Vehicle splashing monitoring method, device, equipment and medium

Abstract

The invention discloses a vehicle splashing monitoring method which comprises the steps that splashing prediction influence information is acquired, and the splashing prediction influence information comprises at least one of vehicle body information, pedestrian information and environment information; predicting the splash prediction influence information by using a splash prediction model to obtain a prediction result containing a splash range; wherein the splash prediction model is obtained by training an initial neural network model by taking splash prediction influence information as input and a splash range as output; according to the method, the splashing range is predicted through the vehicle body information, the pedestrian information and the environment information, the risk that accumulated water splashes to pedestrians when the vehicle passes through the water-accumulated road surface is reduced, and the pedestrian splashing prevention effect is better.

Description

Technical Field

[0001] The present application relates to the field of testing technology, and in particular to a vehicle water splash monitoring method, device, equipment and medium. Background Art

[0002] When driving on the road, if you encounter puddles, especially those with large areas and deep water, there will often be a significant splashing effect. When the vehicle passes through these puddles quickly, the wheels will suddenly lift up the accumulated water, forming a strong water flow and countless tiny water droplets. These splashing water droplets may draw arcs in the air and then fall on the windshield, side windows and rearview mirrors of the vehicle, seriously interfering with the driver's vision.

[0003] What is more serious is that the splashing water may also affect pedestrians and cyclists on the non-motorized vehicle lanes nearby. Pedestrians may get their clothes wet or even slip and get injured due to the sudden splashing water; cyclists may lose balance and have accidents due to obstructed vision or wheel slippage.

[0004] Application Contents

[0005] In view of the above-mentioned shortcomings of the prior art, the present application provides a vehicle water splash monitoring method, device, equipment and medium to solve at least one defect in the prior art.

[0006] To achieve the above objectives and other objectives, the present application provides a vehicle water splash monitoring method, comprising:

[0007] Acquiring water splash prediction impact information, wherein the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information;

[0008] Using a water splash prediction model to predict the water splash prediction impact information, and obtaining a prediction result including a water splash range;

[0009] The water splash prediction model is obtained by training an initial neural network model with the water splash prediction impact information as input and the water splash range as output.

[0010] In one embodiment of the present application, the water splash prediction model includes an input layer, a feature extraction layer, a feature fusion layer, and a prediction layer; the prediction step of the water splash prediction model includes:

[0011] Receiving the water splash prediction impact information through an input layer;

[0012] Using a feature extraction layer to extract features from the water splash prediction impact information to obtain water splash prediction impact features, the water splash prediction impact features including: at least one of a vehicle body feature corresponding to the vehicle body information, a pedestrian feature corresponding to the pedestrian information, and a road feature corresponding to the environmental information;

[0013] The splash prediction impact features are fused through a feature fusion layer to obtain fused features;

[0014] The fused features are mapped into a prediction result of the water splash range based on the prediction layer.

[0015] In one embodiment of the present application, the vehicle body information includes at least one of the following: real-time vehicle speed, vehicle driving direction, vehicle load, and vehicle body height;

[0016] The pedestrian information includes at least one of the following: the relative distance between the pedestrian and the vehicle;

[0017] The environmental information includes at least one of the following: real-time wind speed, real-time wind direction, and water depth in the waterlogged area.

[0018] In one embodiment of the present application, the input layer includes multiple input sub-layers, each input sub-layer corresponds to receiving a type of splash prediction impact feature, and different input sub-layers receive different types of splash prediction impact features.

[0019] In one embodiment of the present application, the step of fusing the water splash prediction impact features through the feature fusion layer to obtain fused features includes:

[0020] Splicing the splash prediction impact features to obtain splicing features;

[0021] The attention mechanism is used to adjust the weight relationship between the splash prediction influencing features to obtain fused features.

[0022] In one embodiment of the present application, the prediction result of the water splash range includes at least one of the following: splash height, splash angle, splash range, and splash direction.

