimension. The vehicle speed dimension data includes at least one of the following: the speed of passing vehicles, or the speeds of vehicles in front of or behind a passing vehicle. The vehicle distance dimension data includes at least one of the following: the distance between a passing vehicle and vehicles in front of or behind, or the distance between a passing vehicle and vehicles in the left or right directions.

[0147] Road surface attribute dimensions include at least one of the following: image dimension, road surface speed limit dimension, traffic flow dimension, duration dimension, or road segment POI dimension.

[0148] Image-level data includes at least one of the following: images of the road surface or images of traffic accidents on that road segment. Speed ​​limit-level data includes at least one of the following: average vehicle speed, speed limit data, or speeding data. Traffic flow-level data includes traffic flow data. Duration-level data includes vehicle waiting times.

[0149] Specifically, the sample road surface data includes data in both the road surface driving dimension and the road surface attribute dimension. The computer equipment extracts features from each sample road surface data in both dimensions, obtaining the corresponding road surface driving features and road surface attribute features for each sample road surface data in the driving dimension. The computer equipment then uses the driving and attribute features corresponding to the same sample road surface data as the sample road surface features in the corresponding second dimension, i.e., as the sample road surface features of the sample sensor in the corresponding second dimension.

[0150] The computer equipment can construct a second correlation matrix based on the road surface characteristics of each sample, and use this second correlation matrix to characterize the second correlation relationship between multiple sample sensors.

[0151] In this embodiment, a second correlation relationship between multiple sample sensors is determined based on multiple sample road features corresponding to each sample road surface data, including: determining the sub-correlation relationship between each sample road surface feature of each sample road surface data; and determining the second correlation relationship between multiple sample sensors based on each sub-correlation relationship.

[0152] Computer equipment can determine the sub-correlation relationships between a sample road surface feature from one sample sensor and each sample road surface feature from other sample sensors. By traversing each sample road surface feature of the sample sensors, the sub-correlation relationships between each sample road surface feature and other sample road surface features are obtained. Based on these sub-correlation relationships, a second correlation relationship is determined among multiple sample sensors.

[0153] In this embodiment, the sub-association relationship between the first sample feature and the second sample feature can be called the first sub-association relationship, and the sub-association relationship between the various sample road surface features can be called the second sub-association relationship.

[0154] In this embodiment, the sub-correlation relationship between the road surface features of each sample can be characterized by the second covariance, and the second correlation relationship between multiple sample sensors can be characterized by the second correlation covariance matrix composed of each second covariance.

[0155] The characteristics of each sample road surface data in each second dimension can be as follows: Figure 4 As shown in the table, for example, sample road surface data 1 corresponds to 8 sample road surface features, namely vehicle speed feature 1, vehicle distance feature 1, vehicle point of interest feature 1, image feature 1 of the acquired image, road speed limit feature 1, traffic flow feature 1, waiting time feature 1, and road segment point of interest feature 1. Based on the 8 sample road surface features corresponding to sample road surface data 1, 2, and 3 respectively, a system can be constructed as follows: Figure 4 The second correlation covariance matrix shown represents Ω R .

[0156] In this embodiment, the multiple second dimensions include road surface driving dimension and road surface attribute dimension. Based on each sample road surface data, features of multiple second dimensions are extracted to obtain the features corresponding to each sample sensor in the road surface driving dimension and road surface attribute dimension. This enables the accurate construction of the second association relationship of multiple sample sensors based on the features corresponding to each sample sensor in the road surface driving dimension and road surface attribute dimension, thereby accurately characterizing the feature correlation of multiple sample sensors in different dimensions through the second association relationship.

[0157] In one embodiment, such as Figure 5 As shown, based on the first initial weight parameter and each second initial weight parameter, and combining the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship, and the second correlation relationship, the weight parameters are adjusted through cross-iteration to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor, including:

[0158] Step S502: Based on the first initial weight parameter and each second initial weight parameter, and combined with the first sample driving data, the second sample driving data, and the road surface data of each sample, determine the initial predicted distance between the first sample vehicle and the second sample vehicle.

[0159] Specifically, the computer equipment can perform regression analysis based on a first initial weight parameter, first sample driving data, and second sample driving data to obtain first sample regression data. Regression analysis is then performed based on each second initial weight parameter and each sample road surface data to obtain second sample regression data. Based on the first and second sample regression data, the initial predicted distance between the first and second sample vehicles is determined.

[0160] In this embodiment, the computer device can multiply the first sample driving data and the second sample driving data by the first initial weight parameter, and sum the two products to obtain the first sample regression data. For each sample sensor, the sample road surface data of a single sample sensor is multiplied by the second initial weight parameter of that sample sensor to obtain the product corresponding to each sample sensor, and the products are summed to obtain the second sample regression data.

[0161] The computer equipment can sum the regression data of the first sample and the regression data of the second sample to obtain the initial predicted distance between the first sample vehicles and the second sample vehicles. The initial predicted distance is an estimated value of the distance between the first sample vehicles and the second sample vehicles, based on the first initial weight parameter and each second initial weight parameter.

[0162] Step S504: Obtain the expected distance between the first sample vehicle and the second sample vehicle. Based on the first initial weight parameter and the initial predicted distance, and combining the expected distance, the first sample driving data, the second sample driving data and the first correlation relationship, perform iterative processing of the weight parameter to obtain the first target weight parameter corresponding to the second sample vehicle.

[0163] The expected distance refers to the true distance between the first sample vehicle and the second sample vehicle. This expected distance is used as a distance label in the iterative processing of the weight parameters.

[0164] Specifically, the computer equipment acquires the expected distance between the first sample vehicle and the second sample vehicle, and determines the distance difference between the expected distance and the initial predicted distance. Based on the first initial weight parameter, the distance difference, the first sample driving data, the second sample driving data, and the first correlation relationship, iterative processing is performed on the weight parameter of the second sample vehicle to obtain the first intermediate weight parameter corresponding to the second sample vehicle in each iteration. After obtaining the first intermediate weight parameter in the previous iteration, the next iteration begins. In the next iteration, the first intermediate weight parameter obtained in the previous iteration is used as the first initial weight parameter in the next iteration, and the weight parameter of the second sample vehicle is adjusted according to the above processing method until the iteration stopping condition is met, thus obtaining the first target weight parameter corresponding to the second sample vehicle.

[0165] In this embodiment, the computer device determines the first product of the distance difference, the first sample driving data, the second sample driving data, and the first correlation relationship. The difference between the first initial weight parameter and the first product is used as the first intermediate weight parameter, which is the first intermediate weight parameter of the second sample vehicle after iterative processing. Since the initial predicted distance is calculated based on the first initial weight parameter and each second initial weight parameter, and the first initial weight parameter and each second initial weight parameter are used to obtain the first intermediate weight parameter and the second intermediate weight parameter after iterative processing, in the next iteration, the first intermediate weight parameter obtained in the previous iteration is used as the first initial weight parameter in the next iteration, and each second intermediate weight parameter obtained in the previous iteration is used as the second intermediate weight parameter in the next iteration. This updates the initial predicted distance in the next iteration, and then calculates the updated distance difference based on the expected distance and the updated initial predicted distance. The difference between the product of the updated distance difference, the first sample driving data, the second sample driving data, and the first correlation relationship, and the first initial weight parameter is used as the first intermediate weight parameter for the next iteration. This process continues until the iteration stopping condition is met, at which point the first target weight parameter corresponding to the second sample vehicle is obtained.

[0166] In one embodiment, the computer device acquires a first learning rate corresponding to the second sample vehicle, where the first learning rate is a pre-set empirical value. The learning rate, as an important hyperparameter in supervised learning and deep learning, determines whether and when the objective function converges to a local minimum. The computer device determines a first product of the first learning rate, distance difference, first sample driving data, second sample driving data, and a first association relationship. The difference between the first initial weight parameter and this first product is used as the updated weight parameter for the second sample vehicle, i.e., the weight parameter for the second sample vehicle obtained through iterative processing.

[0167] Step S506: Based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, the road surface data of each sample, and the second correlation, the weight parameters are iteratively processed to obtain the second target weight parameter corresponding to each sample sensor.

[0168] Specifically, the computer device iteratively processes the weight parameters of each sample sensor based on each second initial weight parameter, distance difference, road surface data of each sample, and second correlation, to obtain the second intermediate weight parameter corresponding to each sample sensor in each iteration. After obtaining the second intermediate weight parameter in the previous iteration, the next iteration begins. In the next iteration, the second intermediate weight parameter obtained in the previous iteration is used as the second initial weight parameter in the next iteration, and the weight parameter of each sample sensor is adjusted according to the above processing method until the iteration stopping condition is met, thus obtaining the second target weight parameter corresponding to each sample sensor.

[0169] In this embodiment, the computer device determines the second product of the distance difference, the road surface data of each sample, and the second correlation. The difference between the second initial weight parameter and the second product is used as the second intermediate weight parameter, which is the second intermediate weight parameter of each sample sensor obtained through iterative processing. Since the initial predicted distance is calculated based on the first initial weight parameter and each second initial weight parameter, and the first initial weight parameter and each second initial weight parameter are used to obtain the first intermediate weight parameter and the second intermediate weight parameter after iterative processing, in the next iteration, the first intermediate weight parameter obtained in the previous iteration is used as the first initial weight parameter in the next iteration, and each second intermediate weight parameter obtained in the previous iteration is used as the second initial weight parameter in the next iteration. This updates the initial predicted distance in the next iteration, and the updated distance difference is calculated based on the expected distance and the updated initial predicted distance. The difference between the updated distance difference, the second product of the road surface data of each sample, and the second correlation, and the second initial weight parameter is used as the second intermediate weight parameter for the next iteration. This process continues until the iteration stopping condition is met, resulting in the second target weight parameter corresponding to each sample sensor.

[0170] In one embodiment, the computer device acquires a second learning rate corresponding to the sample sensor, which is a pre-set empirical value. The first learning rate may be the same as or different from the second learning rate. The computer device determines a second product of the second learning rate, distance difference, each sample road surface data, and the second correlation, and uses the difference between the second initial weight parameter and the second product as a second intermediate weight parameter, i.e., the second intermediate weight parameter of each sample sensor obtained through iterative processing.

[0171] In this embodiment, based on the first initial weight parameter and each second initial weight parameter, and combining the first sample driving data, the second sample driving data, and each sample road surface data, the distance between the first sample vehicle and the second sample vehicle is predicted to obtain an initial predicted distance. The expected distance between the first sample vehicle and the second sample vehicle is obtained as a distance label in the iteration. This allows the difference between the actual distance and the distance predicted based on each weight parameter to be determined. Combining the driving data of each vehicle and the data correlation between them, the weight parameters of the second sample vehicle are iterated, thereby continuously optimizing the weight parameters of the second sample vehicle in multiple iterations, thus accurately obtaining the first target weight parameter of the second sample vehicle. Combining the sample road surface data collected by each sample sensor and the data correlation between them, the weight parameters of each sample sensor are iterated, thereby continuously optimizing the weight parameters of each sample sensor in multiple iterations, thus accurately obtaining the second target weight parameter corresponding to each sample sensor.

[0172] In one embodiment, the second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle; the forward sample vehicle is positioned ahead of the first sample vehicle on the sample road, and the backward sample vehicle is positioned behind the first sample vehicle on the sample road; the second sample driving data includes forward sample driving data of the forward sample vehicle and backward sample driving data of the backward sample vehicle; the initial prediction distance includes the forward initial prediction distance and the backward initial prediction distance between the forward sample vehicle and the first sample vehicle, respectively; the expected distance includes the forward expected distance and the backward expected distance between the forward sample vehicle and the first sample vehicle, respectively; the first association relationship includes the forward association relationship and the backward association relationship between the forward sample vehicle and the first sample vehicle, respectively; the first initial weight parameter includes the forward initial weight parameter of the forward sample vehicle and the backward initial weight parameter of the backward sample vehicle; the first target weight parameter includes the forward target weight parameter of the forward sample vehicle and the backward target weight parameter of the backward sample vehicle.

[0173] The initial predicted distance includes the forward initial predicted distance between the first sample vehicle and the forward sample vehicle, and the backward initial predicted distance between the first sample vehicle and the backward sample vehicle; the expected distance includes the forward expected distance between the first sample vehicle and the forward sample vehicle, and the backward expected distance between the first sample vehicle and the backward sample vehicle; the first association includes the forward association between the first sample vehicle and the forward sample vehicle, and the backward association between the first sample vehicle and the backward sample vehicle.

