Vehicle control method and vehicle

Abstract

The invention relates to the technical field of vehicles, in particular to a vehicle control method and a vehicle. The purpose is to improve the vehicle cleaning effect in a complex vehicle environment. The vehicle control method comprises the steps of obtaining multi-modal environment data of a vehicle, determining a weight corresponding to each modal environment data in the multi-modal environment data, and determining a vehicle environment with the highest priority in at least one vehicle environment where the vehicle is located based on the weights and the multi-modal environment data. And windshield wiper equipment of the vehicle is controlled based on the vehicle environment with the highest priority. Through information complementation among all modal environment data, the stability and accuracy of vehicle environment recognition are improved, more complex vehicle environment recognition requirements are met, then cleaning of windshield wiper equipment is more suitable for cleaning requirements in the complex vehicle environment, and the cleaning effect on a vehicle is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a vehicle control method and a vehicle. BACKGROUND

[0002] The vehicle needs to control the windshield cleaning of the vehicle by the windshield wiper device according to different vehicle environments in the rain, snow, sandstorm and other vehicle environments.

[0003] In a related technology, sensor data is obtained in response to a vehicle mode, the sensor data includes forward-looking sensor data and light rain amount sensor data, data processing is performed based on the rain amount sensor data to obtain a discrimination type, and vehicle control is performed based on the discrimination type in combination with a preset execution mechanism, the execution mechanism including a vehicle body headlamp and a windshield wiper. Specifically, when the discrimination type is a regular type, the execution mechanism is directly controlled to perform vehicle control; when the discrimination type is a feature type, the execution mechanism is controlled based on the forward-looking sensor data. In the above technical solution, the obtained sensor data is single, and is easily affected by environmental interference to affect the recognition accuracy of the discrimination type. At the same time, the execution mechanism is directly controlled based on single sensor data, which cannot adapt to the cleaning demand in a complex scene, and the cleaning effect is poor.

[0004] In another related technology, data of a driving environment of a vehicle is collected by a preset collection device to obtain environmental data, a scene analysis is performed on the vehicle based on the environmental data to determine a driving scene of the vehicle, and an initial windshield wiper motion mode is switched according to the driving scene to control the windshield wiper to clean the vehicle. The environmental data in the above technical solution and the scene obtained based on the environmental data are also single, and the windshield wiper motion mode is relatively fixed, so the cleaning demand in a complex scene cannot be adapted, and the cleaning effect is poor. SUMMARY

[0005] The present application provides a vehicle control method and a vehicle, which improves the cleaning effect of the windshield wiper device on the vehicle in a complex vehicle environment.

[0006] To achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows:

[0007] In a first aspect, the present application provides a vehicle control method, the vehicle control method comprising:

[0008] obtaining multi-modal environmental data of the vehicle, determining a weight corresponding to each modal environmental data in the multi-modal environmental data, determining a vehicle environment with the highest priority in at least one vehicle environment in which the vehicle is located based on the weight and the multi-modal environmental data, and controlling a windshield wiper device of the vehicle based on the vehicle environment with the highest priority.

[0009] The technical scheme provided in the application determines the vehicle environment with the highest priority among at least one vehicle environment in which the vehicle is located based on multi-modal environment data and a weight corresponding to each modal environment data, comprehensively and accurately reflects information about the vehicle environment in which the vehicle is located, reduces errors caused by environmental interference of a single data source by complementing semantic information between each modal environment data, improves the stability and accuracy of vehicle environment identification by balancing the contribution degree of different data sources to vehicle environment identification, and can adapt to more complex vehicle environment identification requirements. In addition, the vehicle wiper device is flexibly controlled to clean the vehicle environment with the highest priority in a complex vehicle environment by determining the vehicle environment with the highest priority, which helps to improve the accuracy of the control strategy of the wiper device in a complex vehicle environment, and the cleaning of the wiper device is more suitable for the cleaning requirements of the complex vehicle environment, thereby improving the cleaning effect of the vehicle.

[0010] In a possible implementation, the vehicle environment with the highest priority among at least one vehicle environment in which the vehicle is located is determined based on the weight and the multi-modal environment data, and can be specifically implemented as follows: the multi-modal environment data is fused based on the weight to obtain target environment data, at least one vehicle environment is determined based on the target environment data, and the vehicle environment with the highest priority is determined from the at least one vehicle environment based on the multi-modal environment data. The weighted fusion of the multi-modal environment data based on the weight can flexibly associate the environment data with the highest correlation degree for different application scenarios, adaptively adjust the contribution of each modal environment data to vehicle environment identification, reduce the weight of the interference modal, reduce the negative impact on the fusion result, and thus improve the accuracy of the vehicle environment identification result, while coordinating the data distribution difference between different modal environment data and unifying the quantitative scale.

[0011] In a possible implementation, the weight corresponding to each modal environment data in the multi-modal environment data is determined, and can be specifically implemented as follows: the confidence of each modal environment data in the multi-modal environment data is obtained, and the weight corresponding to each modal environment data is determined based on the confidence. The weight is determined based on the confidence, the weight of the environment data with higher reliability is improved, and thus the reliability of the data fusion result is improved.

[0012] In a possible implementation, the weight corresponding to each modal environment data is determined based on the confidence, and can be specifically implemented as follows: an initial weight corresponding to each modal environment data is determined, and the weight is determined based on the initial weight and the confidence. The weight is determined based on the initial weight and the confidence, and the weight determination efficiency is improved.

[0013] In a possible implementation, the at least one vehicle environment is determined based on the target environment data, which can be specifically implemented as follows: the at least one vehicle environment is determined based on the target environment data, running information of the vehicle, and a scene classification model. The at least one vehicle environment is determined in combination with the environment data, the running state of the vehicle itself, and the scene classification model, which can more comprehensively reflect the vehicle surrounding environment and the interaction between the vehicle and the environment, and the vehicle control is more in line with the actual cleaning demand. Meanwhile, the scene classification model can improve the efficiency of vehicle environment determination and be updated in an incremental learning manner, flexibly adapt to different data distribution and classification requirements, and improve the recognition accuracy of the vehicle environment.

[0014] In a possible implementation, the wiper device of the vehicle is controlled based on the vehicle environment with the highest priority, which can be specifically implemented as follows: a control strategy corresponding to the vehicle environment with the highest priority is queried, and the wiper device of the vehicle is controlled based on the control strategy. In the case of a complex vehicle environment, the control strategy of the vehicle environment with the highest priority is preferentially ensured, and the conflict between control strategies in different vehicle environments is avoided.