[0023] In one embodiment of the present application, the vehicle water splash monitoring method further includes:

[0024] generating a vehicle control instruction according to the prediction result of the water splashing range, so as to control the vehicle driving by the vehicle control instruction;

[0025] The vehicle control instruction includes: a deceleration instruction or an obstacle avoidance instruction. The deceleration instruction is used to control the vehicle to decelerate and the obstacle avoidance instruction is used to control the vehicle to avoid a flooded area.

[0026] In order to achieve the above-mentioned purpose and other related purposes, the present application provides a vehicle water splash monitoring device, the vehicle water splash monitoring device comprising:

[0027] An information acquisition module, used to acquire water splash prediction impact information, wherein the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information;

[0028] The splash prediction module is used to use the splash prediction model to predict the splash prediction influence information to obtain a prediction result including the splash range; wherein the splash prediction model is obtained by training the initial neural network model with the splash prediction influence information as input and the splash range as output.

[0029] To achieve the above objectives and other related objectives, the present application provides a vehicle water splash monitoring device, including:

[0030] one or more processors; and

[0031] The memory is used to store one or more programs. When the one or more programs are executed by the one or more processors, the memory implements the method described.

[0032] To achieve the above objectives and other related objectives, the present application provides one or more machine-readable media having instructions stored thereon, which, when executed by one or more processors, enable the processors to execute the described methods.

[0033] Beneficial effects of this application:

[0034] A vehicle splash monitoring method of the present application includes: obtaining splash prediction impact information, the splash prediction impact information including: at least one of vehicle body information, pedestrian information, and environmental information; using a splash prediction model to predict the splash prediction impact information to obtain a prediction result including a splash range; wherein the splash prediction model is obtained by training an initial neural network model with the splash prediction impact information as input and the splash range as output; the present application predicts the splash range by using one of vehicle body information, pedestrian information, and environmental information, thereby reducing the risk of pedestrians being splashed by accumulated water when the vehicle passes through a flooded road surface, and the pedestrian anti-splash effect is better.

[0035] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0037] Figure 1 Schematic diagram of the implementation environment of the vehicle water splash monitoring method according to an embodiment of the present application

[0038] Figure 2 A flow chart of a vehicle water splash monitoring method according to an embodiment of the present application;

[0039] Figure 3 A flow chart of a method for predicting a water splash range according to an embodiment of the present application;

[0040] Figure 4 A schematic diagram of a vehicle water splash monitoring device according to an embodiment of the present application;

[0041] Figure 5 A schematic diagram of the structure of a computer system suitable for implementing the memory of an embodiment of the present application is shown. DETAILED DESCRIPTION

[0042] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0043] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0044] Although the terms "first", "second", "A", and "B", etc. may be used herein to describe various elements, these elements should not be limited by these terms and are only used to distinguish one element from another. For example, a first element may be referred to as a second element, and similarly, a second element may be referred to as a first element without departing from the scope of the following technology. The term "and / or" includes a combination of multiple related items or any of the multiple related items.

[0045] As used herein, unless the context indicates otherwise, the singular form is intended to include the plural form, and it will be understood that the term "comprising" means the presence of stated features, quantities, steps, operations, elements, or combinations thereof, but does not exclude the presence or addition of one or more other features, quantities, steps, operations, elements, components, or combinations thereof.

[0046] Before the detailed description, it is intended to clarify that the division of components in this specification is divided only by the main function of each component. That is, two or more components to be described below can be combined into one component, or can be divided into two or more components according to more detailed functions. In addition to the main function of the component, each component to be described below can also perform some or all of the functions of other components, and some of the main functions of each component can be exclusively performed by other components.