[0174] Specifically, the first sample driving data of the first sample vehicle, the forward sample driving data of the sample vehicles preceding the first sample vehicle, and the backward sample driving data of the sample vehicles following the first sample vehicle are obtained; the driving position of the forward sample vehicle on the sample road is before the driving position of the first sample vehicle, and the driving position of the backward sample vehicle on the sample road is after the driving position of the first sample vehicle.

[0175] Obtain the forward initial weight parameters of the forward sample vehicles, the backward initial weight parameters of the backward sample vehicles, and the second initial weight parameters of each sample sensor.

[0176] Based on the forward initial weight parameters, the backward initial weight parameters, and each second initial weight parameter, combined with the first sample driving data, forward sample driving data, backward sample driving data, and road surface data of each sample, the forward predicted distance between the first sample vehicle and the forward sample vehicle, and the backward predicted distance between the first sample vehicle and the backward sample vehicle are determined. The forward expected distance between the first sample vehicle and the forward sample vehicle, and the backward expected distance between the first sample vehicle and the backward sample vehicle are obtained. Based on the forward initial weight parameters, the backward initial weight parameters, the forward predicted distance, and the backward predicted distance, combined with the forward expected distance, the backward expected distance, the first sample driving data, the forward sample driving data, the backward sample driving data, the forward correlation relationship, and the backward correlation relationship, iterative processing of the weight parameters is performed to obtain the forward target weight parameters corresponding to the forward sample vehicle and the backward target weight parameters corresponding to the backward sample vehicle.

[0177] Based on each second initial weight parameter, forward prediction distance, and backward prediction distance, combined with forward expected distance, backward expected distance, road surface data of each sample, and second correlation, the weight parameters are iteratively processed to obtain the second target weight parameters corresponding to each sample sensor.

[0178] In this embodiment, based on the forward initial weight parameter, the backward initial weight parameter, and each second initial weight parameter, combined with the first sample driving data, the forward sample driving data, the backward sample driving data, and the road surface data of each sample, the distance between the first sample vehicle and the forward sample vehicle, and between the first sample vehicle and the backward sample vehicle are predicted to obtain the forward predicted distance and the backward predicted distance. By acquiring the forward and backward expected distances as distance labels in the iteration, the difference between the true forward distance and the forward distance predicted based on the weight parameters of the forward sample vehicles and the weight parameters of each sample sensor can be determined. Similarly, the difference between the true backward distance and the backward distance predicted based on the weight parameters of the backward sample vehicles and the weight parameters of each sample sensor can be determined. Combining the driving data of each vehicle and the data correlation between driving data, and the data correlation between sample road surface data collected by each sample sensor and the data correlation between sample road surface data, the weight parameters of the forward sample vehicles, backward sample vehicles, and each sample sensor are iterated. This process continuously optimizes each weight parameter in multiple iterations, thereby accurately obtaining the target weight parameters corresponding to each of the forward sample vehicles, backward sample vehicles, and each sample sensor.

[0179] In one embodiment, based on a first initial weight parameter and each second initial weight parameter, and combining first sample driving data, second sample driving data, and road surface data of each sample, the initial predicted distance between the first sample vehicle and the second sample vehicle is determined, including:

[0180] Based on the forward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, the forward sample driving data, and the road surface data of each sample, the forward initial predicted distance between the first sample vehicle and the forward sample vehicle is determined.

[0181] Based on the backward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, the backward sample driving data, and the road surface data of each sample, the backward initial predicted distance between the first sample vehicle and the backward sample vehicle is determined.

[0182] Specifically, the computer equipment can perform regression analysis based on the forward initial weight parameters, the first sample driving data, and the forward sample driving data to obtain forward sample regression data. Regression analysis is then performed based on each second initial weight parameter and each sample road surface data to obtain second sample regression data. Based on the forward sample regression data and the second sample regression data, the forward initial predicted distance between the first sample vehicle and the forward sample vehicle is determined.

[0183] The computer equipment can perform regression analysis based on the backward initial weight parameters, the first sample driving data, and the backward sample driving data to obtain backward sample regression data. Regression analysis is then performed based on each second initial weight parameter and each sample road surface data to obtain second sample regression data. Based on the backward sample regression data and the second sample regression data, the backward initial predicted distance between the first sample vehicles and the backward sample vehicles is determined.

[0184] In this embodiment, the computer device can multiply the first sample driving data and the forward sample driving data by the forward initial weight parameter, and sum the two products to obtain the forward sample regression data. For each sample sensor, the sample road surface data of a single sample sensor is multiplied by the second initial weight parameter of that sample sensor to obtain the product corresponding to each sample sensor. The products are then summed to obtain the second sample regression data. The computer device acquires a forward random sequence and sums the forward sample regression data, the second sample regression data, and the forward random sequence to obtain the forward initial predicted distance between the first sample vehicle and the forward sample vehicle. The forward random sequence indicates that the distance between the first sample vehicle and the forward sample vehicle follows a standard normal distribution.

[0185] The computer equipment can multiply the first sample driving data and the backward sample driving data by the backward initial weight parameter, and then sum the two products to obtain the backward sample regression data. The computer equipment acquires a backward random sequence, sums the backward sample regression data, the second sample regression data, and the backward random sequence to obtain the backward initial predicted distance between the first sample vehicle and the backward sample vehicle. The backward random sequence indicates that the distance between the first sample vehicle and the backward sample vehicle follows a standard normal distribution.

[0186] For example, a computer device can calculate the initial forward prediction distance using the following forward distance formula:

[0187]

[0188] Computer equipment can calculate the backward initial prediction distance using the following backward distance formula:

[0189]

[0190] In this embodiment, based on the forward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, forward sample driving data, and each sample road surface data, the forward initial predicted distance between the first sample vehicle and the forward sample vehicle can be accurately calculated. Based on the backward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, backward sample driving data, and each sample road surface data, the backward initial predicted distance between the first sample vehicle and the backward sample vehicle can be accurately calculated. Therefore, based on the predicted distance between the first sample vehicle and the forward and backward vehicles, the weight parameters of the forward and backward vehicles and the weight parameters of the sample sensors are adjusted to obtain the optimal solution for each weight parameter.

[0191] In one embodiment, based on a first initial weight parameter and an initial predicted distance, combined with the expected distance, first sample driving data, second sample driving data, and a first correlation, iterative processing of the weight parameter is performed to obtain the first target weight parameter corresponding to the second sample vehicle, including:

[0192] Based on the forward initial weight parameters and the forward initial predicted distance, combined with the forward expected distance, the first sample driving data, the forward sample driving data and the forward correlation, the weight parameters are iteratively processed to obtain the forward target weight parameters corresponding to the forward sample vehicles.

[0193] Based on the backward initial weight parameters and the backward initial predicted distance, combined with the backward expected distance, the first sample driving data, the backward sample driving data, and the backward correlation, the weight parameters are iteratively processed to obtain the backward target weight parameters corresponding to the backward sample vehicles.

[0194] Specifically, the computer equipment acquires the forward expected distance between the first sample vehicle and the forward sample vehicle, and determines the forward distance difference between the forward expected distance and the forward initial predicted distance. Based on the forward initial weight parameters, the forward distance difference, the first sample driving data, the forward sample driving data, and the forward correlation, iterative processing is performed on the weight parameters of the forward sample vehicles to obtain the forward intermediate weight parameters corresponding to the forward sample vehicles in each iteration. After obtaining the forward intermediate weight parameters in the previous iteration, the next iteration begins. In the next iteration, the forward intermediate weight parameters obtained in the previous iteration are used as the forward initial weight parameters in the next iteration, and the weight parameters of the forward sample vehicles are adjusted according to the above processing method until the iteration stopping condition is met, thus obtaining the forward target weight parameters corresponding to the forward sample vehicles.

[0195] The computer equipment acquires the expected backward distance between the first sample vehicle and the subsequent sample vehicles, and determines the backward distance difference between the expected backward distance and the initial backward predicted distance. Based on the initial backward weight parameters, the backward distance difference, the first sample driving data, the subsequent sample driving data, and the backward correlation, iterative processing is performed on the weight parameters of the subsequent sample vehicles to obtain the intermediate backward weight parameters corresponding to the subsequent sample vehicles in each iteration. After obtaining the intermediate backward weight parameters in the previous iteration, the next iteration begins. In the next iteration, the intermediate backward weight parameters obtained in the previous iteration are used as the initial backward weight parameters for the next iteration, and the weight parameters of the subsequent sample vehicles are adjusted according to the above processing method until the iteration stopping condition is met, at which point the target backward weight parameters corresponding to the subsequent sample vehicles are obtained.

[0196] In this embodiment, the computer device determines the forward distance difference, the first sample driving data, the forward sample driving data, and the forward product of the forward correlation. The difference between the forward initial weight parameter and this forward product is used as the forward intermediate weight parameter. Since the forward initial prediction distance is calculated based on the forward initial weight parameter and each second initial weight parameter, and the forward initial weight parameter and each second initial weight parameter are iteratively processed to obtain the forward intermediate weight parameter and the second intermediate weight parameter, in the next iteration, the forward intermediate weight parameter obtained in the previous iteration is used as the forward initial weight parameter in the next iteration, and each second intermediate weight parameter obtained in the previous iteration is used as each second initial weight parameter in the next iteration. This updates the forward initial prediction distance in the next iteration, and the updated forward distance difference is calculated based on the forward expected distance and the updated forward initial prediction distance. The difference between the updated forward distance difference, the first sample driving data, the forward sample driving data, and the forward correlation product, and the forward initial weight parameter is used as the forward intermediate weight parameter for the next iteration. This process continues until the iteration stopping condition is met, at which point the forward target weight parameters corresponding to the forward sample vehicles are obtained.

[0197] The iteration process of the rearward initial weight parameters of the following vehicles is similar to the iteration process of the forward initial weight parameters of the forward vehicles. By adjusting the data of all forward vehicles to the data of the following vehicles, the forward target weight parameters corresponding to the forward sample vehicles can be obtained.

[0198] In this embodiment, based on the initial forward weight parameters and the initial forward predicted distance, combined with the expected forward distance, the first sample driving data, the forward sample driving data, and the forward correlation, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of the forward sample vehicles are optimized by incorporating various data points, resulting in the optimal solution for the weight parameters corresponding to the forward sample vehicles, i.e., the forward target weight parameters. Similarly, based on the initial backward weight parameters and the initial backward predicted distance, combined with the expected backward distance, the first sample driving data, the backward sample driving data, and the backward correlation, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of the backward sample vehicles are optimized by incorporating various data points, resulting in the optimal solution for the weight parameters corresponding to the backward sample vehicles, i.e., the backward target weight parameters.

[0199] In one embodiment, based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, the road surface data of each sample, and the second correlation, iterative processing of the weight parameters is performed to obtain the second target weight parameters corresponding to each sample sensor, including:

[0200] Based on each second initial weight parameter, forward initial prediction distance, and backward initial prediction distance, combined with forward expected distance, backward expected distance, road surface data of each sample, and second correlation, the weight parameters are iteratively processed to obtain the second target weight parameters corresponding to each sample sensor.

[0201] Specifically, the computer equipment calculates the global initial predicted distance between forward and backward sample vehicles, and the global expected distance between them, based on the forward initial predicted distance, backward initial predicted distance, forward expected distance, and backward expected distance. The global distance difference is then calculated based on the global initial predicted distance and the global expected distance. This global distance difference characterizes the difference between the global expected distance and the global initial predicted distance.

[0202] The computer equipment iteratively processes the weight parameters of each sample sensor based on each second initial weight parameter, global distance difference, road surface data of each sample, and second correlation, obtaining the second intermediate weight parameter corresponding to each sample sensor in each iteration. After obtaining the second intermediate weight parameter in the previous iteration, the next iteration begins. In the next iteration, the second intermediate weight parameter obtained in the previous iteration is used as the second initial weight parameter, and the weight parameter of each sample sensor is adjusted according to the above processing method until the iteration stopping condition is met, thus obtaining the second target weight parameter corresponding to each sample sensor.

[0203] In this embodiment, based on each second initial weight parameter, forward initial prediction distance, and backward initial prediction distance, combined with forward expected distance and backward expected distance, each sample road surface data, and the second correlation, the weight parameters are iteratively processed. In this way, the weight parameters of each sample sensor are optimized by combining various data in multiple iterations, so as to obtain the optimal solution of the weight parameters corresponding to each sample sensor, that is, each second target weight parameter.