[0015] In a possible implementation, the wiper device of the vehicle is controlled based on the control strategy, which can be specifically implemented as follows: the change of the vehicle environment in a future time period is predicted based on the multi-modal environment data. The control strategy is adjusted based on the change, and the wiper device of the vehicle is controlled by using the adjusted control strategy. The control demand in the future time period is predicted in advance, the adjustment time of the control strategy is reduced, and the adaptability of the control strategy to different vehicle environment changes is improved.

[0016] In a possible implementation, the vehicle environment includes: a congestion environment, a night rainy environment, a sandstorm environment, a haze environment, a low-temperature icing environment, a tunnel environment, an automatic car washing environment, a front vehicle splashing water environment, and a rain and fog environment. A plurality of complex vehicle environments are covered, so that the wiper device adapts to the cleaning demand in the complex vehicle environment.

[0017] In a possible implementation, the control strategy is used to indicate a control scheme of a wiper and / or a water spraying component in the wiper device. The control scheme of the wiper includes a cleaning speed, a cleaning mode, a cleaning pressure, and a heating wire temperature of the wiper. The control scheme of the water spraying component includes a water spraying amount, a water spraying time, and a water spraying pressure of the water spraying component. The control strategy is more flexible and diverse, different control strategies can be flexibly selected for different vehicle environments, the cleaning problem in different vehicle environments is solved in a targeted manner, and thus the cleaning effect is improved.

[0018] In a second aspect, the present application provides a vehicle control device, which includes an acquisition module, a processing module, and an execution module.

[0019] The acquisition module is configured to acquire multi-modal environment data of the vehicle.

[0020] The processing module is configured to determine a weight corresponding to each of the multi-modal environment data, and determine a vehicle environment with a highest priority from at least one vehicle environment of the vehicle based on the weight and the multi-modal environment data.

[0021] The execution module is configured to control a wiper device of the vehicle based on the vehicle environment with the highest priority.

[0022] In a possible implementation, the processing module is further configured to: fuse the multi-modal environment data based on the weight to obtain target environment data, determine the at least one vehicle environment based on the target environment data, and determine the vehicle environment with the highest priority from the at least one vehicle environment based on the multi-modal environment data.

[0023] In a possible implementation, the processing module is further configured to: obtain a confidence degree of each of the multi-modal environment data, and determine the weight corresponding to each of the multi-modal environment data based on the confidence degree.

[0024] In a possible implementation, the processing module is further configured to: determine an initial weight corresponding to each of the multi-modal environment data, and determine the weight based on the initial weight and the confidence degree.

[0025] In a possible implementation, the processing module is further configured to: determine the at least one vehicle environment based on the target environment data, running information of the vehicle, and a scene classification model.

[0026] In a possible implementation, the execution module is further configured to: query a control strategy corresponding to the vehicle environment with the highest priority, and control the wiper device of the vehicle based on the control strategy.

[0027] In a possible implementation, the execution module is further configured to: predict a change of the vehicle environment in a future period of time based on the multi-modal environment data, adjust the control strategy based on the change, and control the wiper device of the vehicle by using the adjusted control strategy.

[0028] In a possible implementation, the vehicle environment includes a congestion environment, a night rainy environment, a sandstorm environment, a haze environment, a low-temperature icing environment, a tunnel environment, an automatic car washing environment, a front vehicle splashing water environment, and a rain and fog environment.

[0029] In a possible implementation, the control strategy is used to indicate a control scheme of a wiper and / or a water spraying component in the wiper device, the control scheme of the wiper includes a cleaning speed, a cleaning mode, a cleaning pressure, and a heating wire temperature of the wiper, and the control scheme of the water spraying component includes a water spraying amount, a water spraying time, and a water spraying pressure of the water spraying component.

[0030] The technical effects of any one of the implementation manners of the second aspect can refer to the technical effects of any one of the implementation manners of the first aspect, which will not be described herein.

[0031] In a third aspect, the present application provides a vehicle, which is cleaned based on the vehicle control method provided in any one of the first aspect.

[0032] In a fourth aspect, the present application provides an electronic device, which comprises a processor and a memory, and the memory stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the vehicle control method of the first aspect.

[0033] In a fifth aspect, a computer readable storage medium is provided, and the computer readable storage medium stores at least one computer program, and the at least one computer program is loaded and executed by the processor to implement the vehicle control method of the first aspect.

[0034] In a sixth aspect, a computer program product is provided, and the computer program product comprises a computer program or instructions, and when the computer program or instructions are executed by the processor, the vehicle control method of the first aspect is implemented.

[0035] The schemes provided in the fourth aspect to the sixth aspect are used to implement the vehicle control method provided in the second aspect, and the specific implementation will not be described one by one. The technical effects of any one of the implementation manners of the fourth aspect to the sixth aspect can refer to the technical effects of any one of the implementation manners of the second aspect, which will not be described herein.

[0036] It should be noted that the various possible implementation manners of any one of the aspects can be combined on the premise that the schemes are not contradictory. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 A structural schematic diagram of a vehicle control system provided for an exemplary embodiment;

[0038] Figure 2 A flowchart of a vehicle control method provided for an exemplary embodiment;

[0039] Figure 3 A flowchart of a risk score classification method provided for an exemplary embodiment;

[0040] Figure 4 A flowchart of a control method for a sandstorm environment provided for an exemplary embodiment;

[0041] Figure 5 A flowchart of a control strategy triggering method provided for an exemplary embodiment;

[0042] Figure 6 A flowchart of another vehicle control method provided for the exemplary embodiments;

[0043] Figure 7 A structural diagram of a vehicle control device provided for the exemplary embodiments;

[0044] Figure 8 A structural diagram of an electronic device provided for the exemplary embodiments. DETAILED DESCRIPTION

[0045] In the embodiments of the present application, in order to clearly describe the technical solutions of the embodiments of the present application, the same items or similar items with basically the same functions and effects are distinguished by using "first", "second", and the like. Those skilled in the art can understand that "first", "second", and the like do not limit the quantity and execution order, and "first", "second", and the like do not necessarily mean different. The technical features described by "first" and "second" have no order or size order.

[0046] In the embodiments of the present application, the words "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design presented as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or advantageous than other embodiments or design solutions. Rather, the use of the words "exemplary" or "for example" is intended to present related concepts in a specific way, which is convenient for understanding.

[0047] In the embodiments of the present application, at least one can also be described as one or more, and the plurality can be two, three, four, or more, which is not limited by the present application.