[0047] Figure 1 FIG. 1 is a schematic diagram of an implementation environment of a vehicle water splash monitoring method according to an embodiment of the present application. Figure 1 The implementation environment includes an information collection terminal 110, a service terminal 120, and a terminal device 130. The information collection terminal 110 and the service terminal 120 are used to provide water splash prediction impact information, which includes: at least one of vehicle body information, pedestrian information, and environmental information; the vehicle body information includes at least one of the following: real-time vehicle speed, vehicle driving direction, vehicle load, and vehicle body height; pedestrian information includes at least one of the following: the relative distance between pedestrians and vehicles; environmental information includes at least one of the following: real-time wind speed, real-time wind direction, and water depth in the water accumulation area; wherein tire information and vehicle body height can be obtained from the service terminal. The data processing end 140 receives the splash prediction impact information and inputs the splash prediction impact information into a pre-trained splash prediction model, predicts the splash range through the splash prediction model, and obtains a prediction result of the splash range, and the prediction result of the splash range includes at least one of the following: splash height, splash angle, splash range, and splash direction; the splash prediction model is obtained by training the initial neural network model with the splash prediction impact information as input and the splash range as output; the terminal device 130 can also be used to display the prediction result of the splash range, so that the vehicle driver can intuitively see whether the surrounding pedestrians will be accidentally splashed, so as to take timely measures to protect the pedestrians. The present application predicts the splash range through vehicle body information, pedestrian information, and environmental information, which reduces the risk of pedestrians being splashed by accumulated water when the vehicle passes through a flooded road surface, and the pedestrian splash protection effect is better.

[0048] It should be understood that Figure 1 The number of information collection terminals 110 and service terminals 120 is only for illustration purposes. Any number of information collection terminals 110 and service terminals 120 may be provided according to actual needs.

[0049] Among them, the terminal device can be any electronic device with a user input interface, including but not limited to smart phones, tablets, laptops, computers, car computers, etc., among which the user input interface includes but is not limited to touch screens, keyboards, physical buttons, audio pickup devices, etc.

[0050] Among them, the server can be a server that provides various services. It can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network) and big data and artificial intelligence platforms. This section does not impose any restrictions on this.

[0051] See also Figure 2 , Figure 2 This is a flow chart of a vehicle splash monitoring method according to an embodiment of the present application. Figure 2 As shown, a vehicle splash monitoring method at least includes steps S210 to S220:

[0052] Step S210, obtaining water splash prediction impact information, where the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information;

[0053] In the embodiment of the present application, the vehicle body information includes at least one of the following: real-time vehicle speed, vehicle travel direction, vehicle load, and vehicle height; pedestrian information includes at least one of the following: relative distance between pedestrians and vehicles; environmental information includes at least one of the following: real-time wind speed, real-time wind direction, and water depth in the waterlogged area. Of course, in other embodiments, the vehicle body information may also include tire information.

[0054] It should be noted that the real-time vehicle speed, vehicle direction, vehicle load, relative distance between pedestrians and vehicles, real-time wind speed, real-time wind direction, and water depth in the flooded area can be collected through the information collection terminal; and the vehicle height and tire information can be called from the server 120.

[0055] The information acquisition module includes vehicle speed sensor, wind speed and direction sensor, inertial sensor, load sensor and lidar.

[0056] The vehicle speed sensor works by indirectly calculating the vehicle speed by measuring the rotation speed of the wheels, and converting this information into an electrical signal to transmit to a receiver. In this embodiment, the real-time vehicle speed is collected by the vehicle speed sensor.

[0057] Wind speed and direction sensor: The wind force generated by air flow drives the sensor to rotate, and the central axis drives the internal sensing element to generate a pulse signal. The wind speed and pulse frequency are linearly related, so the wind speed is calculated, and the wind direction is measured through a wind vane or a magnetic sensor. This embodiment monitors the real-time wind speed and real-time wind direction through the wind speed and direction sensor.

[0058] Inertial sensor, the working principle of inertial wind speed sensor is based on Newton's first law, that is, an object will remain stationary or move in a uniform linear motion without external force. This sensor indirectly measures physical quantities such as acceleration, angular velocity and direction of the object by measuring the inertial force generated by the object during movement. This embodiment monitors the direction of the vehicle through an inertial sensor.

[0059] Load sensor, the working principle of load sensor is based on the principle of strain measurement. When an object is acted upon by a force, its shape and size will change slightly. This deformation is called strain, and the strain is proportional to the magnitude and direction of the force on the object. The load sensor measures the weight or carrying capacity by measuring the strain caused by the force on the object. In this embodiment, the load sensor is used to measure the vehicle load.