[0204] In one embodiment, based on each second initial weight parameter, the forward initial prediction distance, and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, the road surface data of each sample, and the second correlation, iterative processing of the weight parameters is performed to obtain the second target weight parameters corresponding to each sample sensor, including:

[0205] Determine the global expected distance based on the forward expected distance and the backward expected distance;

[0206] The global initial prediction distance is determined based on the forward initial prediction distance and the backward initial prediction distance;

[0207] Based on each second initial weight parameter and the global initial prediction distance, combined with the global expected distance, the road surface data of each sample, and the second correlation, the weight parameters are iteratively processed to obtain the second target weight parameter corresponding to each sample sensor.

[0208] Specifically, the computer device uses the sum of the forward initial prediction distance and the backward initial prediction distance as the global initial prediction distance between the forward and backward sample vehicles. It also uses the sum of the forward expected distance and the backward expected distance as the global expected distance between the forward and backward sample vehicles. The difference between the global expected distance and the global initial prediction distance is used as the global distance difference.

[0209] The computer equipment determines the second product of the global distance difference, the road surface data of each sample, and the second correlation. The difference between the second initial weight parameter and this second product is used as the second intermediate weight parameter, which is the second intermediate weight parameter of each sample sensor obtained through iterative processing. Since the global initial predicted distance is calculated based on the forward initial weight parameter, the backward initial weight parameter, and each second initial weight parameter, and these parameters are iteratively processed to obtain the forward intermediate weight parameter, the backward intermediate weight parameter, and the second intermediate weight parameter, respectively, in the next iteration, the forward intermediate weight parameter, the backward intermediate weight parameter, and each second intermediate weight parameter obtained in the previous iteration are used as the forward initial weight parameter, the backward initial weight parameter, and each second initial weight parameter in the next iteration, respectively. This updates the global initial predicted distance in the next iteration, and then calculates the updated global distance difference based on the global expected distance and the updated global initial predicted distance. The difference between the updated global distance difference, the second product of the road surface data of each sample, and the second correlation, and the second initial weight parameter is used as the second intermediate weight parameter for the next iteration. This process continues until the iteration stopping condition is met, at which point the second target weight parameters corresponding to each sample sensor are obtained.

[0210] In this embodiment, the global initial predicted distance between the first sample vehicle and the vehicles in front and behind is calculated based on the forward and backward expected distances. The global expected distance between the first sample vehicle and the vehicles in front and behind is then calculated based on the forward and backward initial predicted distances, thus simultaneously considering the impact of the vehicles in front and behind on the first sample vehicle. Based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, road surface data for each sample, and the second correlation relationship, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of each sample sensor are optimized using data from the first sample vehicle itself, the vehicles in front and behind, the correlation between vehicles, the correlation between sensors, and the vehicle data collected by each sample sensor. This yields the optimal solution for the weight parameters corresponding to each sample sensor, making the obtained second target weight parameters suitable for predicting the distance between the first vehicle and the vehicles in front and behind when the first vehicle simultaneously has vehicles in front and behind.

[0211] In one embodiment, such as Figure 6 As shown, a vehicle data processing method is provided, which is applied to a computer device (the computer device may be...) Figure 1 Taking a terminal or server as an example, the following steps are included:

[0212] Step S602: Obtain the first driving data of the first vehicle and the second driving data of the second vehicle; the first vehicle and the second vehicle have a driving order on the target road.

[0213] In this context, "first vehicle" and "second vehicle" refer to the vehicles requiring distance detection. The second vehicle includes at least one of the vehicles traveling in the direction of the first vehicle's movement, either forward or backward. The forward vehicle's position on the target road is ahead of the first vehicle's position, and the backward vehicle's position on the target road is behind the first vehicle's position. The order of travel refers to the sequence in which the vehicles travel on the road.

[0214] The first driving data refers to driving data related to the first vehicle, which may include data in at least one dimension. The data in at least one dimension includes at least one of the following: data of the first vehicle in the vehicle driving dimension, or vehicle attribute data of the first vehicle in the vehicle attribute dimension.

[0215] Second driving data refers to driving data related to the second vehicle, which may include data in at least one dimension. The data in at least one dimension includes at least one of the following: data on the second vehicle in the vehicle driving dimension, or vehicle attribute data on the second vehicle in the vehicle attribute dimension.

[0216] Data in the vehicle driving dimension includes at least one of the following: vehicle speed, speeds of vehicles in front and behind, road surface images collected by the vehicle, or vehicle POI data. Vehicle attribute data in the vehicle attribute dimension includes at least one of the following: vehicle type, vehicle length, height, width, vehicle capacity, actual number of passengers, maximum speed, or engine displacement.

[0217] Specifically, the first vehicle collects data along the vehicle driving dimension to obtain corresponding first driving data. Further, the first vehicle collects data along the vehicle attribute dimension, that is, it obtains the vehicle's own attribute data to obtain vehicle attribute data. The relevant data collected along the vehicle driving dimension and the vehicle attribute data are used as the first driving data of the first vehicle.

[0218] The second vehicle collects data along the vehicle driving dimension to obtain corresponding second driving data. Further, the second vehicle collects data along the vehicle attribute dimension, i.e., it acquires the vehicle's own attribute data to obtain vehicle attribute data. The relevant data collected along the vehicle driving dimension and the vehicle attribute data are used as the second driving data for the second vehicle.

[0219] The computer equipment can determine that there is a first vehicle and a second vehicle traveling in a sequence on the target road, and obtain the first driving data of the first vehicle and the second driving data of the second vehicle.

[0220] Step S604: Obtain target road surface data collected by multiple sensors associated with the first vehicle for the target road.

[0221] The sensor refers to the sensor installed on the target road to collect data on vehicles on the target road.

[0222] The system includes multiple sensors associated with the first vehicle, including a target sensor closest to the first vehicle's driving position, and at least one of a forward sensor or a backward sensor corresponding to the target sensor. The forward sensor is positioned ahead of the target sensor on the target road, and the backward sensor is positioned behind the target sensor on the target road.

[0223] Alternatively, multiple sensors associated with the first vehicle, including a target sensor within a preset range of the first vehicle, and at least one of a forward sensor or a backward sensor corresponding to the target sensor.

[0224] For example, multiple sensors associated with the first vehicle, including a target sensor that the first vehicle is passing, and at least one of a forward sensor or a backward sensor corresponding to the sensor that the first vehicle is passing.

[0225] Road surface data refers to data collected by sensors from vehicles passing through a target road. Road surface data can include data in at least one dimension, which includes at least one of the following: road surface driving data in the dimension of vehicle driving on the target road, or road surface attribute data in the dimension of road surface attributes.

[0226] Road driving data in the road driving dimension includes at least one of the following: vehicle speed of passing vehicles, speed of vehicles in front and behind, distance between vehicles in front and behind, distance between vehicles to the left and right, or vehicle POI data.

[0227] Road surface attribute data in the road surface attribute dimension includes at least one of the following: road surface images, average vehicle speed, speed limit data, speeding data, traffic accident images, traffic flow data, waiting time, or road segment POI information.

[0228] Specifically, sensors on the target road collect vehicle driving data in the road surface driving dimension, obtaining road surface driving data. Sensors on the target road can also collect data in the road surface attribute dimension, obtaining road surface attribute data.

[0229] For example, a sensor collects vehicle driving data from a passing vehicle to obtain road surface driving data corresponding to the vehicle, and also collects road surface attribute data corresponding to the target road at the time the vehicle passes. At least one of the road surface driving data or road surface attribute data is used as the road surface data corresponding to the sensor. Each sensor performs the same data collection and processing on passing vehicles, resulting in road surface data for each sensor.

[0230] The computer equipment identifies a target sensor on the target road that is closest to the driving position of the first vehicle, and at least one of a forward sensor or a backward sensor corresponding to the target sensor. The target sensor and the forward sensor may be considered as multiple sensors associated with the first vehicle, or the target sensor and the backward sensor may be considered as multiple sensors associated with the first vehicle, or the target sensor, the forward sensor, and the backward sensor may be considered as multiple sensors associated with the first vehicle.

[0231] In this embodiment, the computer device identifies a target sensor within a preset range of the first vehicle on the road, and identifies at least one of the forward or backward sensors corresponding to the target sensor.

[0232] The computer equipment can acquire the road surface data collected by each sensor, resulting in multiple road surface data sets.

[0233] Step S606: According to the driving sequence, obtain the first target weight parameters obtained by pre-training for the second vehicle and the second target weight parameters obtained by pre-training for each sensor.

[0234] Here, the first target weight parameter refers to the target value of the weight parameter corresponding to the second vehicle, and the second target weight parameter refers to the target value of the weight parameter corresponding to the sensor. The target value of this weight parameter is generated during pre-training by adjusting the initial values ​​of the weight parameters.

[0235] The first target weight parameter is obtained through iterative adjustment of the first initial weight parameter corresponding to the second sample vehicle during pre-training. The second target weight parameter is obtained through iterative adjustment of the second initial weight parameter corresponding to the sample sensor during pre-training.

[0236] The pre-training process includes: acquiring first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle; the first and second sample vehicles having a driving order on the sample road; acquiring sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road; determining a first correlation between the first and second sample vehicles based on the first and second sample driving data; determining a second correlation between multiple sample sensors based on the multiple sample road surface data; acquiring a first initial weight parameter for the second sample vehicle and a second initial weight parameter for each sample sensor; and performing cross-iterative weight parameter adjustments based on the first initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the second sample driving data, each sample road surface data, the first correlation, and the second correlation, to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor.

[0237] Specifically, the computer equipment obtains the first target weight parameters obtained through pre-training for the second vehicle, and the second target weight parameters obtained through pre-training for each sensor, according to the driving sequence.

[0238] Step S608: Perform regression analysis based on the first driving data, the second driving data, and the first target weight parameter to obtain the first regression data; perform regression analysis based on each road surface data and the second target weight parameter to obtain the second regression data.

[0239] Specifically, the computer device can multiply the first driving data and the second driving data by the first target weight parameter, and then sum the two products to obtain the first regression data. For each sensor, the road surface data of a single sensor is multiplied by the second target weight parameter of that sensor to obtain the product corresponding to each sensor, and the products are summed to obtain the second regression data.

[0240] Step S610: Determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0241] Specifically, the computer equipment can sum the first regression data and the second regression data to obtain the vehicle distance between the first vehicle and the second vehicle.

[0242] In this embodiment, the computer device sums the first regression data, the second regression data, and a preset random sequence to obtain the vehicle distance between the first vehicle and the second vehicle. The preset random sequence indicates that the vehicle distance follows a standard normal distribution.

[0243] In this embodiment, feature extraction is performed on the first driving data and the second driving data to obtain the first feature corresponding to the first vehicle and the second feature corresponding to the second vehicle; feature extraction is performed on each road surface data to obtain the road surface feature corresponding to each sensor; regression analysis is performed based on the first feature, the second feature and the first target weight parameter to obtain the first regression data; regression analysis is performed based on each road surface feature and the second target weight parameter to obtain the second regression data.

[0244] In this embodiment, second driving data for the first and second vehicles, which have a driving sequence on the target road, are acquired, along with pre-trained first target weight parameters for the second vehicle. Regression analysis is performed based on the first driving data, second driving data, and first target weight parameters, using the first target weight parameters as regression coefficients to regress the first and second driving data, thus obtaining the correlation between the vehicle distance to be solved and the driving data of each vehicle—this is the first regression data. Target road surface data collected by multiple sensors associated with the first vehicle on the target road are acquired, along with pre-trained second target weight parameters for each sensor. Regression analysis is performed based on the road surface data and second target weight parameters, using the second target weight parameters as regression coefficients to regress the road surface data of each sensor, thus obtaining the correlation between the vehicle distance to be solved and the data collected by each sensor—this is the second regression data. Based on the correlation between the vehicle distance to be solved and the driving data of each vehicle, and the correlation with the data collected by each sensor, the vehicle distance between the first and second vehicles can be calculated more accurately.

[0245] In one embodiment, the method further includes: providing a prompt when the distance between the first vehicle and the second vehicle meets the distance prompting condition.

[0246] Among these, distance indication conditions refer to conditions related to vehicle distance used to indicate the distance between vehicles. Specifically, a distance indication condition can be that the distance between vehicles is less than a distance threshold.

[0247] When the distance between the first vehicle and the second vehicle meets the distance warning condition, it indicates that the distance between the first vehicle and the second vehicle is relatively short and there is a safety hazard of vehicle collision. In this case, a voice prompt will be made through the vehicle terminal of the first vehicle, or the driver will be prompted through the screen of the vehicle terminal of the first vehicle, so as to effectively prevent the collision between the vehicles and improve traffic safety.

