Vehicle active splash-proof control method and device and vehicle

Abstract

The invention discloses a vehicle active splash-proof control method and device and a vehicle, and relates to the technical field of vehicle control. The method comprises the steps of obtaining road information of a first road section and information of traffic participants through multi-sensor fusion and cloud collaboration, determining a road ponding type according to the road information, determining a vehicle target control quantity according to the distance and the relative speed between a vehicle and the traffic participants based on different conditions of small-range ponding or large-range ponding, and determining a vehicle target control quantity according to the vehicle target control quantity. And finally, the suspension height and the vehicle speed of the vehicle are controlled according to the target control quantity, the driving experience is guaranteed, meanwhile, water is actively prevented from being splashed to pedestrians, and the accuracy of avoiding pedestrians or non-motor vehicles is improved.

Description

Technical Field

[0001] This application relates to the field of vehicle control technology, and more specifically, to a vehicle active water splash control method, device, and vehicle. Background Technology

[0002] Currently, active water splash prevention in vehicles mainly relies on alarm prompts from sensor signals, prompting drivers to take appropriate evasive actions. However, in most cases, it is not possible to take reasonable evasive actions based solely on alarm prompts. Currently, some drivers can precisely control the vehicle based on the alarm signal level, but this technology comes at the cost of sacrificing the driving experience.

[0003] Therefore, how to proactively prevent water from splashing onto pedestrians while minimizing the loss of driving experience has become an urgent problem to be solved. Summary of the Invention

[0004] In view of this, embodiments of this application propose a vehicle active water splash control method, device and vehicle, which can ensure the driving experience while more accurately avoiding water splashing onto pedestrians.

[0005] The following technical solution is adopted in this application.

[0006] In a first aspect, embodiments of this application provide a vehicle active water splash control method, applied to a vehicle control unit, the method comprising:

[0007] When the vehicle is traveling on the first road segment, road information and traffic participant information for the first road segment are acquired; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants; based on the road information, the type of road water accumulation is determined; the type of road water accumulation includes small-area water accumulation or large-area water accumulation; based on the type of road water accumulation and the traffic participant information, a target control quantity for the vehicle is generated; the target control quantity is used to indicate: the suspension height and speed of the vehicle determined based on the type of road water accumulation, the distance and relative speed between the vehicle and the traffic participants; the vehicle is controlled according to the target control quantity.

[0008] In some embodiments, obtaining the road information and traffic participant information of the first road segment includes: Initial images of traffic participants on the first road segment are acquired using a vehicle-mounted camera; target features are extracted and fused from the initial images to obtain target images; and information about the traffic participants is determined based on the target images.

[0009] In some embodiments, the vehicle communicates with the cloud, and obtaining road information for the first road segment includes: Historical road information of the first road segment is obtained from the cloud; a first road surface image of the first road segment is obtained through an in-vehicle camera; and the road information is determined based on the historical road information and the first road surface image.

[0010] In some embodiments, the method for determining the type of road flooding based on the road information includes: A second road surface image of the first road section is acquired using a radar sensor; the type of road flooding is determined based on the road information and the second road surface image.

[0011] In some embodiments, the method for determining the type of road flooding based on the road information and the second road surface image includes: Data fusion and pothole feature extraction are performed on the second road surface image to obtain the pothole contours and pothole depths of the flooded road; the type of road flooding is determined based on the road information, the pothole contours and pothole depths.

[0012] In some embodiments, generating the target control quantity for the vehicle based on the road flooding type and information about traffic participants includes: Based on the fact that the road flooding is of a small-scale nature, and considering that the traffic participant is located to the side of the vehicle's direction of travel, a first control quantity for the vehicle is calculated. The first control quantity includes the lift height of the vehicle's suspension. The first control quantity is determined based on the distance and relative speed between the vehicle and the traffic participant, the area of ​​the flooding, and the distance between the flooding location and the traffic participant.

[0013] In some embodiments, generating the target control quantity for the vehicle based on the road flooding type and information about traffic participants includes: Based on the fact that the road flooding is of a large scale, a second control quantity for the vehicle is calculated according to the location of the traffic participants in front of and / or to the side of the vehicle's direction of travel. The second control quantity includes the lifting height of the vehicle suspension and the reduction in vehicle speed. The second control quantity is determined based on the distance and relative speed between the vehicle and the traffic participants, as well as the area of ​​the flooding.