[0060] Laser radar, the working principle of laser radar is mainly based on emitting a laser beam to a target and detecting the target's position, speed and other characteristic quantities by receiving the signal reflected from the target. In this embodiment, the depth of water in the waterlogged area and the relative distance between pedestrians and vehicles are measured by laser radar.

[0061] Step S220, using a splash prediction model to predict the splash prediction influence information to obtain a prediction result including a splash range; wherein the splash prediction model is obtained by training an initial neural network model with the splash prediction influence information as input and the splash range as output.

[0062] In the embodiment of the present application, by collecting data related to the splash range, and then using an intelligent algorithm to predict the splash range of the vehicle, the owner can take measures such as speed reduction and obstacle avoidance in advance to avoid accidentally injuring pedestrians. The prediction result of the splash range uses an intelligent algorithm, which is a neural network algorithm. The core of the neural network algorithm is to simulate the structure and function of the human brain neural network, process and learn the complex relationships in a large amount of data, and then predict unknown data. Before using the neural network model to predict the splash range, it is necessary to first train the neural network-based splash prediction model, which specifically includes the following steps:

[0063] 1) Dataset collection and preprocessing

[0064] Collect data collected by the cloud and the information collection end, namely, vehicle speed, direction of travel of pedestrians and vehicles, vehicle load, vehicle height, tire information, relative distance between pedestrians and vehicles, wind speed, wind direction, water depth in the waterlogged area, etc., and then pre-process the data. The pre-processing steps include:

[0065] ① Noise filtering: Remove noise from sensor data caused by environmental factors, equipment failure or transmission errors to improve data accuracy.

[0066] ②Data synchronization: Ensure that data from different sensors are synchronized in time so that they can be correctly associated and fused in subsequent processing.

[0067] ③Data standardization / normalization: Unify the dimensions and ranges of different sensor data to the same scale to avoid certain features dominating the neural network training process.

[0068] ④ Feature extraction: Extract features useful for predicting the splash range from the original data, reduce the data dimension and improve the training efficiency of the model.

[0069] ⑤Data segmentation: Split the dataset into training set, validation set, and test set to facilitate cross-validation and performance evaluation during model training.

[0070] ⑥Data enhancement: Improve the generalization ability of the model by increasing the diversity of the data set.

[0071] 2) Training and Validation

[0072] ① Model training: According to the designed neural network structure, the model parameters are initialized, including the initial value setting of weights and biases, as well as the initial configuration of hyperparameters. The training process includes forward propagation, loss calculation, back propagation, and iterative training.

[0073] The water splash prediction model includes an input layer, a feature extraction layer, a feature fusion layer, and a prediction layer; the input layer is used to receive the water splash prediction impact information; the feature extraction layer is used to extract the features of the water splash prediction impact information to obtain the water splash prediction impact features, and the water splash prediction impact features include: vehicle body features corresponding to the vehicle body information, pedestrian features corresponding to the pedestrian information, and road features corresponding to the environmental information; the feature fusion layer is used to fuse the vehicle body features, pedestrian features, and road features to obtain fusion features; the prediction layer maps the fusion features to the prediction results of the water splash range. The following is a detailed description of each layer of the water splash prediction model.

[0074] Input layer: Since the model needs to process data from different sensors, the input layer should be designed to receive multiple types of data. This can be achieved through a parallel input layer, that is, the input layer includes multiple input sublayers, each of which corresponds to a type of sensor data. According to the characteristics of the sensor data, the input layer needs to adapt to different data formats. For example, lidar data needs to be input in the form of a 3D point cloud or a 2D image, while time series data (such as vehicle speed, wind speed and direction) can be directly input in the form of a numerical sequence.

[0075] Feature extraction layer: For data that provides spatial information, such as lidar, a convolutional neural network (CNN) can be used for feature extraction. CNN can automatically learn the spatial hierarchy from the raw data and extract spatial features that are useful for predicting the range of water splashes. For time series data (such as vehicle speed, wind speed and direction, and inertial sensor data), a recurrent neural network (RNN) can be used for feature extraction. These networks can capture the time dependencies in the data and extract dynamic features in the time series. For data that provides a single value, such as load sensors, feature extraction can be performed directly through the fully connected layer (FCN). FCN can learn the nonlinear relationships in the data and convert them into higher-level feature representations.