[0248] In one embodiment, the second vehicle includes the vehicle traveling in front of the first vehicle and the vehicle traveling behind the first vehicle. The vehicle traveling in front is positioned ahead of the first vehicle on the target road, and the vehicle traveling behind the first vehicle is positioned behind the first vehicle on the target road. The second driving data includes the forward driving data of the vehicle traveling in front and the backward driving data of the vehicle traveling behind. The first target weight parameter includes the forward target weight parameter of the vehicle traveling in front and the backward target weight parameter of the vehicle traveling behind. The first regression data includes the forward regression data of the vehicle traveling in front and the backward regression data of the vehicle traveling behind. The vehicle distance includes the forward distance between the first vehicle and the vehicle traveling in front, and the backward distance between the first vehicle and the vehicle traveling behind.

[0249] Specifically, the first driving data of the first vehicle, the forward driving data of the vehicle in front of the first vehicle, and the backward driving data of the vehicle behind the first vehicle are obtained; the driving position of the vehicle in front on the target road is before the driving position of the first vehicle, and the driving position of the vehicle behind on the target road is after the driving position of the first vehicle.

[0250] Acquire target road surface data collected by multiple sensors associated with the first vehicle for the target road. Following the driving sequence, acquire the pre-trained forward target weight parameters for the preceding vehicle, the pre-trained backward target weight parameters for the following vehicle, and the pre-trained second target weight parameters for each sensor.

[0251] Regression analysis is performed based on the first driving data, forward driving data, backward driving data, forward target weight parameters, and backward target weight parameters to obtain forward regression data for forward vehicles and backward regression data for backward vehicles. Regression analysis is then performed based on road surface data and the second target weight parameters to obtain second regression data. Based on the forward regression data, backward regression data, and second regression data, the forward distance between the first vehicle and the vehicle in front, and the backward distance between the first vehicle and the vehicle behind, are determined. In this embodiment, the driving data of the first vehicle, the vehicles in front of the first vehicle, and the vehicles behind the first vehicle, which have a driving sequence on the target road, are obtained, along with the pre-trained target weight parameters for the vehicles in front and behind. Regression analysis is performed based on the driving data of each vehicle and the corresponding target weight parameters, using the target weight parameters as regression coefficients to regress the corresponding driving data and obtain the correlation between the vehicle distance to be solved and the driving data of each vehicle. Target road surface data collected by multiple sensors associated with the first vehicle on the target road, along with pre-trained second target weight parameters for each sensor, are acquired. Regression analysis is performed based on the road surface data and the second target weight parameters, using the second target weight parameters as regression coefficients to regress the road surface data from the sensors. This yields the correlation between the vehicle distance to be solved and the data collected by each sensor. By analyzing the correlation between the forward distance to be solved and the driving data of the vehicle in front, as well as the correlation between the forward distance to be solved and the data collected by each sensor, the forward distance between the first vehicle and the vehicle in front can be calculated more accurately. Similarly, by analyzing the correlation between the backward distance to be solved and the driving data of the vehicle in front, as well as the correlation between the backward distance to be solved and the data collected by each sensor, the forward distance between the first vehicle and the vehicle behind can be calculated more accurately.

[0252] In one embodiment, regression analysis is performed based on first driving data, second driving data, and a first target weight parameter to obtain first regression data, including:

[0253] Regression analysis is performed based on the first driving data, forward driving data, and forward target weight parameters to obtain forward regression data; regression analysis is performed based on the first driving data, backward driving data, and backward target weight parameters to obtain backward regression data.

[0254] Based on the first regression data and the second regression data, the vehicle distance between the first vehicle and the second vehicle is determined, including:

[0255] Based on the forward regression data and the second regression data, the forward distance between the first vehicle and the vehicle in front is determined; based on the backward regression data and the second regression data, the backward distance between the first vehicle and the vehicle behind is determined.

[0256] Specifically, the computer equipment can multiply the first driving data and the forward driving data by the forward target weight parameter, and sum the two products to obtain the forward regression data. For each sensor, the road surface data of a single sensor is multiplied by the second target weight parameter of that sensor to obtain the product corresponding to each sensor, and the products are summed to obtain the second regression data. The computer equipment acquires a forward random sequence, and sums the forward regression data, the second regression data, and the forward random sequence to obtain the forward distance between the first vehicle and the vehicle in front. The forward random sequence indicates that the distance between the first vehicle and the vehicle in front follows a standard normal distribution.

[0257] The computer equipment can multiply the first driving data and the backward driving data by the backward target weight parameter, and then sum the two products to obtain the backward regression data. The computer equipment acquires a backward random sequence, sums the backward regression data, the second regression data, and the backward random sequence to obtain the backward distance between the first vehicle and the following vehicle. The backward random sequence indicates that the distance between the first vehicle and the following vehicle follows a standard normal distribution.

[0258] In this embodiment, the forward distance between the first vehicle and the preceding vehicle can be calculated using the forward distance formula:

[0259]

[0260] The backward distance between the first vehicle and the following vehicle can be calculated using the backward distance formula:

[0261]

[0262] Among them, X c,k This represents the first driving data, forward driving data, and backward driving data, or it represents the first feature corresponding to the first driving data, the forward feature corresponding to the forward driving data, and the backward feature corresponding to the backward driving data. When k = 0, it represents the first vehicle, then X c,k This represents the first driving data or first characteristic of the first vehicle; when k=1, it represents a vehicle traveling in the forward direction, then X... c,k X represents forward driving data or forward characteristics; when k = -1, it represents backward vehicles. c,k Represents backward driving data or backward features. R represents the sensor, X represents the sensor. R,i This represents the road surface data or road surface features of the i-th sensor.

[0263] In this embodiment, the forward distance between the first vehicle and the vehicle ahead can be calculated using a forward distance model. The forward distance model uses the forward distance formula to calculate the forward distance from the input data. Similarly, the backward distance between the first vehicle and the vehicle behind can be calculated using a backward distance model. The backward distance model uses the backward distance formula to calculate the backward distance from the input data.

[0264] In this embodiment, regression analysis is performed based on the first driving data, forward driving data, and forward target weight parameters. The forward target weight parameters are then used as regression coefficients to regress the first driving data and forward driving data, yielding forward regression data representing the correlation between the unsolved forward distance and the vehicle's driving data. Regression analysis is then performed based on each road surface data and the second target weight parameters, using the second target weight parameters as regression coefficients to regress the sensor's road surface data. This yields the second regression data representing the correlation between the unsolved vehicle distance and the data collected by each sensor. Based on the correlation between the unsolved forward distance and the driving data of the first vehicle and the vehicles ahead, as well as the correlation with the data collected by each sensor, the forward distance between the first vehicle and the vehicles ahead of it can be calculated more accurately.

[0265] Regression analysis is performed based on the first driving data, the subsequent driving data, and the subsequent target weight parameters. The subsequent target weight parameters are used as regression coefficients to regress the first and subsequent driving data, yielding the correlation between the desired subsequent distance and the vehicle's driving data—this is the subsequent regression data. By analyzing the correlation between the desired subsequent distance and the driving data of the first and subsequent vehicles, as well as its correlation with data collected by various sensors, the subsequent distance between the first vehicle and any vehicles following it can be calculated more accurately.

[0266] In one embodiment, the method further includes providing a prompt when at least one of the forward distance or the backward distance meets the distance prompt condition.

[0267] Specifically, the distance indication condition can be either the forward distance between vehicles being less than a distance threshold, or the backward distance between vehicles being less than a distance threshold.

[0268] When the forward distance between the first vehicle and the vehicle in front meets the distance warning condition, it indicates that the distance between the first vehicle and the vehicle in front is relatively short, and there is a safety hazard of a collision between the first vehicle and the vehicle in front. In this case, a warning is issued through the first vehicle's onboard terminal to effectively prevent a collision. Similarly, when the backward distance between the first vehicle and the vehicle behind meets the distance warning condition, it indicates that the distance between the first vehicle and the vehicle behind is relatively short, and there is a safety hazard of a collision between the first vehicle and the vehicle behind. In this case, a warning is issued through the first vehicle's onboard terminal to effectively prevent a collision.

[0269] like Figure 7 As shown, the vehicle-mounted terminal of the first vehicle collects data from itself to obtain first driving data. Vehicles in front of and behind the first vehicle also collect data, obtaining forward and backward driving data. The vehicle-mounted terminal of the second vehicle collects data from itself to obtain second driving data. Sensors on the road collect data from the road surface to obtain road surface data. When the first vehicle passes sensor 2, sensor 2, its forward sensor 3, and its rearward sensor 1 are used as multiple sensors corresponding to the first vehicle. Each sensor collects data to obtain various road surface data. The forward distance between the first vehicle and the vehicle in front, and the backward distance between the first vehicle and the vehicle behind are calculated according to the processing method of the above embodiment. If the forward or backward distance is less than 1 meter, no warning is issued. If the forward or backward distance is less than 1 meter, a warning is issued through the vehicle-mounted terminal to remind the driver of the first vehicle of a potential collision.

[0270] In one embodiment, a vehicle data processing method is provided, applied to a computer device, comprising:

[0271] Pre-training phase:

[0272] Acquire the first sample driving data of the first sample vehicle, the forward sample driving data of the sample vehicles preceding the first sample vehicle, and the backward sample driving data of the sample vehicles following the first sample vehicle; the driving position of the forward sample vehicle on the sample road is before the driving position of the first sample vehicle, and the driving position of the backward sample vehicle on the sample road is after the driving position of the first sample vehicle.

[0273] Acquire sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road.

[0274] Based on the first sample driving data, feature extraction is performed on multiple first dimensions to obtain the first sample features of the first sample driving data in each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension; feature extraction is performed on the forward sample driving data and the backward sample driving data in each first dimension to obtain the forward sample features of the forward sample driving data in each first dimension, and the backward sample features of the backward sample driving data in each first dimension.

[0275] Based on the features of each first sample and each forward sample, the forward correlation between the first sample vehicle and the forward sample vehicle is determined; based on the features of each first sample and each backward sample, the backward correlation between the first sample vehicle and the backward sample vehicle is determined.

[0276] Based on each sample road surface data, multiple second-dimensional features are extracted to obtain the sample road surface features corresponding to each sample road surface data in each second dimension; the multiple second dimensions include road surface driving dimension and road surface attribute dimension; based on the multiple sample road surface features corresponding to each sample road surface data, the second correlation relationship between multiple sample sensors is determined.

[0277] Obtain the forward initial weight parameters of the forward sample vehicles, the backward initial weight parameters of the backward sample vehicles, and the second initial weight parameters of each sample sensor.

[0278] Based on the forward initial weight parameters and each second initial weight parameter, combined with the features of the first sample, the features of the forward sample, and the road surface features of each sample, the forward initial predicted distance between the first sample vehicle and the forward sample vehicle is determined.

[0279] Based on the backward initial weight parameters and each second initial weight parameter, combined with the features of the first sample, the features of the backward sample, and the road surface features of each sample, the backward initial predicted distance between the first sample vehicle and the backward sample vehicle is determined.

[0280] Based on the forward initial weight parameters and the forward initial prediction distance, combined with the forward expected distance, the first sample features, the forward sample features and the forward correlation, the weight parameters are iteratively processed to obtain the forward target weight parameters corresponding to the forward sample vehicles.

[0281] Based on the backward initial weight parameters and the backward initial prediction distance, combined with the backward expected distance, the first sample features, the backward sample features, and the backward correlation, the weight parameters are iteratively processed to obtain the backward target weight parameters corresponding to the backward sample vehicles.

[0282] The global expected distance is determined based on the forward expected distance and the backward expected distance; the global initial predicted distance is determined based on the forward initial predicted distance and the backward initial predicted distance; based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, the road surface features of each sample, and the second correlation, the weight parameters are iteratively processed to obtain the second target weight parameters corresponding to each sample sensor.

[0283] Application phase:

[0284] Acquire the first feature of the first vehicle, the forward feature of the vehicle in front of the first vehicle, and the backward feature of the vehicle behind the first vehicle; the forward vehicle is positioned before the first vehicle on the target road, and the backward vehicle is positioned after the first vehicle on the target road.

[0285] Acquire target road surface data collected by multiple sensors associated with the first vehicle for the target road. Following the driving sequence, acquire the pre-trained forward target weight parameters for the preceding vehicle, the pre-trained backward target weight parameters for the following vehicle, and the pre-trained second target weight parameters for each sensor.