[0014] In some embodiments, generating the target control quantity for the vehicle based on the road flooding type and information about traffic participants includes: The water spray distance is determined based on the vehicle's speed; a target control quantity for the vehicle is determined based on the water spray distance and a first control quantity or a second control quantity for the vehicle; the target control quantity is used to indicate that the actual water spray distance is less than the distance between the vehicle and the traffic participant.

[0015] Secondly, embodiments of this application provide a vehicle active water splash control device, the device comprising: An acquisition module is used to acquire road information and traffic participant information for the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants; a determination module is used to determine the type of road water accumulation based on the road information; the type of road water accumulation includes small-area water accumulation and large-area water accumulation; a generation module is used to generate a target control quantity for the vehicle based on the type of road water accumulation and the traffic participant information; the target control quantity is used to indicate the suspension height and speed of the vehicle determined based on the type of road water accumulation, the distance and relative speed between the vehicle and the traffic participants; a control module is used to control the vehicle based on the target control quantity.

[0016] Thirdly, embodiments of this application provide a vehicle, the vehicle comprising: A control unit; a memory storing computer-readable instructions that, when executed by the control unit, implement the method described above.

[0017] In this application's solution, road information and traffic participant information for the first road segment are acquired through multi-sensor fusion and cloud collaboration to maintain high-precision identification even in adverse environments such as rain, fog, and night. Simultaneously, the type of road water accumulation is determined based on the road information. Depending on whether the water accumulation is small or large, the target control quantity for the vehicle is determined by the distance and relative speed between the vehicle and traffic participants. The control quantity is updated in real time with the dynamic changes of the vehicle, traffic participants, and water accumulation, thus ensuring a smooth driving experience while proactively preventing water from splashing onto pedestrians and improving the accuracy of avoiding pedestrians or non-motorized vehicles.

[0018] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0020] Figure 1 This is a schematic diagram of a scenario for an active water splash control method for vehicles provided in an embodiment of this application.

[0021] Figure 2 This is a flowchart illustrating a vehicle active water splash control method provided in an embodiment of this application.

[0022] Figure 3 This is a flowchart illustrating a method for identifying traffic participants provided in an embodiment of this application.

[0023] Figure 4 This is a flowchart illustrating a method for determining road information provided in an embodiment of this application.

[0024] Figure 5 This is a flowchart illustrating a method for determining the type of road flooding, as provided in an embodiment of this application.

[0025] Figure 6 This is a flowchart illustrating a method for obtaining a target control quantity, provided in an embodiment of this application.

[0026] Figure 7 This is a flowchart illustrating another active water splash control method for vehicles provided in an embodiment of this application.

[0027] Figure 8 This is a schematic diagram of the structure of a vehicle active water splash control device provided in an embodiment of this application.

[0028] Figure 9 This is a structural schematic diagram of a vehicle provided in an embodiment of this application.

[0029] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through specific embodiments. Detailed Implementation

[0030] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0031] Conventional technologies primarily emphasize preventing water splashes on pedestrians or non-motorized vehicles when driving through puddles, without addressing the issue of large bodies of water. However, in reality, drivers often fail to accurately predict the status of pedestrians or non-motorized vehicles, resulting in splashes. Furthermore, considering the potential for decreased accuracy of onboard sensors in extreme weather conditions, vehicles may struggle to promptly and accurately identify pedestrians or non-motorized vehicles, reducing the accuracy of proactive water avoidance. Therefore, the challenge lies in how to proactively prevent water splashes on pedestrians while maintaining a comfortable driving experience, and how to improve the accuracy of pedestrian / non-motorized vehicle avoidance.

[0032] The vehicle active water splash prevention control method provided in this application aims to solve the above-mentioned technical problems of the prior art.

[0033] The technical solution of this application and how it solves the above-mentioned technical problems will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0034] Figure 1 This is a schematic diagram illustrating a scenario for an active water splash control method for vehicles provided in an embodiment of this application. Figure 1 As shown in the figure, the vehicle active water splash control method provided in this application includes a vehicle 110 and a cloud 120 as the execution subjects. The vehicle 110 includes a control unit 111, a sensor module 112, and a chassis system 113.

[0035] Optionally, the vehicle 110 and the cloud 120 can communicate wirelessly. This wireless connection may include protocols such as Transmission Control Protocol / Internet Protocol (TCP / IP), Wireless Local Area Network (WLAN), and Remote Direct Memory Access over Converged Ethernet (RoCE).