[0048] In order to facilitate understanding, the vehicle control method provided by the present application is specifically introduced below in combination with the drawings.

[0049] The scheme provided by the present application can be applied to Figure 1 In the computer system shown, for example Figure 1As shown, the computer system provided by the embodiments of the present application includes a data acquisition device 100, a controller 101 and a windshield wiper device 102. Optionally, the data acquisition device 100 can be a sensor, a radar, a satellite, a drone, a camera, a navigation application (APP) and the like for acquiring multi-modal environmental data. The controller 101 is a device for acquiring multi-modal environmental data of the data acquisition device 100, performing data fusion on the multi-modal environmental data, and performing vehicle environment recognition based on the fused data. The controller 101 can also be a device for weighted fusion of multi-modal data. The controller 101 can also be a device for controlling the windshield wiper device 102 based on the data after data fusion. Optionally, the controller 101 includes a body control module (BCM), an automatic driving domain controller and the like. The windshield wiper device 102 is a device for cleaning the vehicle. Exemplarily, the windshield wiper device 102 includes a windshield wiper blade, a windshield wiper arm, a windshield wiper motor, a connecting rod mechanism and a windshield washer reservoir and the like. In the embodiments of the present application, “acquiring” includes any term with acquisition function such as querying, discovering and extracting, which is not limited in the present application.

[0050] As shown, the vehicle control method provided by the present application includes: Figure 2

[0051] Step S201: The controller acquires multi-modal environmental data of the vehicle.

[0052] The multi-modal environmental data refers to environmental data of different sources and different forms, which can comprehensively and multi-dimensionally reflect the environmental state and changes.

[0053] Different sources refer to different devices or technical means for acquiring multi-modal data. For example, weather data is acquired by monitoring methods such as multiple sensors or satellites, drones and the like.

[0054] Optionally, the sensors include a rainfall sensor, a vibration sensor, a humidity sensor, a temperature sensor, an illumination sensor, a particulate matter 2.5 (PM2.5) sensor and the like, and the monitoring methods include double camera (such as red green blue (RGB) camera + infrared camera) image monitoring, millimeter wave radar monitoring and the like.

[0055] Optionally, the multi-modal environmental data includes rainfall, vehicle vibration intensity, stain area of the vehicle windshield, humidity inside and outside the vehicle, temperature inside and outside the vehicle, vehicle splashing trajectory, dust diffusion speed, air quality and illumination intensity.

[0056] ​Exemplarily, the rain intensity is detected by the rain sensor based on the photoelectric feedback current difference value. The vehicle vibration intensity is monitored by the vibration sensor to distinguish the bumpy road surface from the human operation interference. The dirt type (mud, oil film, ice layer) and distribution area on the vehicle surface are identified by the dual camera. The humidity difference inside and outside the vehicle is detected by the humidity sensor to assist in determining the rain and fog weather. The water splashing trajectory of the preceding vehicle and the dust diffusion speed are detected by the millimeter wave radar.

[0057] Different forms refer to the formats of multi-modal data, including structured data (such as numerical value, table), unstructured data (such as image, audio, video), or semi-structured data (such as text report), and various forms of data.

[0058] In some embodiments, the multi-modal data has a timestamp and geospatial information, facilitating dynamic analysis and trend prediction.

[0059] In some embodiments, the controller directly monitors to obtain real-time multi-modal environment data and query multi-modal environment data at historical time points.

[0060] In some embodiments, the multi-modal data is stored in blocks based on timestamps, facilitating query and alignment.

[0061] Step S202: The controller determines the weight corresponding to each modal environment data in the multi-modal environment data.

[0062] The weight is used to quantify the contribution degree of different modalities or different features in the same modal environment data to vehicle environment recognition. By adjusting the weight, more attention can be paid to higher value data, and noise or redundant data can be suppressed.

[0063] In some embodiments, the weight corresponding to each modal environment data in the multi-modal environment data is determined or adjusted in the following manner, but is not limited thereto.

[0064] Method 1: The confidence of each modal environment data in the multi-modal environment data is obtained, and the weight corresponding to each modal environment data is determined based on the confidence.

[0065] The confidence refers to a quantitative evaluation value of the reliability, accuracy or credibility of each modal environment data.

[0066] Exemplarily, the confidence is determined or adjusted in the following steps, but is not limited thereto.

[0067] Step one: Calculate the historical false positive rate of the sensor that acquires multi-modal environmental data, for example, the false positive rate of a PM2.5 sensor in a sandstorm environment is ≤2% in the past 10 detections, and the confidence benchmark value is set to 0.9, with a full score of 1.0. Calculate the standard deviation of the data detected by the sensor, and adjust the confidence based on the standard deviation, for example: the fluctuation range of the data detected by the sensor is <5% for 5 consecutive minutes, and the confidence is increased by 10%.

[0068] Step two: cross-verify the detection results obtained by detecting the same environmental data in different ways, for example: the visual camera detects that the particle size is 50μm, and the radar detects that the dust diffusion speed is 1.5m / s, after normalization comparison, the data difference is within the allowed range (such as <15%), then the confidence of the visual camera and the radar is weighted by 10%, and the data difference exceeds the threshold (such as >30%), an abnormal log is recorded, and a third sensor (such as a humidity sensor) is started to arbitrate, based on the arbitration result, the confidence is dynamically adjusted, such as the difference between the detection result of the third sensor and the detection result of the radar is smaller, then the confidence of the radar is increased, and the confidence of the visual camera is decreased.

[0069] Step three: determine or adjust the confidence in combination with external data.

[0070] For example, the confidence is corrected in combination with navigation positioning (such as entering a tunnel), weather application programming interface (API) (such as real-time rainfall probability), timestamp (day / night mode), etc. For example: the navigation positioning indicates that the vehicle enters a "sandstorm high-risk area", and the PM2.5 sensor confidence is increased by 20%, and the timestamp indicates that the vehicle is in the night, and the infrared camera confidence weight is increased by 15%.

[0071] The above steps of determining or adjusting the confidence can be used in combination, and the execution order between the steps is not limited.

[0072] For example, the formula for determining the weight corresponding to each modal environmental data based on the confidence is as follows:

[0073]

[0074] wherein w i represents the weight corresponding to the i-th modal environmental data, c i represents the confidence of the i-th modal environmental data, c k represents the confidence of the k-th modal environmental data, and n represents the number of modalities of the multi-modal environmental data.