[0286] Regression analysis is performed based on the first feature, forward feature, and forward target weight parameters to obtain forward regression data; regression analysis is performed based on each road surface feature and the second target weight parameters to obtain second regression data; and the forward distance between the first vehicle and the forward vehicle is determined based on the forward regression data and the second regression data.

[0287] Based on the first feature, the backward feature, and the backward target weight parameters, regression analysis is performed to obtain backward regression data; based on the backward regression data and the second regression data, the backward distance between the first vehicle and the backward vehicle is determined.

[0288] A prompt is given when at least one of the forward distance or backward distance meets the distance prompting condition.

[0289] In this embodiment, during the pre-training phase, first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle are acquired. These first and second sample vehicles have a driving order on the sample road. First sample features of the first sample driving data are extracted in both the vehicle driving dimension and the vehicle attribute dimension. Based on the second sample driving data, multiple first-dimensional feature extractions are performed to obtain second sample features of the second sample driving data in both the vehicle driving dimension and the vehicle attribute dimension. This allows for the accurate construction and determination of a first correlation between the first and second sample vehicles based on their features in both dimensions, thereby accurately representing the feature correlation between the first and second sample vehicles in different dimensions. Based on each sample road surface data, multiple second-dimensional feature extractions are performed to obtain the corresponding features of each sample sensor in both the road driving dimension and the road attribute dimension. This allows for the accurate construction of a second correlation between multiple sample sensors based on their corresponding features in both dimensions, thereby accurately representing the feature correlation between multiple sample sensors in different dimensions.

[0290] Based on the first initial weight parameters of the second sample vehicles and the second initial weight parameters of each sample sensor, and combining the features of the first sample, the features of the second sample, the road surface features of each sample, the first correlation relationship, and the second correlation relationship, cross-iterative weight parameter adjustments are performed. This allows for weight calculation by combining the correlation between the features of each vehicle and the correlation between the road surface features collected by each sensor, resulting in the first target weight parameters of the second sample vehicles and the second target weight parameters of each sample sensor. This avoids the problem of inaccurate weight parameters caused by the repeated superposition and calculation of the same features, effectively improving the accuracy of weight parameter calculation. Using the first target weight parameters and the second target weight parameters to determine the distance between at least two vehicles with a driving sequence can effectively improve the accuracy of vehicle distance prediction.

[0291] In the application phase, regression analysis is performed based on the first feature, forward features, and forward target weight parameters. The forward target weight parameters are then used as regression coefficients to regress the first feature and forward features, yielding the correlation between the desired forward distance and vehicle features—the forward regression feature. Regression analysis is then performed based on various road surface features and the second target weight parameters, using the second target weight parameters as regression coefficients to regress the various road surface features from the sensors. This yields the correlation between the desired vehicle distance and the features collected by each sensor—the second regression feature. By understanding the correlations between the desired forward distance and the features of the first vehicle, the preceding vehicles, and the features collected by each sensor, the forward distance between the first vehicle and the vehicles preceding it can be calculated more accurately.

[0292] Regression analysis is performed based on the first feature, backward features, and backward target weight parameters. The backward target weight parameters are then used as regression coefficients to regress the first feature and backward features, yielding the correlation between the backward distance to be solved and the vehicle's features—the backward regression features. By analyzing the correlation between the backward distance to be solved and the features of the first vehicle and subsequent vehicles, as well as the correlation with the features collected by each sensor, the backward distance between the first vehicle and the vehicles following it can be calculated more accurately.

[0293] When the forward distance between the first vehicle and the vehicle in front meets the distance warning condition, it indicates that the distance between the first vehicle and the vehicle in front is relatively short, and there is a safety hazard of the first vehicle colliding with the vehicle in front. When the backward distance between the first vehicle and the vehicle behind meets the distance warning condition, it indicates that the distance between the first vehicle and the vehicle behind is relatively short, and there is a safety hazard of the first vehicle colliding with the vehicle behind. In this case, the first vehicle's on-board terminal will issue a warning to effectively prevent the first vehicle from colliding with the vehicle behind.

[0294] In one embodiment, a vehicle data processing method is provided for an application scenario applied to the vehicle's in-vehicle infotainment system. This application scenario requires calculating the distances between a first sample vehicle and its preceding and following sample vehicles. The processing flow for this application scenario is as follows: Figure 8 As shown, the model training and prediction stages include steps S802-S818 and steps S820-S822, respectively.

[0295] Model training and prediction refers to using a machine learning or deep learning model, employing gradient descent, to iteratively calculate the model's feature data and labels to obtain model weights, and then using these model weights and prediction samples to make predictions, as detailed below:

[0296] Step S802, Data Input Stage. The data input section includes three parts: first, road information data collected by the vehicle-mounted terminal; second, sample vehicle's own configuration parameter data; and third, road surface data collected by the sample sensors. The road information data collected by the vehicle-mounted terminal and the sample vehicle's own configuration parameter data constitute the sample driving data. That is, the road information data collected by the vehicle-mounted terminal of the first sample vehicle and the vehicle's own configuration parameter data constitute the first sample driving data; the road information data collected by the vehicle-mounted terminal of the forward sample vehicle and the vehicle's own configuration parameter data constitute the forward sample driving data; and the road information data collected by the vehicle-mounted terminal of the backward sample vehicle and the vehicle's own configuration parameter data constitute the backward sample driving data.

[0297] Specifically as follows:

[0298] The road information data collected by the vehicle-mounted system includes: vehicle speed, speeds of vehicles in front and behind, road surface images collected by the vehicle, distances between vehicles in front and behind, distances between vehicles to the left and right, and vehicle POI data.

[0299] Vehicle configuration parameters include: vehicle type, vehicle length, height, width, vehicle capacity, actual number of passengers, maximum speed, and engine displacement.

[0300] The data collected by the roadside sensors include: images of the road surface, road information, average vehicle speed, speed limit data, speeding data, accident images, traffic flow data, waiting time, and POI information for the road section.

[0301] Step S804, the stage of constructing the total distance model and total loss function.

[0302] Total distance model: D=Y+Z+E (1)

[0303] Total loss model:

[0304] Step S806, the stage of constructing the forward distance model and loss function.

[0305] Forward distance formula, i.e., forward distance model:

[0306] Forward loss function model:

[0307] Step S808, backward distance model and loss function construction stage.

[0308] Backward distance formula, i.e., backward distance model:

[0309] Backward loss function model:

[0310] Step S810, the forward and backward distance loss cross model construction stage, that is, the distance loss cross model construction stage for forward and backward vehicles.

[0311] Forward and backward distance loss cross model:

[0312] In the pre-training phase, D represents the total distance between forward and backward sample vehicles, i.e., the global expected distance. Z represents the backward distance between the current sample vehicle and backward sample vehicles, i.e., the backward expected distance. The current sample vehicle is the first sample vehicle. Y represents the forward distance between the current sample vehicle and forward sample vehicles, i.e., the forward expected distance. k represents the k-th vehicle (k=0 represents the current sample vehicle, k=-1 represents the backward sample vehicle of the current sample vehicle, and k=1 represents the forward sample vehicle of the current sample vehicle). X C,k W represents the sample features corresponding to the sample driving data collected by the vehicle's infotainment system of the k-th vehicle, including features of road data and features of the vehicle's own configuration data. C,k V represents the weight parameters corresponding to the vehicle's infotainment system for the forward sample vehicles. C,k This represents the weight parameters corresponding to the vehicle's infotainment system in the backward sample vehicles. R,i X represents the weight parameter corresponding to the i-th sample sensor in the road segment. R,i This represents the sample road surface feature corresponding to the sample road surface data collected by the i-th sample sensor in the road segment. n indicates that there are n sample sensors when the current sample vehicle passes through this road segment. Let represent the weight parameter of the i-th sample sensor (i = -1 represents the backward sample sensor of the current sample sensor, i = 0 represents the current sample sensor, and i = 1 represents the forward sample sensor of the current sample sensor). Let E represent a random number sequence whose total distance follows a standard normal distribution. Y E represents a sequence of random numbers whose forward distances follow a standard normal distribution. Z L represents a sequence of random numbers whose backward distances follow a standard normal distribution. D Represents the total loss function. This represents the distance loss function for the forward sample vehicles. This represents the distance loss function for backward sample vehicles. The crossover loss function α represents the relationship between the forward and backward distances. k Let β represent the parameters of the k-th vehicle. i This represents the parameters of the i-th sample sensor.

[0313] Understandably, in the application phase, Z represents the backward distance between the current vehicle and the following vehicle. Y represents the forward distance between the current vehicle and the preceding vehicle. k represents the k-th vehicle (k=0 represents the current vehicle, k=-1 represents the following vehicle of the current vehicle, k=1 represents the preceding vehicle of the current vehicle, and the current vehicle is the first vehicle). X C,k This represents the features of the driving data collected by the vehicle's infotainment system for the k-th vehicle, including the first feature of the first vehicle, the forward feature of the preceding vehicle, or the backward feature of the following vehicle. W C,k This refers to the weight parameters corresponding to the vehicle's infotainment system for the vehicle in the forward direction, i.e., the target weight parameters. V C,k This refers to the weight parameters corresponding to the vehicle's infotainment system in the rear-facing vehicle, i.e., the target weight parameters. W R,i This represents the weight parameter corresponding to the i-th sensor in the road segment, i.e., the target weight parameter. X R,i This represents the road surface feature corresponding to the road surface data collected by the i-th sensor in the road segment.

[0314] Step S812, Federated Learning Model Construction Stage. Federated learning combines distributed machine learning, cryptography, financial rule-based incentive mechanisms, and game theory to address the problem of using scattered data. Based on the distance model and loss function derived in S804-S210, a federated learning gradient descent parameter estimation model based on the correlation of features between preceding and following vehicles and the correlation of features between adjacent sensors is constructed, as follows:

[0315]

[0316]

[0317]

[0318] In the pre-training phase, W R The weight parameters W corresponding to the sample sensor are represented. C Represents the weight parameters corresponding to the forward sample vehicles, V C This represents the weight parameters corresponding to the vehicles in the backward sample. This represents the total predicted distance between forward sample vehicles and backward sample vehicles. This represents the forward prediction distance between the current sample vehicle and the preceding sample vehicle. This represents the backward prediction distance between the current sample vehicle and the backward sample vehicle.

[0319] Understandably, in the application phase, formula (3) is used to calculate the forward distance between the current vehicle and the vehicle ahead, and formula (5) is used to calculate the backward distance between the current vehicle and the vehicle behind. The current vehicle is the first vehicle. In the application phase, W R This represents the weight parameters W corresponding to the sensor. C This represents the weight parameters corresponding to the forward-facing vehicle, V. C This represents the weight parameters corresponding to the backward vehicles.

[0320] Step S814, adjacent sensor correlation covariance matrix construction stage. Feature extraction is performed on the first sample driving data, the second sample driving data, and the road surface data of each sample, respectively, to obtain the first sample features, the second sample features, and the road surface features of each sample.

[0321] Input the sample road surface features X from each sample sensor R The correlation between the road surface features of each sample sensor is calculated, and the correlation covariance matrix of the road surface features of the sample sensors is obtained:

[0322]

[0323] Step S816: Construction of the correlation covariance matrix between forward and backward sample vehicles. Input the sample features of the forward sample vehicles, the first sample vehicle, and the backward sample vehicles.