[0036] The following is combined Figure 1 The vehicle 110 and cloud 120 shown illustrate the vehicle active water splash control method provided in this application embodiment: First, based on historical road information obtained from the cloud 120 and road surface images obtained by the sensor module 112, the control unit 111 determines the road information on the first road segment. Based on the traffic participant images obtained by the sensor module 112, the control unit 111 determines the road information on the first road segment. The control unit 111 determines the type of road water accumulation based on the road information and, in conjunction with the traffic participant information, determines the target control quantity of the vehicle 110. The control unit 111 transmits the target control quantity to the chassis system 113 for execution to control the suspension height and speed of the vehicle 110.

[0037] Below Figure 1 Based on the vehicle 110 and cloud 120 shown, the active water splash control method for vehicles provided in the embodiments of this application will be further described, such as... Figure 2 The diagram shows a flowchart of a vehicle active water splash control method. In a specific embodiment, this vehicle active water splash control method can be applied to, for example... Figure 8 The vehicle active water splash control device 800 and the vehicle 900 equipped with the vehicle active water splash control device 800 are shown. Figure 9 The specific process of the embodiments of this application will be described below. Of course, it is understood that this method can be executed by a cloud server with computing power. The following will focus on... Figure 2 The process shown is described in detail. The vehicle active water splash control method is applied to the vehicle's control unit and may include the following steps 201 to 204.

[0038] Step 201: When the vehicle is traveling on the first road segment, obtain the road information and traffic participant information of the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants.

[0039] In this embodiment, the road information refers to the road surface information on the current travel segment of the vehicle, including the location and area of ​​water accumulation. The location of the water accumulation is defined as coordinate information in a world coordinate system.

[0040] In this embodiment of the application, the information of traffic participants is based on the position and movement status of pedestrians or non-motorized vehicles around the vehicle on the current road segment.

[0041] For example, to improve the accuracy of road water condition identification, a combination of vehicle sensor modules and cloud technology is used to compensate for the shortcomings of a single sensing source. The vehicle's forward-facing three-nozzle camera acquires real-time images of the road surface of the first road segment. Through coordinate transformation and integral calculation, the image pixel information is converted into the location and area of ​​accumulated water in the world coordinate system. Furthermore, in extreme weather conditions or when there is obstruction in front of the vehicle, the accuracy of the road surface images acquired by the vehicle's forward-facing three-nozzle camera decreases or the road surface images cannot be acquired. In such cases, historical road information for the first road segment obtained from the cloud can be combined to obtain the current road information for the vehicle. This historical road information mainly includes the historical location and area of ​​accumulated water on the road.

[0042] Furthermore, initial images of traffic participants on the first road segment are acquired using the vehicle's forward-facing tri-lens camera. These initial images are then input into a YW-YOLO target detection model improved based on the YOLOv5 network to identify traffic participants around the vehicle and determine their positions and motion states. This improves both recognition accuracy and detection speed, providing a precise data foundation for subsequent water splash control. The positions of traffic participants are determined using coordinates in the world coordinate system, and their motion states primarily include their direction and speed of movement.

[0043] Step 202: Determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding or large-scale flooding.

[0044] In this embodiment, the type of road water accumulation is determined based on characteristics such as the water accumulation area, the contour of the puddles in the water accumulation area, and the depth of the puddles. Among them, small-scale water accumulation usually refers to small, short-term, and scattered water accumulations formed in low-lying areas of the road surface due to local unevenness of the road surface (such as ruts, bumps, potholes, and shoulder subsidence) or momentary poor drainage; large-scale water accumulation can be understood as the formation of large-scale and continuous water accumulation on the road surface due to drainage system failure or terrain limitations (such as intersections, underpasses, and low-lying sections), which prevents surface runoff from being drained in time.

[0045] In this embodiment, the location of water accumulation, the outline of potholes, and the depth of potholes are obtained through multi-source fusion detection of microwave radar, lidar, and millimeter-wave radar of the vehicle, followed by feature extraction and algorithm optimization. Combined with road information, the type of road water accumulation on the first road segment is determined.

[0046] For example, a wide-area coarse scan is performed using microwave radar (24GHz / 77GHz) to locate the target area (water accumulation area) and its preliminary outline. The microwave radar has a detection accuracy of 2 degrees horizontal beam angle and a detection range of up to 10 nautical miles. A millimeter-level precise scan is performed using lidar (532nm blue-green laser) to determine the outline and depth of the target area. Its vertical depth measurement accuracy is 5cm@70m. A millimeter-wave radar (K-band) is used to supplement the shallow water topographic data and detect the water level of the target area. Its ranging resolution is 1mm.