[0075] For example: the confidence of PM2.5 sensor data C1=0.85, the confidence of visual data C2=0.7, the confidence of radar data

[0076] C3=0.65, then the weight of PM2.5 sensor data W1=0.85 / (0.85+0.7+0.65)=0.39. In the case of navigation display entering the high sandstorm area, the weight of PM2.5 sensor data can be increased to 70%, and the weights of visual data and radar data are relatively reduced.

[0077] Method 2: Pre-configure the initial weight of each modal environmental data.

[0078] For example, the initial weight template of each type of vehicle environment is predefined, such as the weight of PM2.5 sensor in sandstorm environment accounts for 70%, when the vehicle environment is pre-identified by external data (such as weather forecast, navigation) as a certain type of vehicle environment, the initial weight template is directly called.

[0079] For example: Table 1 shows an example of an initial weight template.

[0080] Table 1: Initial weight template

[0081] Vehicle environment PM2.5 sensor data weight Vision data weight Radar data weight Sandstorm / haze 70％ 20％ 10％ Rainy night high speed 50％ 30％ 20％ Low temperature icing 40％ 40％ 20％

[0082] In the sandstorm environment, the weight of PM2.5 sensor data accounts for 70%, and based on real vehicle test data, the detection accuracy rate (92%) of PM2.5 sensor on particulate matter concentration in sand environment is significantly higher than that of visual (78%) and radar (65%).

[0083] In some embodiments, the weight is adjusted based on the initial weight and the confidence.

[0084] For example, the initial weight of each modal environmental data is determined, and the weight is determined based on the initial weight and the confidence.

[0085] Method 3: Determine or adjust the weight of each modal environmental data based on the change of the data.

[0086] Optionally, the change of each modal environmental data includes that the data exceeds / lowers a set threshold, the data change rate exceeds a change threshold, etc.

[0087] For example, the weight of each modal environmental data is temporarily determined or adjusted based on the change of the data. For example: the PM2.5 concentration detected by the PM2.5 sensor increases to 200%, the weight of the PM2.5 sensor increases from 60% to 80%, and the weights of the visual sensor and the radar each decrease by 10%, and after 30 seconds, the weights before adjustment are restored.

[0088] Experiments show that when the data of the sensor and the radar conflict, the humidity sensor is preferred for arbitration, and the discrimination accuracy rate of the humidity sensor data in dust and rain and fog environment reaches 89%.

[0089] The above methods can be used in combination, and the combination method is not limited.

[0090] In some embodiments, a weight corresponding to each modality of environmental data in the multimodal environmental data is allocated in real time through a dynamic weighting model.

[0091] Optionally, the dynamic weighted model includes an adaptive weighted model, an uncertainty weighted model, a deep reinforcement learning weighted model, etc.

[0092] Step S203: The controller determines a vehicle environment with the highest priority among at least one vehicle environment in which the vehicle is located based on the weight and the multimodal environment data.

[0093] Among them, the vehicle environment refers to the specific environment, conditions and dynamic interaction status of the vehicle during driving or parking, such as: the physical location of the vehicle, the natural environment, the interaction relationship between the vehicle and traffic participants, the operating status of the vehicle, etc.

[0094] For example, the vehicle environment is categorized based on spatial, temporal, weather, and interactive dimensions. For example, the vehicle environment categorized based on spatial dimensions includes highways, urban and rural roads, ramps, and parking lots. The vehicle environment categorized based on temporal dimensions includes nighttime, daytime, dusk, and winter. The vehicle environment categorized based on weather dimensions includes rainy, foggy, snowy, sandstorms, hail, and low-temperature environments. The vehicle environment categorized based on interactive dimensions includes traffic jams, tailgating, waiting at traffic lights, water splashing from the vehicle ahead, a sudden increase in pedestrian traffic, and car washes and repairs.

[0095] Optionally, the vehicle environment includes a congested environment, a night rainy environment, a sandstorm environment, a haze environment, a low temperature icy environment, a tunnel environment, an automatic car wash environment, an environment where the vehicle in front is splashed with water, and a rain and fog environment.

[0096] In some embodiments, multimodal environmental data is fused based on weights to obtain target environmental data, and at least one vehicle environment is determined based on the target environmental data. Fusion of multimodal environmental data, by integrating environmental data of multiple modalities from different data sources and extracting complementary information between the data, generates more comprehensive and accurate target environmental data. This overcomes the limitations of a single data source and a single modality of environmental data, and improves the reliability of environmental monitoring, analysis, and decision-making. In some embodiments, the steps for fusing multimodal environmental data based on weights are as follows:

[0097] Step 1: Preprocess the multimodal environment data: align different modal environment data and extract data features of the multimodal environment data.

[0098] Step 2: dynamically determine the weight of each modal environment data.

[0099] Step 3: linearly transform the data features, and weight and fuse the data features based on the weights to obtain target environment data.

[0100] In some embodiments, at least one vehicle environment is determined based on the multi-modal environment data / target environment data and the running information of the vehicle.

[0101] The running information of the vehicle refers to information describing the driving state, performance and system state of the vehicle.

[0102] Optionally, the running information of the vehicle includes vehicle speed, vehicle speed gear, power, charging state, window / door state, etc.

[0103] Exemplarily, the manner of determining the vehicle environment based on the multi-modal environment data and the running information of the vehicle is as follows, but is not limited thereto.

[0104] Example 1: the vehicle speed continuously below 15 km / h for more than 5 minutes, the camera detects that the light transmittance of the front windshield of the vehicle decreases by >2% per minute, and the PM2.5 sensor feedbacks that the concentration of particulate matter outside the vehicle is >75 μg / m 3 , it is determined as an urban congestion environment.

[0105] Example 2: the light intensity is <10 lux, the rainfall intensity detected by the rainfall sensor is >4 mm / h, and the radar monitors the vehicle speed to be >80 km / h, which is determined as a rainy night highway environment.

[0106] Example 3: the PM2.5 outside the vehicle is >150 μg / m 3 for 10 minutes, the camera identifies that the average particle size of particulate matter is >50 μm, and the radar detects that the diffusion speed of dust is >1.5 m / s, which is determined as a sandstorm / haze environment.

[0107] Example 4: the surface temperature of the windshield is ≤0℃, the humidity outside the vehicle is >80%, and the infrared camera identifies static crystalline texture, which is determined as a low-temperature icing environment.

[0108] Example 5: the navigation positioning vehicle enters a tunnel, the light intensity suddenly decreases from 800 lux to 50 lux, the camera identifies the reflection features at the top of the tunnel, which is determined as a tunnel environment.