[0324] X C,k (k = -1, 0, 1), calculate the forward correlation covariance matrix between the forward sample vehicles and the first sample vehicle:

[0325]

[0326] And the backward correlation covariance matrix between the backward sample vehicles and the first sample vehicles:

[0327]

[0328] Step S818, Model Weight Cross-Iteration Solution Stage. First, initialize the weight parameters W of the forward sample vehicles. C Obtain the forward initial weight parameters W C,0 Initialize the weight parameter W for each sample sensor. R Obtain the second initial weight parameter W R,0 Therefore, according to the forward distance formula Obtain the forward prediction distance The initial value, i.e., the forward initial prediction distance. Reinitialize the weight parameters V of the backward sample vehicles CObtain the backward initial weight parameters V C,0 Therefore, according to the backward distance formula Obtain the backward prediction distance The initial value, i.e., the backward initial prediction distance. This yields the global initial prediction distances for the forward and backward sample vehicles. Then initialize the weight parameters W of each sample sensor. R Obtain the second initial weight parameter W R,0 Substitute the sample road surface features X from the sample sensor R And the correlation matrix Ω, through the formula Calculate the weight parameter W for the first step. R,1 Substitute the weight parameters W of the sample sensor obtained in the first step. R,1 Thus, Y is obtained. R The value of Y R,1 Substitute the feature data X of the sample vehicle C,Y And correlation matrix Σ C,Y Through formula Calculate the weight parameter W for the first step of the forward sample vehicle. C,1 Substitute the weight parameters W of the sample sensor obtained in the first step. R,1 Thus, Z is obtained. R The value of Z R,1 Substitute the feature X of the sample vehicle C,Z And correlation matrix ∑ C,Z Through formula Calculate the weight parameter V for the first step of the backward sample vehicle. C,1 Similarly, through a cross-iteration method, the target weight parameters corresponding to each sample sensor are obtained. Target weight parameters corresponding to forward sample vehicles Target weight parameters corresponding to backward sample vehicles

[0329] Step S820, the vehicle distance prediction stage, i.e., the application stage. Input the target weight parameters obtained from the cross-iteration solution of the vehicle-mounted system in step S818. and Input the road surface features corresponding to the road surface data collected by each sensor, the first feature corresponding to the first driving data of the current vehicle (i.e., the first vehicle), and the forward feature corresponding to the forward driving data of the vehicle in front, and substitute them into the forward distance formula: Obtain the forward distance output by the forward distance model. Input the road features corresponding to the road surface data collected by each sensor, the first feature corresponding to the first driving data of the current vehicle, and the backward features corresponding to the backward driving data of the following vehicle, and substitute them into the backward distance formula. Obtain the backward distance output by the backward distance model.

[0330] Step S822: Safe Distance Judgment and Prompt Stage. Input the forward and backward distances calculated in step S820. Set a distance threshold (e.g., 5 meters). If the forward distance is greater than the threshold, it indicates that the forward distance between the current vehicle and the vehicle in front is safe, and no prompt is needed; otherwise, it indicates that the forward distance between the current vehicle and the vehicle in front is unsafe, and the driver will be prompted via the vehicle's voice system or screen. Similarly, if the backward distance is greater than the threshold, it indicates that the backward distance between the current vehicle and the vehicle behind is safe, and no prompt is needed; otherwise, it indicates that the backward distance between the current vehicle and the vehicle behind is unsafe, and the driver will be prompted via the vehicle's voice system or screen.

[0331] The vehicle data processing method in this embodiment is based on a multi-vehicle, multi-sensor vehicle-road cooperative safe distance prediction scheme. It addresses the correlation problem between data collected by sensors from adjacent vehicles and adjacent roadside sensors, building upon federated learning in vehicle-road cooperation. Specifically, considering the correlation between data collected by adjacent sensors but not between data collected by non-adjacent sensors, a sensor correlation matrix is ​​constructed. Furthermore, the correlation between the driving characteristics of adjacent vehicles is considered, and a front-to-rear vehicle correlation covariance matrix is ​​constructed based on this correlation. This embodiment employs a federated learning algorithm mechanism, effectively protecting data security between vehicles and sensors. Based on longitudinal federated learning between front and rear vehicles, longitudinal federated learning between adjacent sensors, and lateral federated learning between vehicles and sensors, a hybrid federated learning model is constructed, including a total distance model and loss function for front and rear vehicles, a distance model and loss function for the front vehicle, a distance model and loss function for the rear vehicle, and a cross-loss function for front and rear vehicle distances. Furthermore, based on the constructed correlation matrix and various loss functions, the gradient descent method is used to derive the iterative formula for estimating the weight parameters based on federated learning, thereby solving for the optimal weight parameters of the vehicles in front and behind and each sensor. This effectively solves the problem of predicting safe distances between multiple vehicles and multiple sensors with correlations. The method in this embodiment fully considers the impact of the correlation between the vehicles in front and behind and the correlation between adjacent sensors on the safe distance between the vehicles in front and behind, enabling more accurate calculation of the safe distance between the vehicles in front and behind, and effectively preventing collisions between the vehicles in front and behind.

[0332] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0333] Based on the same inventive concept, this application also provides a vehicle data processing apparatus for implementing the vehicle data processing method described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more vehicle data processing apparatus embodiments provided below can be found in the limitations of the vehicle data processing method described above, and will not be repeated here.

[0334] In one embodiment, such as Figure 9 As shown, a vehicle data processing device 900 is provided, comprising:

[0335] The sample driving data acquisition module 902 is used to acquire the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road.

[0336] The sample road surface data acquisition module 904 is used to acquire sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road.

[0337] The correlation determination module 906 is used to determine the first correlation between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data; and to determine the second correlation between multiple sample sensors based on multiple sample road surface data.

[0338] The weight parameter acquisition module 908 is used to acquire the first initial weight parameter of the second sample vehicle and the second initial weight parameter of each sample sensor.

[0339] The weight parameter adjustment module 910 is used to perform cross-iterative weight parameter adjustment based on the first initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship and the second correlation relationship, to obtain the first target weight parameter corresponding to the second sample vehicle and the second target weight parameter corresponding to each sample sensor; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

[0340] In this embodiment, first sample driving data of a first sample vehicle and second sample driving data of a second sample vehicle are acquired. These two vehicles have a driving order on the sample road. Based on the first and second sample driving data, a first correlation relationship is determined between the first and second sample vehicles, thereby determining the correlation between the driving data collected by each vehicle. Sample road surface data collected by multiple sample sensors associated with the first sample vehicle on the sample road are acquired. Based on this sample road surface data, a second correlation relationship is determined between the multiple sample sensors, thereby determining the correlation between the multiple sample sensors on the road surface data.

[0341] Based on the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor, and combining the first sample driving data, the second sample driving data, the road surface data of each sample, the first correlation relationship, and the second correlation relationship, cross-iterative weight parameter adjustments are performed. This allows for weight calculation by combining the correlation between the driving data of each vehicle and the correlation between the road surface data collected by each sensor, resulting in the first target weight parameters of the second sample vehicle and the second target weight parameters of each sample sensor. This avoids the problem of inaccurate weight parameters caused by the repeated superposition and calculation of the same data, effectively improving the accuracy of weight parameter calculation. Using the first target weight parameters and the second target weight parameters to determine the distance between at least two vehicles with a driving sequence can effectively improve the accuracy of vehicle distance prediction.

[0342] In one embodiment, the association determination module 906 is further configured to perform feature extraction on multiple first dimensions based on the first sample driving data to obtain first sample features of the first sample driving data on each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension; perform feature extraction on multiple first dimensions based on the second sample driving data to obtain second sample features of the second sample driving data on each first dimension; and determine the first association relationship between the first sample vehicle and the second sample vehicle based on each first sample feature and each second sample feature.

[0343] In this embodiment, the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension. Feature extraction is performed on the first sample driving data for multiple first dimensions to extract the first sample features of the first sample driving data in the vehicle driving dimension and vehicle attribute dimension, respectively. Feature extraction is then performed on the second sample driving data for multiple first dimensions to obtain the second sample features of the second sample driving data in the vehicle driving dimension and vehicle attribute dimension, respectively. This allows for the accurate construction and determination of the first association relationship between the first sample vehicle and the second sample vehicle based on their features in the vehicle driving dimension and vehicle attribute dimension. Thus, the association relationship accurately characterizes the feature correlation between the first sample vehicle and the second sample vehicle in different dimensions.

[0344] In one embodiment, the association determination module 906 is further configured to determine the sub-association relationship between each first sample feature and each second sample feature; and to determine the first association relationship between the first sample vehicle and the second sample vehicle based on each sub-association relationship.

[0345] In this embodiment, the sub-association relationship between each first sample feature and each second sample feature is determined, so as to characterize the correlation between any two sample features through the sub-association relationship, thereby accurately reflecting the feature correlation between the first sample vehicle and the second sample vehicle on each feature based on the first association relationship determined by each sub-association relationship.

[0346] In one embodiment, the association determination module 906 is further configured to perform feature extraction of multiple second dimensions based on each sample road surface data to obtain sample road surface features corresponding to each sample road surface data in each second dimension; the multiple second dimensions include road surface driving dimension and road surface attribute dimension; and determine the second association relationship between multiple sample sensors based on the multiple sample road surface features corresponding to each sample road surface data.

[0347] In this embodiment, the multiple second dimensions include road surface driving dimension and road surface attribute dimension. Based on each sample road surface data, features of multiple second dimensions are extracted to obtain the features corresponding to each sample sensor in the road surface driving dimension and road surface attribute dimension. This enables the accurate construction of the second association relationship of multiple sample sensors based on the features corresponding to each sample sensor in the road surface driving dimension and road surface attribute dimension, thereby accurately characterizing the feature correlation of multiple sample sensors in different dimensions through the second association relationship.

[0348] In one embodiment, the weight parameter adjustment module 910 is further configured to: determine the initial predicted distance between the first sample vehicle and the second sample vehicle based on the first initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the second sample driving data, and the sample road surface data; obtain the expected distance between the first sample vehicle and the second sample vehicle; perform iterative processing of the weight parameters based on the first initial weight parameter and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship, to obtain the first target weight parameter corresponding to the second sample vehicle; and perform iterative processing of the weight parameters based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, the sample road surface data, and the second correlation relationship, to obtain the second target weight parameter corresponding to each sample sensor.

[0349] In this embodiment, based on the first initial weight parameter and each second initial weight parameter, and combining the first sample driving data, the second sample driving data, and each sample road surface data, the distance between the first sample vehicle and the second sample vehicle is predicted to obtain an initial predicted distance. The expected distance between the first sample vehicle and the second sample vehicle is obtained as a distance label in the iteration. This allows the difference between the actual distance and the distance predicted based on each weight parameter to be determined. Combining the driving data of each vehicle and the data correlation between them, the weight parameters of the second sample vehicle are iterated, thereby continuously optimizing the weight parameters of the second sample vehicle in multiple iterations, thus accurately obtaining the first target weight parameter of the second sample vehicle. Combining the sample road surface data collected by each sample sensor and the data correlation between them, the weight parameters of each sample sensor are iterated, thereby continuously optimizing the weight parameters of each sample sensor in multiple iterations, thus accurately obtaining the second target weight parameter corresponding to each sample sensor.

[0350] In one embodiment, the second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle; the forward sample vehicle is positioned ahead of the first sample vehicle on the sample road, and the backward sample vehicle is positioned behind the first sample vehicle on the sample road; the second sample driving data includes forward sample driving data of the forward sample vehicle and backward sample driving data of the backward sample vehicle; the initial prediction distance includes the forward initial prediction distance and the backward initial prediction distance between the forward sample vehicle and the first sample vehicle, respectively; the expected distance includes the forward expected distance and the backward expected distance between the forward sample vehicle and the first sample vehicle, respectively; the first association relationship includes the forward association relationship and the backward association relationship between the forward sample vehicle and the first sample vehicle, respectively; the first initial weight parameter includes the forward initial weight parameter of the forward sample vehicle and the backward initial weight parameter of the backward sample vehicle; the first target weight parameter includes the forward target weight parameter of the forward sample vehicle and the backward target weight parameter of the backward sample vehicle.

[0351] In this embodiment, based on the forward initial weight parameter, the backward initial weight parameter, and each second initial weight parameter, combined with the first sample driving data, the forward sample driving data, the backward sample driving data, and the road surface data of each sample, the distance between the first sample vehicle and the forward sample vehicle, and between the first sample vehicle and the backward sample vehicle are predicted to obtain the forward predicted distance and the backward predicted distance. By acquiring the forward and backward expected distances as distance labels in the iteration, the difference between the true forward distance and the forward distance predicted based on the weight parameters of the forward sample vehicles and the weight parameters of each sample sensor can be determined. Similarly, the difference between the true backward distance and the backward distance predicted based on the weight parameters of the backward sample vehicles and the weight parameters of each sample sensor can be determined. Combining the driving data of each vehicle and the data correlation between driving data, and the data correlation between sample road surface data collected by each sample sensor and the data correlation between sample road surface data, the weight parameters of the forward sample vehicles, backward sample vehicles, and each sample sensor are iterated. This process continuously optimizes each weight parameter in multiple iterations, thereby accurately obtaining the target weight parameters corresponding to each of the forward sample vehicles, backward sample vehicles, and each sample sensor.