[0047] For example, in order to obtain more accurate data such as the concave features of the target area, as well as the location, contour, depth and water level of the target area, the image data directly acquired by the vehicle's microwave radar, lidar and millimeter-wave radar are optimized.

[0048] Specifically, the data acquired by the lidar uses the RIATT (Robust Integrated Adaptive Threshold Transform) compression algorithm to eliminate water scattering noise, such as water surface reflection and suspended particles, which may cause clutter in the lidar point cloud. At the same time, single-photon level signal processing is used to improve the penetration of turbid water, which can increase the point cloud density by 300% in 60 meters of turbid water, ensuring that the lidar can capture the contour of the depression at the bottom of the target area, rather than just detecting the water surface.

[0049] Meanwhile, the data acquired by microwave radar utilizes the characteristic of the terahertz band (0.1-10THz) to penetrate water molecules. By inverting the concave contour of the target area through absorption-scattering signals, the output SAR (Synthetic Aperture Radar) image resolution can reach 0.31m. At the same time, basic grayscale enhancement is performed on the image to improve the contrast of subsequent edge detection.

[0050] In addition, the system combines millimeter-wave radar with the Doppler effect to calculate water surface velocity, corrects lidar blind zone data, and measures water level changes in the target area.

[0051] Also, by way of example, based on microwave radar, lidar, millimeter-wave radar... After the multi-source data was optimized (validation), the data sampling time was synchronized and the coordinate systems of each radar were aligned. The generalized likelihood ratio fusion was used to fuse the processed multi-source data and extract the features of the puddles, such as the precise contour and depth of the puddles.

[0052] Optionally, the time synchronization processing adopts hardware PPS pulses (Pulse Per Second), which can achieve microsecond-level time synchronization and control the data sampling interval within 10ms; the alignment processing of each radar coordinate system adopts the ICP (Iterative Closest Point) algorithm to ensure that the three-dimensional point cloud data of the lidar and the three-dimensional image data of the microwave radar are detection results in the same time and the same spatial coordinate system, avoiding positional deviations during fusion.

[0053] Furthermore, when performing data fusion on the processed multi-source data, the generalized likelihood ratio fusion method L is used. fusion The expression for is formula (1).

[0054] L fusion =α L microwave +β L lidar +γ D cross-validation Formula (1) Among them, L microwave For microwave radar data, α is the weighting coefficient of the microwave radar data, and α takes a value of 0.4; L lidar For LiDAR data, β is the weighting coefficient of the LiDAR data, and β takes a value of 0.5; D cross-validation The data is millimeter-wave radar data, and γ is the weighting coefficient of the millimeter-wave radar data, with a value of 0.1.

[0055] Among them, the multi-source data after data fusion processing realizes the initial fusion and credibility screening at the data level. The cross-validation penalty term can reduce the false detection rate to 5%, and the fused data with low false detection rate is output to the hole feature extraction stage, which lays the data weight foundation for subsequent edge fusion and reduces the computational amount of hole feature extraction.

[0056] Furthermore, the precise contours and depths of the concave areas in the target region are extracted, and then processed by lidar edge detection, microwave lidar edge enhancement, and weighted fusion (the lidar weight is 0.7 and the microwave weight is 0.3) to output the complete concave contours and depths.

[0057] For example, based on the extracted contour and depth of the crater, combined with the location and area of ​​the water accumulation, the puddles are divided into large-area water accumulation and small-area water accumulation, providing a basis for subsequent suspension height adjustment.

[0058] Step 203: Generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity is used to indicate the suspension height and speed of the vehicle determined based on the road flooding type, the distance between the vehicle and the traffic participants, and the relative speed.

[0059] In the embodiments of this application, the target control quantities are the vehicle's suspension height and speed, which are determined by the distance and relative speed between the vehicle and traffic participants.

[0060] For example, based on the position and movement of pedestrians or non-motorized vehicles around the vehicle on the current road segment, the distance and relative speed between the vehicle and the traffic participants are obtained through the difference frequency of the radar frequency modulation signal, providing dynamic data for adjusting the suspension height and vehicle speed to prevent water splashing.

[0061] The distance between the vehicle and traffic participants is estimated according to formula (2).