[0109] Example 6: the parking radar detects that the distance of the bilateral obstacles is <30 cm, the vehicle speed is 0 and the gear is N, which lasts for 15 seconds, which is determined as an automatic car washing environment.

[0110] Example 7: the millimeter wave radar detects that the diffusion speed of the water splashing trajectory of the front vehicle is >3 m / s, the camera identifies the fan-shaped water splash shape, and the vehicle body vibrates at high frequency, which is determined as a front vehicle water splashing environment.

[0111] Example 8: The humidity difference between inside and outside the vehicle is greater than 30%, the camera detects that the windshield is uniformly blurred, and the navigation weather API returns a regional rainfall probability greater than 70%, which is determined as a rain and fog weather environment.

[0112] In some embodiments, at least one vehicle environment is determined based on the target environment data, the running information of the vehicle, and a scene classification model.

[0113] Exemplarily, the target environment data and the running information are input into the scene classification model, a predefined vehicle environment label is identified by the scene classification model, and the vehicle environment is determined. Optionally, the scene classification model includes a multi-modal and attention model and a convolutional neural network (CNN) based recognition model.

[0115] In some embodiments, the scene classification model and the dynamic weighting model are trained based on a historical multi-modal environment data set.

[0116] Exemplarily, the scene classification model combines the dynamic weighting model to determine at least one vehicle environment based on the multi-modal environment data and the running information of the vehicle. For example, the dynamic weighting model determines the weight corresponding to each modal environment data based on an attention mechanism, Bayesian inference, etc., performs feature-level fusion or decision-level fusion on the multi-modal environment data, inputs the feature vector (target environment data) obtained after the fusion into the scene classification model, and outputs multiple vehicle environment labels.

[0117] In some embodiments, based on the multi-modal environment data, a vehicle environment with the highest priority is determined from at least one vehicle environment.

[0118] The priority of the vehicle environment is used to represent the size of the cleaning demand corresponding to different vehicle environments in a complex vehicle environment. For example, the priority of the rainy environment is higher than that of the congestion environment, and the demand for cleaning in the rainy environment is higher, and the wiper device needs to be controlled according to the wiper device control strategy corresponding to the rainy environment.

[0119] Optionally, the priority of the vehicle environment is pre-configured, or the priority of the vehicle environment is dynamically adjusted based on the identification of the vehicle environment.

[0120] In some embodiments, the risk score of each vehicle environment is calculated based on the multi-modal environment data, and the priority of the vehicle environment is determined or adjusted based on the risk score.

[0121] Exemplarily, the initial risk score is calculated based on the weight of each modal environment data by the dynamic weighting model, wherein Si is the normalized data value of the i-th sensor, w i is its dynamic weight (calculated by environmental confidence).

[0122] In some embodiments, the risk score is corrected by a long short-term memory (LSTM) time series prediction.

[0123] Illustratively, the LSTM network predicts the trend of environmental changes in the next 5 minutes based on historical data, and outputs a correction coefficient K (range 0.8-1.2). The corrected risk score is obtained by multiplying the risk score by the correction coefficient.

[0124] For example, in a rain and fog environment, if the LSTM predicts that the visibility will continue to decrease, the correction coefficient K = 1.2, and the corrected risk score is increased by 20%.

[0125] As an example, in a rain night highway environment, the rainfall sensor data: 4mm / h (weight 50%), the normalized value is 0.8; the infrared camera data (raindrop coverage rate): 30% (weight 30%), the normalized value is 0.7; the vibration sensor data (vehicle speed associated with jolt value): 0.6 (weight 20%), the normalized value is 0.5, then the risk score is 0.8x0.5+0.7x0.3+0.5x0.2=0.71. The LSTM predicts a correction coefficient K = 1.1 (rainfall enhancement trend), and the corrected risk score based on the correction coefficient is 0.71x1.1=0.781. In the case of full score 1.0 corresponding to 100 points, the corrected risk score is 78.1 points.

[0126] Illustratively, the vehicle environment is classified based on the risk score, and the priority of the vehicle environment is determined based on the type of the vehicle environment.

[0127] As shown in Figure 3 , the step of classifying based on the risk score includes:

[0128] Step S301: classifying the vehicle environment based on the risk score.

[0129] Illustratively, the vehicle environment is classified into a safety scenario, an efficiency scenario, and a maintenance scenario.

[0130] Step S302: the risk score is greater than or equal to 70, and the vehicle environment is a safety scenario. There are safety hazards in the safety scenario, for example, rain night highway, front car splashing water, sandstorm.

[0131] Step S303: the risk score is less than 70 and greater than or equal to 40, and the vehicle environment is an efficiency type scene. The efficiency type scene focuses on energy efficiency and user experience, for example: urban congestion, tunnel passing. The wiper device is adjusted to a low-frequency wiping and a mute mode, and the water spraying amount is limited.

[0132] Step S304: the risk score is less than 40, and the vehicle environment is a maintenance type scene. Or the vehicle is in a non-driving state, for example: automatic car washing, parking charging. The wiper motor is locked, and the wiper motor is adjusted to a sleep mode.

[0133] Among them, the priority of the safety type scene is higher than the priority of the efficiency type scene, and the priority of the efficiency type scene is higher than the priority of the maintenance type scene.

[0134] For example: the risk score of the rainy night high-speed environment is 78.1, triggering the control strategy of the safety type scene (such as high-frequency wiping + water spraying delay), and the risk score of the urban congestion environment is 35, triggering the control strategy of the efficiency type scene (such as low-frequency wiping + water spraying amount limitation). For example: when the vehicle speed is greater than 80 km / h, the control strategy corresponding to the high-speed environment is preferentially executed, such as water spraying delay anti-reflection; when the navigation displays that the vehicle enters a tunnel, the control strategy is forcibly covered to clean in a mute mode, and the water spraying is disabled.

[0135] In some embodiments, if the risk score of a certain vehicle environment suddenly increases (such as the risk score of the sandstorm environment or the rainy high-speed environment increases from 60 to 85), the priority of the vehicle environment is immediately raised to the priority corresponding to the safety type scene.

[0136] Step S204: the controller controls the wiper device of the vehicle based on the vehicle environment with the highest priority.

[0137] Among them, the wiper device is a cleaning device installed on the front windshield and / or rear windshield of the vehicle, used to remove water droplets, snow or dust in rainy, snowy or dirty conditions, to ensure the driver's clear vision and ensure driving safety.