[0352] In one embodiment, the weight parameter adjustment module 910 is further configured to determine the forward initial prediction distance between the first sample vehicle and the forward sample vehicle based on the forward initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the forward sample driving data, and the sample road surface data; and to determine the backward initial prediction distance between the first sample vehicle and the backward sample vehicle based on the backward initial weight parameter and each second initial weight parameter, combined with the first sample driving data, the backward sample driving data, and the sample road surface data.

[0353] In this embodiment, based on the forward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, forward sample driving data, and each sample road surface data, the forward initial predicted distance between the first sample vehicle and the forward sample vehicle can be accurately calculated. Based on the backward initial weight parameters and each second initial weight parameter, combined with the first sample driving data, backward sample driving data, and each sample road surface data, the backward initial predicted distance between the first sample vehicle and the backward sample vehicle can be accurately calculated. Therefore, based on the predicted distance between the first sample vehicle and the forward and backward vehicles, the weight parameters of the forward and backward vehicles and the weight parameters of the sample sensors are adjusted to obtain the optimal solution for each weight parameter.

[0354] In one embodiment, the weight parameter adjustment module 910 is further configured to perform iterative processing of the weight parameters based on the forward initial weight parameters and the forward initial predicted distance, combined with the forward expected distance, the first sample driving data, the forward sample driving data, and the forward correlation relationship, to obtain the forward target weight parameters corresponding to the forward sample vehicle; and to perform iterative processing of the weight parameters based on the backward initial weight parameters and the backward initial predicted distance, combined with the backward expected distance, the first sample driving data, the backward sample driving data, and the backward correlation relationship, to obtain the backward target weight parameters corresponding to the backward sample vehicle.

[0355] In this embodiment, based on the initial forward weight parameters and the initial forward predicted distance, combined with the expected forward distance, the first sample driving data, the forward sample driving data, and the forward correlation, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of the forward sample vehicles are optimized by incorporating various data points, resulting in the optimal solution for the weight parameters corresponding to the forward sample vehicles, i.e., the forward target weight parameters. Similarly, based on the initial backward weight parameters and the initial backward predicted distance, combined with the expected backward distance, the first sample driving data, the backward sample driving data, and the backward correlation, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of the backward sample vehicles are optimized by incorporating various data points, resulting in the optimal solution for the weight parameters corresponding to the backward sample vehicles, i.e., the backward target weight parameters.

[0356] In one embodiment, the weight parameter adjustment module 910 is further configured to perform iterative processing of the weight parameters based on each second initial weight parameter, the forward initial prediction distance and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, the road surface data of each sample and the second correlation relationship, to obtain the second target weight parameter corresponding to each sample sensor.

[0357] In this embodiment, based on each second initial weight parameter, forward initial prediction distance, and backward initial prediction distance, combined with forward expected distance and backward expected distance, each sample road surface data, and the second correlation, the weight parameters are iteratively processed. In this way, the weight parameters of each sample sensor are optimized by combining various data in multiple iterations, so as to obtain the optimal solution of the weight parameters corresponding to each sample sensor, that is, each second target weight parameter.

[0358] In one embodiment, the weight parameter adjustment module 910 is further configured to determine the global expected distance based on the forward expected distance and the backward expected distance; determine the global initial predicted distance based on the forward initial predicted distance and the backward initial predicted distance; and perform iterative processing of the weight parameters based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, the road surface data of each sample, and the second correlation relationship, to obtain the second target weight parameter corresponding to each sample sensor.

[0359] In this embodiment, the global initial predicted distance between the first sample vehicle and the vehicles in front and behind is calculated based on the forward and backward expected distances. The global expected distance between the first sample vehicle and the vehicles in front and behind is then calculated based on the forward and backward initial predicted distances, thus simultaneously considering the impact of the vehicles in front and behind on the first sample vehicle. Based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, road surface data for each sample, and the second correlation relationship, iterative processing of the weight parameters is performed. Through multiple iterations, the weight parameters of each sample sensor are optimized using data from the first sample vehicle itself, the vehicles in front and behind, the correlation between vehicles, the correlation between sensors, and the vehicle data collected by each sample sensor. This yields the optimal solution for the weight parameters corresponding to each sample sensor, making the obtained second target weight parameters suitable for predicting the distance between the first vehicle and the vehicles in front and behind when the first vehicle simultaneously has vehicles in front and behind.

[0360] In one embodiment, such as Figure 10 As shown, a vehicle data processing device 1000 is provided, comprising:

[0361] The driving data acquisition module 1002 is used to acquire the first driving data of the first vehicle and the second driving data of the second vehicle; the first vehicle and the second vehicle have a driving sequence on the target road.

[0362] The road surface data acquisition module 1004 is used to acquire target road surface data collected by multiple sensors related to the first vehicle for the target road.

[0363] The target weight acquisition module 1006 is used to acquire the first target weight parameters obtained by pre-training for the second vehicle and the second target weight parameters obtained by pre-training for each sensor, according to the driving sequence.

[0364] The processing module 1008 is used to perform regression analysis based on the first driving data, the second driving data and the first target weight parameter to obtain the first regression data; and to perform regression analysis based on the road surface data and the second target weight parameter to obtain the second regression data.

[0365] The distance determination module 1010 is used to determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

[0366] In this embodiment, second driving data for the first and second vehicles, which have a driving sequence on the target road, are acquired, along with pre-trained first target weight parameters for the second vehicle. Regression analysis is performed based on the first driving data, second driving data, and first target weight parameters, using the first target weight parameters as regression coefficients to regress the first and second driving data, thus obtaining the correlation between the vehicle distance to be solved and the driving data of each vehicle—this is the first regression data. Target road surface data collected by multiple sensors associated with the first vehicle on the target road are acquired, along with pre-trained second target weight parameters for each sensor. Regression analysis is performed based on the road surface data and second target weight parameters, using the second target weight parameters as regression coefficients to regress the road surface data of each sensor, thus obtaining the correlation between the vehicle distance to be solved and the data collected by each sensor—this is the second regression data. Based on the correlation between the vehicle distance to be solved and the driving data of each vehicle, and the correlation with the data collected by each sensor, the vehicle distance between the first and second vehicles can be calculated more accurately.

[0367] In one embodiment, the second vehicle includes the vehicle traveling in front of the first vehicle and the vehicle traveling behind the first vehicle. The vehicle traveling in front is positioned ahead of the first vehicle on the target road, and the vehicle traveling behind the first vehicle is positioned behind the first vehicle on the target road. The second driving data includes the forward driving data of the vehicle traveling in front and the backward driving data of the vehicle traveling behind. The first target weight parameter includes the forward target weight parameter of the vehicle traveling in front and the backward target weight parameter of the vehicle traveling behind. The first regression data includes the forward regression data of the vehicle traveling in front and the backward regression data of the vehicle traveling behind. The vehicle distance includes the forward distance between the first vehicle and the vehicle traveling in front, and the backward distance between the first vehicle and the vehicle traveling behind.

[0368] In this embodiment, the driving data of a first vehicle, its preceding vehicle, and its following vehicle, which have a driving sequence on the target road, are acquired. Pre-trained target weight parameters for the preceding and following vehicles are also obtained. Regression analysis is performed based on the driving data and target weight parameters of each vehicle, using the target weight parameters as regression coefficients to regress the corresponding driving data and obtain the correlation between the vehicle distance to be solved and the driving data of each vehicle. Target road surface data collected by multiple sensors associated with the first vehicle on the target road are acquired, along with pre-trained second target weight parameters for each sensor. Regression analysis is performed based on the road surface data and the second target weight parameters, using the second target weight parameters as regression coefficients to regress the road surface data of each sensor and obtain the correlation between the vehicle distance to be solved and the data collected by each sensor. Based on the correlation between the forward distance to be solved and the driving data of the preceding vehicle, and the correlation between the forward distance and the data collected by each sensor, the forward distance between the first vehicle and the preceding vehicle can be calculated more accurately. Based on the correlation between the backward distance to be solved and the driving data of the forward vehicle, as well as the correlation between the data collected by each sensor, the forward distance between the first vehicle and the backward vehicle can be calculated more accurately.

[0369] In one embodiment, the processing module 1008 is further configured to perform regression analysis processing based on the first driving data, forward driving data, and forward target weight parameters to obtain forward regression data; and to perform regression analysis processing based on the first driving data, backward driving data, and backward target weight parameters to obtain backward regression data.

[0370] The distance determination module 1010 is also used to determine the forward distance between the first vehicle and the vehicle in front based on the forward regression data and the second regression data; and to determine the backward distance between the first vehicle and the vehicle behind based on the backward regression data and the second regression data.

[0371] In this embodiment, regression analysis is performed based on the first driving data, forward driving data, and forward target weight parameters. The forward target weight parameters are then used as regression coefficients to regress the first driving data and forward driving data, yielding forward regression data representing the correlation between the unsolved forward distance and the vehicle's driving data. Regression analysis is then performed based on each road surface data and the second target weight parameters, using the second target weight parameters as regression coefficients to regress the sensor's road surface data. This yields the second regression data representing the correlation between the unsolved vehicle distance and the data collected by each sensor. Based on the correlation between the unsolved forward distance and the driving data of the first vehicle and the vehicles ahead, as well as the correlation with the data collected by each sensor, the forward distance between the first vehicle and the vehicles ahead of it can be calculated more accurately.

[0372] Regression analysis is performed based on the first driving data, the subsequent driving data, and the subsequent target weight parameters. The subsequent target weight parameters are used as regression coefficients to regress the first and subsequent driving data, yielding the correlation between the desired subsequent distance and the vehicle's driving data—this is the subsequent regression data. By analyzing the correlation between the desired subsequent distance and the driving data of the first and subsequent vehicles, as well as its correlation with data collected by various sensors, the subsequent distance between the first vehicle and any vehicles following it can be calculated more accurately.

[0373] In one embodiment, the device further includes a prompting module; the prompting module is used to provide a prompt when the vehicle distance meets the distance prompting condition.

[0374] In one embodiment, the device further includes a prompting module; the prompting module is used to provide a prompt when at least one of the forward distance or backward distance meets the distance prompting condition.

[0375] Each module in the aforementioned vehicle data processing device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0376] In one embodiment, a computer device is provided, which may be a terminal or a server. Taking a terminal as an example, its internal structure diagram can be as follows: Figure 11As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a vehicle data processing method. The display unit of the computer device is used to form a visually visible image. It can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0377] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0378] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0379] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0380] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0381] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data shall comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0382] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0383] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0384] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A vehicle data processing method, characterized in that, The method includes: Obtain the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road; Acquire sample road surface data collected by multiple sample sensors associated with the first sample vehicle for the sample road; Based on the first sample driving data and the second sample driving data, a first correlation relationship is determined between the first sample vehicle and the second sample vehicle; based on the multiple sample road surface data, a second correlation relationship is determined between the multiple sample sensors. Obtain the first initial weight parameters of the second sample vehicle and the second initial weight parameters of each sample sensor; The first initial weight parameter is weighted and summed with the first sample driving data and the second sample driving data respectively to obtain the first sample regression data; The sample road surface data of each sample sensor and the corresponding second initial weight parameter are weighted and summed to obtain the second sample regression data; The regression data of the first sample and the regression data of the second sample are summed to obtain the initial predicted distance between the first sample vehicle and the second sample vehicle; Obtain the expected distance between the first sample vehicle and the second sample vehicle. Based on the first initial weight parameter and the initial predicted distance, and combined with the expected distance, the first sample driving data, the second sample driving data and the first correlation relationship, perform iterative processing of the weight parameter to obtain the first target weight parameter corresponding to the second sample vehicle. Based on each of the second initial weight parameters and the initial predicted distance, combined with the expected distance, each of the sample road surface data and the second correlation, the weight parameters are iteratively processed to obtain the second target weight parameter corresponding to each sample sensor; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

2. The method according to claim 1, characterized in that, The step of determining the first association relationship between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data includes: Based on the first sample driving data, features of multiple first dimensions are extracted to obtain the first sample features of the first sample driving data in each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension. Based on the second sample driving data, feature extraction is performed on the multiple first dimensions to obtain the second sample features of the second sample driving data in each first dimension; Based on the first sample features and the second sample features, a first association relationship is determined between the first sample vehicle and the second sample vehicle.

3. The method according to claim 2, characterized in that, The step of determining the first association relationship between the first sample vehicle and the second sample vehicle based on each of the first sample features and each of the second sample features includes: Determine the sub-association relationship between each feature of the first sample and each feature of the second sample; Based on the sub-associations, a first association relationship is determined between the first sample vehicle and the second sample vehicle.