[0076] Feature fusion layer: Features from different feature extraction layers are concatenated to form a feature vector containing all sensor information. The attention mechanism is then introduced to dynamically adjust the importance of different features. The attention mechanism can automatically learn the weight relationship between different features according to the needs of the current task, thereby improving the prediction performance of the model.

[0077] After using the feature extraction layer to extract the splash prediction influence features, the vehicle body feature a, pedestrian feature b, and road feature c are obtained. Then the vehicle body feature a, pedestrian feature b, and road feature c are spliced ​​to obtain the fusion feature, fusion feature = (a, b, c). The importance of different features is determined according to the current task. The importance of features is different for different tasks. The greater the importance, the greater the weight. On the contrary, the smaller the importance, the smaller the weight. The attention weight of the vehicle body feature is w1, the attention weight of the vehicle body feature is w2, and the attention weight of the road feature is w3. The fusion feature can be expressed as: fusion feature = (a*w1, b*w2, c*w3).

[0078] Prediction layer: After the feature fusion layer, FCN is used as the prediction layer. The prediction layer maps the fused feature vector to the prediction result of the splash range. The prediction result can be specific indicators such as the bounding box of the splash range and the splash height. According to the specific requirements of the prediction task, the output layer is designed as a regression task (predicting continuous values).

[0079] The prediction layer FCN (fully convolutional neural network) maps the fusion features of the vehicle body features, pedestrian features, and road features into a prediction result of the splash range, including at least one of the following: splash height, splash angle, splash range, and splash direction.

[0080] ② Model optimization: Select a suitable loss function to evaluate the predictive performance of the model, and use efficient optimization algorithms to train the model. These optimization algorithms can automatically adjust the learning rate and accelerate the convergence process of the model.

[0081] In this embodiment, since the prediction of the water splash range is a regression task, the loss function may be one of a mean square error loss function and an absolute error loss function.

[0082] ③Model evaluation and tuning: Use cross-validation to evaluate the performance of the model. By dividing the training set and validation set multiple times, the generalization ability of the model can be more accurately evaluated. Then, the hyperparameters of the model are tuned through grid search, Bayesian optimization and other methods to further improve the performance of the model. The hyperparameters include learning rate, batch size, number of network layers, number of neurons, etc. In order to prevent the model from overfitting, regularization technology can be used to limit the complexity of the model. If the performance of a single model cannot meet the requirements, you can consider using model integration technology to improve the prediction performance.

[0083] ④ Model validation: Divide the data set into training set, validation set and test set. The validation set is used to evaluate the performance of the model during the training process, so as to select and tune the model. Carry out performance evaluation, use quantitative indicators to evaluate the performance of the model on the validation set, and then qualitatively evaluate the prediction effect of the model by comparing the visual prediction results with the true labeled values. Select the optimal model based on the performance evaluation results on the validation set.

[0084] In one embodiment, the performance evaluation of the water splash prediction model can be verified by one or more of accuracy, precision, recall, F1 score, ROC curve and AUC, confusion matrix, mean square error, mean absolute error, cross-validation score, etc. The specific verification process is not repeated here.

[0085] ⑤ Model testing and deployment: Evaluate the performance of the final selected model on an independent test set to ensure that the model has good generalization ability. Deploy the trained water splash prediction model to the actual vehicle and integrate it with the sensor system to achieve the function of real-time prediction of the vehicle's water splash range.

[0086] 3) Deployment and practical application

[0087] Deploy the trained model to the actual vehicle, integrate it with the sensor system, collect sensor data in real time while the vehicle is driving, and use the model to predict the splash range. Adjust the vehicle behavior based on the prediction results, such as slowing down and avoiding water accumulation areas, and collect actual feedback to further optimize the model.

[0088] 4) Continuous Optimization

[0089] New data is continuously collected and used to update and improve the model, and the model structure and parameters are adjusted according to the performance in actual applications.