[0138] Exemplarily, the wiper device includes a wiper and a water spraying component. Specifically, the wiper includes a wiper arm, a wiper blade, a wiper motor, etc., and the water spraying component includes a cleaning liquid storage tank, an electric water spraying pump, a water spraying port, a water conveying pipeline, etc.

[0139] In some embodiments, the control strategy corresponding to the vehicle environment with the highest priority is queried, and the wiper device of the vehicle is controlled based on the control strategy.

[0140] Among them, the control strategy is used to indicate the control scheme of the wiper and / or water spraying component in the wiper device.

[0141] Optionally, the control scheme of the wiper includes the cleaning speed, the cleaning mode, the cleaning pressure and the heating wire temperature of the wiper.

[0142] wherein the wiper cleaning mode includes single stroke, intermittent stroke, low speed continuous stroke, high speed continuous stroke, manual single stroke, etc. The cleaning pressure refers to the contact pressure between the wiper blade and the windshield, which is provided by the spring tension of the wiper arm, ensuring that the wiper blade closely fits the glass surface and effectively removes water and stains. The heating wire temperature refers to the temperature of the heating wire embedded in the wiper blade or wiper arm, which is suitable for low-temperature icing environment.

[0143] Optionally, the control scheme of the water spraying component includes the water spraying amount, water spraying time and water spraying pressure of the water spraying component.

[0144] In some embodiments, the controller controls the vehicle's windows, air conditioner and sensors, etc. in linkage based on the control strategy.

[0145] Exemplarily, the vehicle environment and the corresponding control strategy are as follows:

[0146] Urban congestion environment: the wiper switches to intermittent mode (such as single stroke every 45 seconds); the automatic water spraying function of the water spraying component is disabled, and the water spraying amount is limited to 50% of the regular mode when manually triggered; the vehicle is in hybrid / electric mode and the battery power is <30%, the wiper motor power is limited to 70%. The above control strategy makes the windshield transmittance recover from 70% to 85% and the motor temperature drop by 40% when the vehicle is in urban congestion environment.

[0147] Rainy night high-speed environment: in response to the light sensor detecting strong light on the opposite lane, the water spraying action of the water spraying component is delayed until the strong light ends; the infrared camera is enabled to assist in raindrop detection, and the wiper frequency is increased to 2 times per second; double wiper strokes are immediately executed after water spraying to eliminate water stain reflection. The above control strategy reduces the number of false water spraying by 90% and improves the accuracy of raindrop recognition to 98% in strong light environment.

[0148] Sandstorm / haze environment: the water spraying pressure of the water spraying component is increased to 1.5 times the regular value, and the water amount is increased by 30%; the wiper arm pressure is increased to 18N to ensure the peeling of attached sand particles; the air conditioner automatically switches to internal circulation and prompts the driver "filter core remaining life: 150km". The above control strategy improves the sand particle removal efficiency by 60% and shortens the filter clogging alarm response time to 5 seconds.

[0149] Low-temperature icing environment: the wiper blade heating wire is heated at a rate of 5℃ / min for 10 minutes; the wiper motor power is cut off when the vibration sensor detects a resistance >2.5N; the vehicle central control screen pops up "ice layer detected, please manually remove ice". In the above control strategy, the ice layer melting rate is >95% after 10 minutes of heating wire preheating, and the motor stall failure rate is reduced to 0.1%.

[0150] In tunnels: Automatic water spray is disabled; wiper motor speed is reduced by 20%, switching to silent mode; manual water spray is followed by a single, powerful wipe, maintaining silent mode for five minutes. This control strategy has reduced complaints about glare from water stains in tunnels by 85% and reduced cabin noise by 10 decibels.

[0151] In an automated car wash environment, the wipers are forced off, the wiper arm motor enters a mechanical lock state, and the vehicle's central control screen displays "Car wash mode activated, please shift to D to exit." This control strategy reduces the number of false wiper triggerings to zero, eliminating the risk of motor mechanical damage.

[0152] In the event of water splashing from a preceding vehicle, the system activates high-frequency wipers (3 times / second) 200ms in advance for 10 seconds; automatically closes the driver's side window; and activates the front and rear fog lights in response to camera field of view blur exceeding 40%. This control strategy reduces the duration of water splash obstruction to 0.5 seconds and reduces side window mist intrusion by 70%.

[0153] In foggy and rainy weather, the system automatically turns on the front and rear fog lights; reduces the wiper interval from 30 seconds to 15 seconds; and activates the front windshield defogger mode using the vehicle's air conditioning. This control strategy improves visibility by 50% and triples defogger efficiency in foggy conditions.

[0154] For example, if Figure 4 As shown, the PM2.5 sensor 400 detects that the PM2.5 concentration is greater than 150 μg / m 3 , camera 401 detects particles with a diameter greater than 50 μm 3 The millimeter-wave radar 402 detects that the dust speed is greater than 1.5m / s, identifies the vehicle environment as a sandstorm environment, triggers the controller 403 to control the wiper equipment to spray water at high pressure, increases the wiper arm pressure to 18N, and controls the air conditioner 404 to switch to internal circulation, and the central control screen 405 displays the filter life warning.

[0155] In some embodiments, different control strategies are triggered based on each multimodal data in turn. Figure 5 As shown, the triggering methods include:

[0156] Step S500: Start.

[0157] Step S501: The temperature sensor detects the temperature.

[0158] Step S502: If the temperature is less than or equal to 0°C, execute step S503; otherwise, execute step S513.

[0159] Step S503: The humidity sensor detects the humidity.

[0160] Step S504: If the humidity is greater than 80°C, execute step S505; otherwise, execute step S513.

[0161] Step S505: Infrared camera detects crystalline texture.

[0162] Step S506: If crystalline texture is identified, proceed to Step S508, otherwise proceed to Step S507.

[0163] Step S507: Prompt manual de-icing.

[0164] Step S508: Pre-heat wiper heater, proceed to Step S509.

[0165] Step S509: Vibration sensor detects resistance.

[0166] Step S510: If resistance is greater than 2.5 N, proceed to Step S512, otherwise proceed to Step S511.

[0167] Step S511: Continue heating the heater.

[0168] Step S512: Switch wiper motor power, proceed to Step S513.

[0169] Step S513: End.

[0170] In some embodiments, a high-priority vehicle environment can interrupt the control strategy of a low-priority vehicle environment, for example: detecting a preceding vehicle splashing water in an automatic car washing mode, immediately exiting the car washing mode and its control strategy.

[0171] In some embodiments, the risk score is also used to adjust or link control strategies.