4. The method according to claim 1, characterized in that, Determining the second correlation between the multiple sample sensors based on the multiple sample road surface data includes: Based on each of the sample road surface data, multiple second-dimensional features are extracted to obtain the sample road surface features corresponding to each of the sample road surface data in each second dimension; the multiple second dimensions include road surface driving dimension and road surface attribute dimension. Based on the multiple sample road features corresponding to each of the sample road surface data, a second correlation relationship is determined among the multiple sample sensors.

5. The method according to any one of claims 1 to 4, characterized in that, The second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle; the forward sample vehicle is positioned before the first sample vehicle on the sample road, and the backward sample vehicle is positioned after the first sample vehicle on the sample road. The second sample driving data includes the forward sample driving data of the forward sample vehicle and the backward sample driving data of the backward sample vehicle; the initial predicted distance includes the forward initial predicted distance and the backward initial predicted distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the expected distance includes the forward expected distance and the backward expected distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively. The first association relationship includes the forward association relationship and the backward association relationship between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the first initial weight parameter includes the forward initial weight parameter of the forward sample vehicle and the backward initial weight parameter of the backward sample vehicle; The first target weight parameter includes the forward target weight parameter of the forward sample vehicle and the backward target weight parameter of the backward sample vehicle.

6. The method according to claim 5, characterized in that, The step of weighting and summing the first initial weight parameter with the first sample driving data and the second sample driving data respectively to obtain the first sample regression data includes: The first sample driving data and the forward sample driving data are respectively weighted and summed with the forward initial weight parameter to obtain the forward sample regression data; The first sample driving data, the backward sample driving data, and the backward initial weight parameter are weighted and summed to obtain the backward sample regression data. The step of summing the regression data of the first sample and the regression data of the second sample to obtain the initial predicted distance between the first sample vehicle and the second sample vehicle includes: The forward sample regression data, the second sample regression data, and the forward random sequence are summed to obtain the forward initial predicted distance between the first sample vehicle and the forward sample vehicle; The backward sample regression data, the second sample regression data, and the backward random sequence are summed to obtain the backward initial predicted distance between the first sample vehicle and the backward sample vehicle.

7. The method according to claim 5, characterized in that, The step involves iteratively processing the weight parameters based on the first initial weight parameters and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship, to obtain the first target weight parameters corresponding to the second sample vehicle, including: Based on the forward initial weight parameters and the forward initial predicted distance, combined with the forward expected distance, the first sample driving data, the forward sample driving data and the forward correlation, the weight parameters are iteratively processed to obtain the forward target weight parameters corresponding to the forward sample vehicle. Based on the backward initial weight parameters and the backward initial prediction distance, combined with the backward expected distance, the first sample driving data, the backward sample driving data and the backward correlation, the weight parameters are iteratively processed to obtain the backward target weight parameters corresponding to the backward sample vehicle.

8. The method according to claim 5, characterized in that, The step involves iteratively processing the weight parameters based on each of the second initial weight parameters and the initial predicted distance, combined with the expected distance, each of the sample road surface data, and the second correlation relationship, to obtain the second target weight parameters corresponding to each sample sensor, including: Based on each of the second initial weight parameters, the forward initial prediction distance and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, each of the sample road surface data and the second correlation relationship, the weight parameters are iteratively processed to obtain the second target weight parameters corresponding to each of the sample sensors.

9. The method according to claim 8, characterized in that, The step involves iteratively processing the weight parameters based on each of the second initial weight parameters, the forward initial prediction distance, and the backward initial prediction distance, combined with the forward expected distance, the backward expected distance, each of the sample road surface data, and the second correlation relationship, to obtain the second target weight parameters corresponding to each sample sensor, including: The global expected distance is determined based on the forward expected distance and the backward expected distance; The global initial prediction distance is determined based on the forward initial prediction distance and the backward initial prediction distance; Based on each of the second initial weight parameters and the global initial prediction distance, combined with the global expected distance, each of the sample road surface data and the second correlation, the weight parameters are iteratively processed to obtain the second target weight parameters corresponding to each of the sample sensors.

10. A vehicle data processing method, characterized in that, The method includes: Acquire first driving data of the first vehicle and second driving data of the second vehicle; the first vehicle and the second vehicle have a driving order on the target road; Acquire target road surface data collected by multiple sensors associated with the first vehicle for the target road; According to the driving sequence, the first target weight parameter obtained from pre-training for the second vehicle and the second target weight parameter obtained from pre-training for each of the sensors are obtained. The first target weight is obtained by iterative processing of the weight parameters based on the first initial weight parameter and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship. The initial predicted distance is obtained by summing the first sample regression data and the second sample regression data. The first sample regression data is obtained by weighted summing the first initial weight parameter with the first sample driving data and the second sample driving data, respectively. The second sample regression data is obtained by weighted summing the sample road surface data of each sample sensor with the corresponding second initial weight parameter. The second target weight parameter is obtained by iterative processing of the weight parameters based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, each of the sample road surface data, and the second correlation relationship. Regression analysis is performed based on the first driving data, the second driving data, and the first target weight parameter to obtain the first regression data; Regression analysis is performed based on the road surface data and each of the second target weight parameters to obtain the second regression data. Based on the first regression data and the second regression data, the vehicle distance between the first vehicle and the second vehicle is determined.

11. A vehicle data processing device, characterized in that, The device includes: The sample driving data acquisition module is used to acquire the first sample driving data of the first sample vehicle and the second sample driving data of the second sample vehicle; the first sample vehicle and the second sample vehicle have a driving order on the sample road; The sample road surface data acquisition module is used to acquire sample road surface data collected by multiple sample sensors related to the first sample vehicle for the sample road. The correlation determination module is used to determine a first correlation between the first sample vehicle and the second sample vehicle based on the first sample driving data and the second sample driving data; and to determine a second correlation between the multiple sample sensors based on the multiple sample road surface data. The weight parameter acquisition module is used to acquire the first initial weight parameter of the second sample vehicle and the second initial weight parameter of each sample sensor; The weight parameter adjustment module is used to perform weighted summation of the first initial weight parameter with the first sample driving data and the second sample driving data to obtain first sample regression data; to perform weighted summation of the sample road surface data of each sample sensor with the corresponding second initial weight parameter to obtain second sample regression data; to sum the first sample regression data and the second sample regression data to obtain the initial predicted distance between the first sample vehicle and the second sample vehicle; to obtain the expected distance between the first sample vehicle and the second sample vehicle; and to perform iterative processing of the weight parameters based on the first initial weight parameter and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship, to obtain the first target weight parameter corresponding to the second sample vehicle; and to perform iterative processing of the weight parameters based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, each sample road surface data, and the second correlation relationship, to obtain the second target weight parameter corresponding to each sample sensor; the first target weight parameter and the second target weight parameter are used to determine the distance between at least two vehicles with a driving sequence.

12. The apparatus according to claim 11, characterized in that, The association determination module is further configured to perform feature extraction on multiple first dimensions based on the first sample driving data to obtain first sample features of the first sample driving data on each first dimension; the multiple first dimensions include vehicle driving dimension and vehicle attribute dimension; perform feature extraction on the multiple first dimensions based on the second sample driving data to obtain second sample features of the second sample driving data on each first dimension; and determine a first association between the first sample vehicle and the second sample vehicle based on each first sample feature and each second sample feature.

13. The apparatus according to claim 12, characterized in that, The association determination module is further configured to determine the sub-association relationship between each of the first sample features and each of the second sample features; and to determine the first association relationship between the first sample vehicle and the second sample vehicle based on each of the sub-association relationships.

14. The apparatus according to claim 11, characterized in that, The association determination module is also used to perform feature extraction of multiple second dimensions based on each of the sample road surface data, so as to obtain the sample road surface features corresponding to each of the sample road surface data in each second dimension. The multiple second dimensions include road surface driving dimension and road surface attribute dimension; based on the multiple sample road surface features corresponding to each sample road surface data, the second correlation relationship between the multiple sample sensors is determined.

15. The apparatus according to any one of claims 11 to 14, characterized in that, The second sample vehicle includes a forward sample vehicle and a backward sample vehicle of the first sample vehicle; the forward sample vehicle is positioned before the first sample vehicle on the sample road, and the backward sample vehicle is positioned after the first sample vehicle on the sample road. The second sample driving data includes the forward sample driving data of the forward sample vehicle and the backward sample driving data of the backward sample vehicle; the initial predicted distance includes the forward initial predicted distance and the backward initial predicted distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the expected distance includes the forward expected distance and the backward expected distance between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively. The first association relationship includes the forward association relationship and the backward association relationship between the forward sample vehicle and the backward sample vehicle and the first sample vehicle, respectively; the first initial weight parameter includes the forward initial weight parameter of the forward sample vehicle and the backward initial weight parameter of the backward sample vehicle; The first target weight parameter includes the forward target weight parameter of the forward sample vehicle and the backward target weight parameter of the backward sample vehicle.

16. The apparatus according to claim 15, characterized in that, The weight parameter adjustment module is further configured to perform a weighted summation of the first sample driving data and the forward sample driving data with the forward initial weight parameter to obtain forward sample regression data. The first sample driving data, the backward sample driving data, and the backward initial weight parameter are weighted and summed to obtain the backward sample regression data. The forward sample regression data, the second sample regression data, and the forward random sequence are summed to obtain the forward initial predicted distance between the first sample vehicle and the forward sample vehicle; The backward sample regression data, the second sample regression data, and the backward random sequence are summed to obtain the backward initial predicted distance between the first sample vehicle and the backward sample vehicle.

17. The apparatus according to claim 15, characterized in that, The weight parameter adjustment module is further configured to perform iterative processing of the weight parameters based on the forward initial weight parameters and the forward initial predicted distance, combined with the forward expected distance, the first sample driving data, the forward sample driving data, and the forward correlation, to obtain the forward target weight parameters corresponding to the forward sample vehicle; and to perform iterative processing of the weight parameters based on the backward initial weight parameters and the backward initial predicted distance, combined with the backward expected distance, the first sample driving data, the backward sample driving data, and the backward correlation, to obtain the backward target weight parameters corresponding to the backward sample vehicle.

18. The apparatus according to claim 15, characterized in that, The weight parameter adjustment module is further configured to perform iterative processing of the weight parameters based on each of the second initial weight parameters, the forward initial prediction distance and the backward initial prediction distance, combined with the forward expected distance and the backward expected distance, each of the sample road surface data and the second correlation relationship, to obtain the second target weight parameter corresponding to each of the sample sensors.

19. The apparatus according to claim 18, characterized in that, The weight parameter adjustment module is further configured to determine the global expected distance based on the forward expected distance and the backward expected distance; determine the global initial predicted distance based on the forward initial predicted distance and the backward initial predicted distance; and perform iterative processing of the weight parameters based on each second initial weight parameter and the global initial predicted distance, combined with the global expected distance, each of the sample road surface data and the second correlation relationship, to obtain the second target weight parameter corresponding to each sample sensor.

20. A vehicle data processing device, characterized in that, The device includes: The driving data acquisition module is used to acquire first driving data of the first vehicle and second driving data of the second vehicle; the first vehicle and the second vehicle have a driving sequence on the target road; The road surface data acquisition module is used to acquire target road surface data collected by multiple sensors related to the first vehicle for the target road. The target weight acquisition module is used to acquire, according to the driving order, the pre-trained first target weight parameters corresponding to the second vehicle and the pre-trained second target weight parameters corresponding to each of the sensors. The first target weight is obtained by iterative processing of weight parameters based on the first initial weight parameters and the initial predicted distance, combined with the expected distance, the first sample driving data, the second sample driving data, and the first correlation relationship. The initial predicted distance is obtained by summing the first sample regression data and the second sample regression data. The first sample regression data is obtained by weighted summing the first initial weight parameters with the first sample driving data and the second sample driving data, respectively. The second sample regression data is obtained by weighted summing the sample road surface data of each sample sensor with the corresponding second initial weight parameters. The second target weight parameters are obtained by iterative processing of weight parameters based on each second initial weight parameter and the initial predicted distance, combined with the expected distance, each of the sample road surface data, and the second correlation relationship. The processing module is used to perform regression analysis processing based on the first driving data, the second driving data, and the first target weight parameter to obtain first regression data; and to perform regression analysis processing based on each of the road surface data and each of the second target weight parameters to obtain second regression data. The distance determination module is used to determine the vehicle distance between the first vehicle and the second vehicle based on the first regression data and the second regression data.

21. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 10.

22. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

23. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 10.

Images