[0090] See also Figure 3 , Figure 3This is a flow chart of a method for predicting a splash range according to an embodiment of the present application. In the embodiment of the present application, the splash prediction model includes an input layer, a feature extraction layer, a feature fusion layer, and a prediction layer; the splash prediction impact information is input into the pre-trained splash prediction model, and the steps of predicting the splash range by the splash prediction model to obtain the prediction result of the splash range include:

[0091] Step S310, receiving water splash prediction impact information through the input layer;

[0092] Since the splash prediction model needs to process data from different sensors, the input layer should receive multiple types of data. This can be achieved through parallel input layers, each of which corresponds to one type of sensor data. The input layer includes multiple input sublayers, each of which receives one type of splash prediction impact feature, and different input sublayers receive different types of splash prediction impact features. According to the characteristics of the sensor data, the input layer needs to adapt to different data formats. For example, lidar data needs to be input in the form of a three-dimensional point cloud or a two-dimensional image, while time series data (such as vehicle speed, wind speed and direction) can be directly input in the form of a numerical sequence.

[0093] Step S320, using the feature extraction layer to extract features of the water splash prediction influence information to obtain water splash prediction influence features, where the water splash prediction influence features include: at least one of: vehicle body features corresponding to the vehicle body information, pedestrian features corresponding to the pedestrian information, and road features corresponding to the environmental information;

[0094] The feature extraction layer uses different feature extraction networks to extract features from the splash prediction impact information received by the input layer according to the type of data.

[0095] For data that provides spatial information, such as lidar, a convolutional neural network (CNN) can be used for feature extraction. CNN can automatically learn the spatial hierarchy from the raw data and extract spatial features that are useful for predicting the range of water splashes. For time series data (such as vehicle speed, wind speed and direction, and inertial sensor data), a recurrent neural network (RNN) can be used for feature extraction. Recurrent neural networks can capture the time dependency in the data and extract dynamic features in the time series. For data that provides a single value, such as load sensors, feature extraction can be performed directly through a fully connected layer (fully convolutional neural network, FCN). FCN can learn nonlinear relationships in the data and convert them into higher-level feature representations.

[0096] Step S330, fusing the splash prediction impact features through a feature fusion layer to obtain fused features;

[0097] Specifically, the vehicle body features, pedestrian features, and road features are spliced ​​to obtain spliced ​​features; the attention mechanism is used to adjust the weight relationship between the vehicle body features, pedestrian features, and road features to obtain fused features.

[0098] The features from different feature extraction layers are concatenated to form a feature vector containing all sensor information. The attention mechanism is then introduced to dynamically adjust the importance of different features. The attention mechanism can automatically learn the weight relationship between different features according to the needs of the current task, thereby improving the prediction performance of the model.

[0099] Step S340: Use the prediction layer to map the fused features into a prediction result of the splash range.

[0100] After the feature fusion layer, FCN is used as the prediction layer. The prediction layer maps the fused feature vector into the prediction result of the splash range. The prediction result can be one or more specific indicators such as splash height, splash angle, splash range, splash direction, etc.

[0101] In an embodiment of the present application, the vehicle water splash monitoring method further includes:

[0102] A vehicle control instruction is generated based on the prediction result of the water splashing range to control the vehicle's driving through the vehicle control instruction; the vehicle control instruction includes: a deceleration instruction or an obstacle avoidance instruction, the deceleration instruction is used to control the vehicle to decelerate, and the obstacle avoidance instruction is used to control the vehicle to avoid the water accumulation area.

[0103] In summary, a vehicle splash monitoring method of the present application obtains splash prediction impact information, and the splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information; the splash prediction impact information is predicted using a splash prediction model to obtain a prediction result including the splash range; in this way, in rainy days or muddy road conditions, it can detect in real time whether there are water splashes around the wheels, helping the driver to predict the road conditions ahead and avoid driving hazards caused by sudden puddles or slippery areas; through early warning, the driver can take measures to slow down or avoid, reducing the possible damage to the vehicle or dirtying of pedestrians' clothes caused by wheel splashing, reflecting the dual care for vehicles and pedestrians; further, in severe weather conditions, the driver does not need to be distracted to pay attention to whether the wheels are splashing water, and can focus more on driving itself and enjoy a more comfortable driving experience, which also makes the driving process more convenient, reduces the inconvenience caused by human judgment errors, and improves driving comfort and convenience.