[0172] Illustratively, the higher the risk score, the more aggressive the cleaning action of the wiper device. For example: risk score > 70, wiper device performs high-frequency wiping (1.5 times / second) and high-pressure water spraying (1.5 times pressure); risk score is between 40-70, wiper device performs regular wiping (1 time / second), standard water spraying; risk score < 40, wiper device performs low-frequency wiping (1 time / 30 seconds) and limits water spraying. For another example: when the risk score is below a score threshold (e.g., the vehicle is in a city traffic environment), reduce the motor power, reduce energy consumption. For another example: when the risk score is > 80, the vehicle's central control screen prompts "high-risk environment, suggest slowing down"; when the risk score is > 90 and lasts for 5 minutes, automatically trigger the vehicle emergency mode (e.g., double flash lights on, wiper running at full speed).

[0173] Illustratively, the controller generates a priority instruction (e.g., "safety class: rainy night on highway"), calls a pre-defined control strategy (e.g., high-frequency wiping + infrared camera auxiliary enhancement), and issues it to the wiper device and linked devices (e.g., fog lamp, air conditioner) for control.

[0174] In some embodiments, based on the multi-modal environment data, a change of the vehicle environment in a future time period is predicted, and a control strategy is adjusted based on the change; and the rain brush device of the vehicle is controlled by using the adjusted control strategy.

[0175] Exemplarily, the change of the vehicle environment in the future time period is predicted by using LSTM time series.

[0176] Optionally, the change of the vehicle environment in the future time period includes a rainfall trend, a freezing rate, a visibility attenuation, etc.

[0177] Exemplarily, based on historical rainfall sensor data, a rainfall intensity change in a future 5 minutes is predicted, and a rain brush frequency is dynamically adjusted, for example, when the rainfall is predicted to be enhanced, the rain brush frequency is adjusted to a high frequency in advance. In a low temperature environment, a windshield surface freezing rate is predicted by using temperature, humidity and infrared camera data, and when the freezing rate exceeds a threshold value, a preheating of an electric heating wire is performed. In a rain and fog environment, a visibility decline trend is predicted based on humidity difference and camera blurring data, and when the visibility decline reaches an attenuation threshold value, a fog lamp and a defogging function are turned on.

[0178] For example, when a windshield light transmittance decline rate > 2% / min and lasts for 3 minutes, a risk score > 60 (full score 100), an adjustment is made to the control strategy, and the adjustment process is triggered only in a non-silent vehicle environment (such as a non-tunnel, a non-automatic car washing mode) to avoid invalid cleaning. When a dirt accumulation rate is in a low rate range (2-3% / min), the control strategy is adjusted to low pressure water spraying (80% of a regular pressure) and intermittent wiping (1 time / 10 seconds), and an air conditioner is started in an external circulation to reduce internal fogging. When the dirt accumulation rate is in a medium rate range (3-5% / min), the wiping frequency is adjusted to 1 time / 5 seconds, an infrared camera detection is enhanced, and a water stain removal effect is optimized.

[0179] Exemplarily, the control strategy adjustment includes an adjustment of water spraying, wiping and linkage components. For example, for mud type dirt, high pressure pulse water spraying (1.2 times pressure for a short time) is used to strip the adhering objects, for oil film type dirt, atomized water spraying (water amount increased by 20%) is used to cooperate with a wiper blade slow wiping, in a sandstorm environment, the wiping arm pressure is increased to 18N to ensure sand particle stripping, in a low temperature environment, the electric heating wire is preheated to 40℃ before wiping to avoid scratching the glass by ice layer, and navigation data is synchronized (such as disabling water spraying when entering a tunnel).

[0180] Based on the above implementation manner, as shown in Figure 6 Another vehicle control method provided by the present application includes:

[0181] Step S600: Start.

[0182] Step S610: Collect multi-modal environment data by using a sensor.

[0183] The sensors include a rain sensor, a vibration sensor, a camera, a humidity sensor, and a millimeter wave radar.

[0184] Step S620: The dynamic weighting model performs data fusion.

[0185] The step of data fusion performed by the dynamic weighting model includes determining confidence, weight distribution, and data fusion.

[0186] Step S630: The scene classification model determines at least one vehicle environment.

[0187] Step S640: A risk score is generated to determine the priority of the at least one vehicle environment.

[0188] Step S650: The vehicle environment with the highest priority is determined.

[0189] Step S660: A control strategy for the vehicle environment with the highest priority is determined.

[0190] Step S670: The control strategy is adjusted through LSTM time series prediction.

[0191] Step S680: The wiper device and other vehicle components are connected based on the control strategy.

[0192] Step S690: End.

[0193] In summary, the technical scheme provided by the embodiments of the present application determines the vehicle environment in which the vehicle is located based on multi-modal environmental data, comprehensively and multi-dimensionally reflects the information of the vehicle environment in which the vehicle is located, supplements semantic information through the information complementation between each modal environmental data, reduces the error caused by environmental interference of a single data source, improves the stability and accuracy of vehicle environment recognition, and thus adapts to more complex vehicle environment recognition requirements, so that the wiper device of the vehicle can be flexibly controlled according to the complex vehicle environment, the cleaning of the wiper device is more adaptive to the cleaning requirements of the vehicle environment, and the cleaning effect on the vehicle is improved.

[0194] As shown in Figure 7 The vehicle control device provided by the present application can include an acquisition module 701, a processing module 702, and an execution module 703. The acquisition module 701 is configured to perform the operation of step S201 in the schematic method; the processing module 702 is configured to perform the operations of steps S202 and S203 in the schematic method; and the execution module 703 is configured to perform the operation of step S204 in the schematic method. Figure 2 Figure 2 Figure 2

[0195] ​​​The above describes the solutions provided by the embodiments of the present application mainly from the perspective of systems and methods. To implement the above functions, the vehicle control method, device or controller includes hardware structures and / or software modules corresponding to each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of the examples described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a certain function is implemented in hardware or computer software driven hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0196] The embodiments of the present application can divide the functional modules of the vehicle control method, device or controller according to the vehicle control method described above. For example, the vehicle control method or controller can include functional modules corresponding to each function division, or two or more functions can be integrated into one processing module. The integrated module can be implemented in the form of hardware or software functional module. It should be noted that the division of modules in the embodiments of the present application is illustrative, and is only a logical function division. Actual implementation can have another division method.