[0062] R=cT(f b+ +f b- ) / 2ΔF formula (2) Where R is the distance between the traffic participant and the vehicle; c is the speed of light; T is the Doppler frequency period; ΔF is the frequency modulation bandwidth of the modulation signal, i.e., the modulation signal period; f b+ f is the difference frequency of the rising edge of the triangular wave after mixing; b- It is the difference frequency of the falling edge of the triangular wave after mixing.

[0063] The relative speed between the vehicle and traffic participants is determined according to formula (3).

[0064] v=c(f b- f b+ ) / 4f0 formula (3) Where v is the relative speed between traffic participants and vehicles; f0 is the radar center operating frequency; c is the speed of light; f b+ f is the difference frequency of the rising edge of the mixed triangular wave; b- This is the difference frequency of the falling edge of the triangular wave after mixing.

[0065] Step 204: Control the vehicle according to the target control quantity.

[0066] For example, if the road flooding is of a small scale and there are pedestrians on the side of the vehicle's direction of travel, and it is determined that the vehicle does not have the conditions to change its driving path, the suspension height is raised based on the distance and relative speed between the vehicle and the traffic participants, combined with the water spray distance, without forcibly slowing down, so as to just avoid splashing water onto pedestrians, thus preserving the driving experience to the greatest extent and ensuring driving smoothness.

[0067] For example, if the road flooding is extensive and there are pedestrians on the side of the vehicle's direction of travel, and it is determined that the vehicle does not have the conditions to change its driving path, the suspension height is adjusted appropriately and the vehicle speed is reduced based on the distance and relative speed between the vehicle and the traffic participants, taking into account both splash protection and driving smoothness.

[0068] In this embodiment, firstly, road water condition recognition through multi-sensor fusion and cloud-based collaboration maintains high-precision recognition even in adverse environments such as rain, fog, and nighttime. Simultaneously, millimeter-level fine scanning with lidar and large-area coarse scanning with microwave radar are employed to achieve 3D modeling of water accumulation areas and potholes, providing precise data support for chassis control. Secondly, an improved YW-YOLO target detection model based on the YOLOv5 network enhances the detection capability and speed for pedestrians and non-motorized vehicles, meeting real-time requirements and ensuring timely issuance of control commands. Finally, based on road information, the type of road water accumulation is determined as small-scale or large-scale. The vehicle's suspension height and speed are determined based on the distance and relative speed between the vehicle and traffic participants, providing dynamic input for splash prediction and precise data for vehicle chassis control. This achieves proactive prevention of vehicle splashing onto pedestrians while ensuring a comfortable driving experience, improving the accuracy of avoiding pedestrians or non-motorized vehicles.

[0069] Regarding how to determine the information of traffic participants, embodiments of this application provide an optional implementation method, such as... Figure 3 The flowchart shown is a method for identifying traffic participants, which may specifically include the following steps 301 to 306.

[0070] Step 301: Acquire initial images of traffic participants on the first road segment using the vehicle-mounted camera.

[0071] In this embodiment of the application, the initial images of traffic participants on the first road segment are obtained by the vehicle's forward-facing tri-lens camera.

[0072] Step 302: Perform target feature extraction and target feature fusion processing on the initial image to obtain the target image.

[0073] For example, in order to improve the accuracy and efficiency of traffic participant detection, a YW-YOLO target detection model based on the YOLOv5 network is used to extract target features and fuse target features from the initial image to obtain the target image.

[0074] The YOLOv5 model is a real-time object detection network consisting of four core modules: input, feature extraction, feature fusion, and detection. The input module includes Mo-saic data augmentation, adaptive anchor box calculation, and adaptive image scaling to improve the model's ability to identify traffic participants in different scenarios. Feature extraction employs the New CSP-Darknet53 structure (a novel cross-stage partially connected Darknet-53 network) and the SPPF structure (Spatial Pyramid Pooling Fast) to extract traffic participant features. Feature fusion uses a PANet network combining a Feature Pyramid Network (FPN) and a Pyramid Attention Network (PAN) to achieve top-down transmission of high-level semantic information and bottom-up supplementation of low-level localization information. The detection module uses a bounding box loss function and non-maximum suppression to complete the final detection of target traffic participants.