[0104] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0105] Figure 4 is a block diagram of a vehicle water splash monitoring device shown in an embodiment of the present application. The device can be applied to Figure 1 The device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applied.

[0106] like Figure 4 As shown, a vehicle splash monitoring device comprises:

[0107] An information acquisition module 410 is used to acquire water splash prediction impact information, where the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information;

[0108] A splash prediction module 420, configured to predict the splash prediction impact information using a splash prediction model to obtain a prediction result including a splash range;

[0109] The water splash prediction model is obtained by training an initial neural network model with the water splash prediction impact information as input and the water splash range as output.

[0110] In an embodiment of the present application, the water splash prediction module 420 receives water splash prediction influence information through an input layer; uses a feature extraction layer to perform feature extraction on the water splash prediction influence information to obtain water splash prediction influence features, and the water splash prediction influence features include: vehicle body features corresponding to vehicle body information, pedestrian features corresponding to pedestrian information, and at least one of road features corresponding to environmental information; fuses the water splash prediction influence features through a feature fusion layer to obtain a fused feature; and uses a prediction layer to map the fused features into a prediction result of the water splash range.

[0111] In an embodiment of the present application, the vehicle body information includes at least one of the following: real-time vehicle speed, vehicle driving direction, vehicle load, and vehicle body height; pedestrian information includes at least one of the following: the relative distance between the pedestrian and the vehicle; environmental information includes at least one of the following: real-time wind speed, real-time wind direction, and water depth in the flooded area.

[0112] In an embodiment of the present application, the input layer includes multiple input sub-layers, each input sub-layer corresponds to receiving a type of splash prediction impact feature, and different input sub-layers receive different types of splash prediction impact features.

[0113] In an embodiment of the present application, the feature fusion layer splices the splash prediction influencing features to obtain spliced ​​features; and uses the attention mechanism to adjust the weight relationship between the splash prediction influencing features to obtain fused features.

[0114] In an embodiment of the present application, the prediction result of the water splashing range includes at least one of the following: splashing height, splashing angle, splashing range, and splashing direction.

[0115] In an embodiment of the present application, after obtaining the predicted result of the splashing range, a vehicle control instruction is generated according to the predicted result of the splashing range to control the vehicle's driving through the vehicle control instruction; the vehicle control instruction includes: a deceleration instruction or an obstacle avoidance instruction, the deceleration instruction is used to control the vehicle to decelerate, and the obstacle avoidance instruction is used to control the vehicle to avoid the water accumulation area.

[0116] It should be noted that the vehicle splash monitoring device provided in the above embodiment and the vehicle splash monitoring method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiment and will not be repeated here. In actual application, the vehicle splash monitoring device provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0117] An embodiment of the present application also provides a device, including: one or more processors; and a memory for storing one or more programs, when the one or more programs are executed by one or more processors, the memory implements a vehicle water splash monitoring method in the above embodiment.

[0118] Embodiments of the present application further provide one or more machine-readable media having instructions stored thereon, which, when executed by one or more processors, enable the processors to execute a vehicle water splash monitoring method in the above-mentioned embodiment.

[0119] Figure 5 The schematic diagram of the structure of a computer system suitable for implementing the memory of the embodiment of the present application is shown. It should be noted that: Figure 5 The computer system of the memory shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0120] like Figure 5As shown, the computer system includes a central processing unit (CPU), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) or the program loaded from the storage part to the random access memory (RAM), such as the method in the above embodiment. In the RAM, various programs and data required for system operation are also stored. The CPU, ROM and RAM are connected to each other through a bus. The input / output (I / O) interface is also connected to the bus.

[0121] The following components are connected to the I / O interface: an input part including a keyboard, a mouse, etc.; an output part including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker; a storage part including a hard disk, etc.; and a communication part including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication part performs communication processing via a network such as the Internet. A drive is also connected to the I / O interface as needed. Removable media, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., are installed on the drive as needed so that a computer program read therefrom is installed into the storage part as needed.