[0197] The power supply method and power supply system provided by the embodiments of the present application can be applied in a vehicle. The vehicle can also be referred to as a vehicle, a mobile carrier, an electric vehicle (EV), a hybrid electric vehicle (HEV), a plug-in hybrid electric vehicle (PHEV), a fuel cell vehicle (FCV), an autonomous vehicle, an intelligent and connected vehicle (ICV), a driverless vehicle, etc.

[0198] In the embodiments of the present application, the vehicle can be a car, a sport utility vehicle (SUV), a truck, an electric vehicle, a motorcycle, a tricycle, a special vehicle (such as an ambulance, a fire truck, a police car, etc.), a driverless taxi, an intelligent and connected bus, an autonomous logistics vehicle, an electric truck, etc. In addition, the method is also applicable to various special vehicles, such as agricultural vehicles, mining vehicles, forestry vehicles, airport vehicles, port vehicles, etc. The present application does not make specific limitations in this regard.

[0199] As Figure 8 shown, the electronic device provided by the embodiments of the present application can include a processor 801, a bus 802, a communication interface 803, and a memory 804. The processor 801, the memory 804, and the communication interface 803 communicate through the bus 802. It should be understood that the number of processors and memories in the network device is not limited by the present application.

[0200] The bus 802 can be a PCI bus or an extended industry standard architecture (EISA) bus, or a USB bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 only one line is used in the figure, but it does not mean that there is only one bus or only one type of bus. The bus 802 can include a path for transmitting information between various components (for example, the memory 804, the processor 801, and the communication interface 803) of the network device.

[0201] The processor 801 can include any one or more of a CPU, a graphics processing unit (GPU), a microprocessor (MP), or a digital signal processor (DSP), etc.

[0202] The memory 804 can include a volatile memory, such as a random access memory (RAM). The processor 801 can also include a non-volatile memory, such as a read-only memory (ROM), a flash memory, a mechanical hard disk drive (HDD), or a solid state drive (SSD).

[0203] The communication interface 803 uses a transceiver module such as, but not limited to, a network interface card, a transceiver, etc. to realize the communication between the network device and other devices or communication networks.

[0204] The memory 804 stores executable program codes, and the processor 801 executes the executable program codes to realize the functions of the foregoing method embodiments, respectively. That is, the memory 804 stores instructions for executing the vehicle control method.

[0205] In still another aspect, a computer readable storage medium is provided, and the computer readable storage medium stores at least one computer program, and the at least one computer program is loaded and executed by a processor to implement the vehicle control method provided by any of the above method embodiments.

[0206] In still another aspect, a computer program product is provided, and the computer program product includes a computer program or instructions, and when the computer program or instructions are executed by a processor, the vehicle control method provided by any of the above method embodiments is implemented.

[0207] It should be noted that the instructions in the computer readable storage medium or the one or more instructions in the computer program product are executed by the processor of the electronic device to implement each process of the above method embodiments, and the same technical effects as the above method can be achieved. To avoid repetition, details are not described here.

[0208] Through the description of the above embodiments, those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional modules is taken as an example for illustration, and in actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete the full classification part or part of the functions described above.

[0209] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules or units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another device, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed each other can be through some interface, indirect coupling or communication connection between the devices or units, which can be electrical, mechanical or other forms.

[0210] The units described as separate components can or can not be physically separate, and the components shown as units can be one physical unit or multiple physical units, that is, can be located in one place, or can be distributed to multiple different places. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment.

[0211] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit. When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present application essentially or say the part that contributes to the prior art or the whole classification part or part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions to make a device (which can be a single-chip microcomputer, a chip, etc.) or a processor execute the whole classification part or part of the steps of the method of each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a ROM, a RAM, a magnetic disk or an optical disk and various storage program codes.

[0212] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, any change or replacement within the technical scope disclosed in the present application should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A vehicle control method, characterized in that: The vehicle control method includes: Acquire multimodal environmental data of the vehicle; Determining a weight corresponding to each modal environment data in the multimodal environment data; determining, based on the weight and the multimodal environment data, a vehicle environment with the highest priority among the at least one vehicle environment in which the vehicle is located; Based on the vehicle environment with the highest priority, a wiper device of the vehicle is controlled.

2. The vehicle control method according to claim 1, characterized in that: The determining, based on the weight and the multimodal environment data, a vehicle environment with the highest priority among the at least one vehicle environment in which the vehicle is located, includes: Based on the weights, the multimodal environmental data are fused to obtain target environmental data; determining at least one vehicle environment based on the target environment data; A vehicle environment with the highest priority is determined from the at least one vehicle environment based on the multimodal environment data.

3. The vehicle control method according to claim 1, characterized in that: The determining of the weight corresponding to each modal environment data in the multimodal environment data includes: Obtaining a confidence level of each modal environment data in the multimodal environment data; A weight corresponding to each modal environment data is determined based on the confidence level.

4. The vehicle control method according to claim 3, characterized in that: The determining, based on the confidence level, a weight corresponding to each modal environment data includes: Determining an initial weight corresponding to each modal environment data; The weight is determined based on the initial weight and the confidence level.

5. The vehicle control method according to claim 2, characterized in that: The determining of at least one vehicle environment based on the target environment data includes: At least one vehicle environment is determined based on the target environment data, the operating information of the vehicle, and a scene classification model.

6. The vehicle control method according to claim 1, characterized in that: The controlling of the wiper device of the vehicle based on the vehicle environment with the highest priority includes: Query the control strategy corresponding to the vehicle environment with the highest priority; Based on the control strategy, a windshield wiper device of the vehicle is controlled.

7. The vehicle control method according to claim 6, characterized in that: The controlling of the wiper device of the vehicle based on the control strategy includes: Predicting changes in the vehicle environment in a future time period based on the multimodal environmental data; adjusting the control strategy based on the changes; The adjusted control strategy is used to control the wiper device of the vehicle.

8. The vehicle control method according to claim 6, characterized in that: The control strategy is used to indicate the control scheme of the wiper and / or water spray component in the wiper device; the control scheme of the wiper includes the cleaning speed, cleaning mode, cleaning pressure and heating wire temperature of the wiper; the control scheme of the water spray component includes the water spraying amount, water spraying time and water spraying pressure of the water spray component.

9. The vehicle control method according to claim 1, characterized in that: The vehicle environment includes: congested environment, night rainy environment, sandstorm environment, haze environment, low temperature icing environment, tunnel environment, automatic car wash environment, water splashing environment of the vehicle in front and rain and fog environment.

10. A vehicle, characterized in that: The vehicle applies the vehicle control method according to any one of claims 1 to 9 to control the wiper device of the vehicle to perform a cleaning action.

Images