[0122] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing a vehicle splash monitoring method implemented as described above. In such an embodiment, the computer program can be downloaded and installed from a network through a communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), various functions defined in the system of the present application are executed.

[0123] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. This propagated data signal can take a variety of forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. A computer program contained on a computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0124] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0125] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0126] Another aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor of a computer, causes the computer to execute a vehicle splash monitoring method as described above. The computer-readable storage medium may be included in the memory described in the above embodiment, or may exist independently without being assembled into the memory.

[0127] Another aspect of the present application further provides a computer program product or a computer program, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes a vehicle splash monitoring method provided in each of the above embodiments.

[0128] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A vehicle splash monitoring method, characterized in that: The vehicle splash monitoring method comprises: Acquiring water splash prediction impact information, wherein the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information; Using a water splash prediction model to predict the water splash prediction impact information, and obtaining a prediction result including a water splash range; The water splash prediction model is obtained by training an initial neural network model with the water splash prediction impact information as input and the water splash range as output.

2. The vehicle water splash monitoring method according to claim 1, characterized in that: The water splash prediction model includes an input layer, a feature extraction layer, a feature fusion layer, and a prediction layer; the prediction steps of the water splash prediction model include: Receiving the water splash prediction impact information through an input layer; Using a feature extraction layer to extract features from the water splash prediction impact information to obtain water splash prediction impact features, the water splash prediction impact features including: at least one of a vehicle body feature corresponding to the vehicle body information, a pedestrian feature corresponding to the pedestrian information, and a road feature corresponding to the environmental information; The splash prediction impact features are fused through a feature fusion layer to obtain fused features; The fused features are mapped into a prediction result of the water splash range based on the prediction layer.

3. The vehicle water splash monitoring method according to claim 2, characterized in that: The vehicle body information includes at least one of the following: real-time vehicle speed, vehicle driving direction, vehicle load, and vehicle body height; The pedestrian information includes at least one of the following: the relative distance between the pedestrian and the vehicle; The environmental information includes at least one of the following: real-time wind speed, real-time wind direction, and water depth in the waterlogged area.

4. The vehicle water splash monitoring method according to claim 2, characterized in that: The input layer includes a plurality of input sub-layers, each of which receives a type of splash prediction impact feature, and different input sub-layers receive different types of splash prediction impact features.

5. The vehicle water splash monitoring method according to claim 2, characterized in that: The step of fusing the splash prediction impact features through the feature fusion layer to obtain fusion features includes: Splicing the splash prediction impact features to obtain splicing features; The attention mechanism is used to adjust the weight relationship between the splash prediction influencing features to obtain fused features.

6. The vehicle water splash monitoring method according to claim 1 or 2, characterized in that: The prediction result of the splash range includes at least one of the following: Splash height, splash angle, splash range, splash direction.

7. The vehicle water splash monitoring method according to claim 1, characterized in that: The vehicle splash monitoring method further comprises: generating a vehicle control instruction according to the prediction result of the water splashing range, so as to control the vehicle driving by the vehicle control instruction; The vehicle control instruction includes: a deceleration instruction or an obstacle avoidance instruction. The deceleration instruction is used to control the vehicle to decelerate and the obstacle avoidance instruction is used to control the vehicle to avoid a flooded area.

8. A vehicle splash monitoring device, characterized in that: The vehicle water splash monitoring device comprises: An information acquisition module, used to acquire water splash prediction impact information, wherein the water splash prediction impact information includes: at least one of vehicle body information, pedestrian information, and environmental information; A water splash prediction module, used to predict the water splash prediction influence information using a water splash prediction model to obtain a prediction result including a water splash range; The water splash prediction model is obtained by training an initial neural network model with the water splash prediction impact information as input and the water splash range as output.

9. A vehicle splash monitoring device, characterized in that: include: one or more processors; and A memory, used to store one or more programs, when the one or more programs are executed by the one or more processors, the memory implements the method according to any one of claims 1 to 7.

10. A machine-readable medium, characterized in that Instructions are stored thereon, which, when executed by one or more processors, cause the processors to perform the method according to any one of claims 1 to 7.

Images