[0075] Furthermore, RepGFPN is introduced into the feature fusion stage of the original YOLOv5 model. By controlling the number of channels at different scales and setting proportional feature mappings for different channel sizes, high-level semantic information and low-level spatial information are exchanged more efficiently. An adaptive fusion mechanism is also added to the feature fusion stage to improve the adaptability of feature fusion in complex road scenarios. At the same time, the SimAM attention module is introduced into the original YOLOv5 model to enhance the algorithm's feature extraction capability and reduce interference factors. In addition, Optimal Transport Assignment is used to optimize the loss function of the original YOLOv5 model, thereby improving the localization accuracy and classification accuracy of object detection.

[0076] Step 303: Determine the information of the traffic participants based on the target image.

[0077] For example, traffic participants around the vehicle are identified based on the target image, and the positions and motion states of the traffic participants are determined.

[0078] Step 304: Determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding or large-scale flooding.

[0079] Step 305: Generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity is used to indicate the suspension height and speed of the vehicle, determined based on the road flooding type, the distance between the vehicle and the traffic participants, and the relative speed.

[0080] Step 306: Control the vehicle according to the target control quantity.

[0081] The specific steps of steps 304 to 306 can be found in steps 202 to 204, and will not be repeated here.

[0082] In this embodiment, the YW-YOLO target detection model based on the YOLOv5 network is used to identify traffic participants. On the dataset of road pedestrians / non-motorized vehicles, the recognition accuracy is increased from 38.1% to 52.6%, and the detection speed is increased from 29.4fps to 30.8fps, providing accurate information on traffic participants for subsequent splash prevention control.

[0083] Based on the above, this application provides an optional implementation method for obtaining information on water accumulation in the current road segment where the vehicle is located, such as... Figure 4 The flowchart shown is a method for determining road information, which may specifically include the following steps 401 to 406.

[0084] Step 401: Obtain historical road information of the first road segment from the cloud.

[0085] In this embodiment of the application, historical road surface images of the first road segment, as well as the historical water accumulation locations and historical water accumulation areas corresponding to the road surface images, are obtained from the cloud.

[0086] Step 402: Obtain the first road surface image of the first road section using the vehicle-mounted camera.

[0087] In this embodiment of the application, the current first road surface image is obtained by the vehicle's forward-facing three-lens camera.

[0088] Step 403: Determine the road information based on the historical road information and the first road surface image.

[0089] In this embodiment, the 2D pixel coordinates of the first road surface image are converted into 3D road surface coordinates in the world coordinate system by the perspective transformation matrix, thereby realizing the conversion of pixel information to actual spatial information and obtaining the water accumulation position in the world coordinate system. The perspective transformation formula is referenced from formula (4).

[0090] Formula (4) Where (u, v) are pixel coordinates; P is the camera intrinsic parameter matrix; d is the depth value; (x, y, z) is the world coordinate system, where z=0 is the road surface plane, i.e. the plane where the water is located.

[0091] Among them, a parallax map is generated by the main camera, wide-angle camera and telephoto camera, and then the depth value d is obtained, referring to formula (5).

[0092] Formula (5) Where f is the camera focal length; b is the baseline distance of the binocular camera; and D is the pixel displacement (parallax) of the matching points in the left and right images.

[0093] Furthermore, in the transformed 3D road surface coordinate system, the projected area of ​​the water accumulation area is calculated by integration, referring to formula (6).

[0094] Formula (6) in, Let be the area of ​​the i-th water accumulation point cloud projection unit, in square centimeters; The road surface inclination angle is obtained by the vehicle inertial measurement unit; the projected area of ​​the accumulated water is obtained by integrating the area of ​​each unit and correcting for the road surface inclination angle.

[0095] For example, real-time cloud-based road information, such as road flooding reports and traffic department monitoring data, is obtained. This information is then combined with historical road surface images of the first road segment obtained from the cloud, as well as the historical flooding locations and areas corresponding to the road surface images. This comprehensive judgment, along with the road information determined through the first road surface images, addresses the issue of decreased accuracy of vehicle-mounted sensors under extreme weather conditions.

[0096] Step 404: Determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding or large-scale flooding.

[0097] Step 405: Generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity is used to indicate the suspension height and speed of the vehicle, determined based on the road flooding type, the distance between the vehicle and the traffic participants, and the relative speed.

[0098] Step 406: Control the vehicle according to the target control quantity.

[0099] The specific steps of steps 404 to 406 can be found in steps 202 to 204, and will not be repeated here.

[0100] Based on the above, this application provides an optional implementation method for determining the type of road flooding based on road information, such as... Figure 5 The flowchart shown is a method for determining the type of road flooding, which may specifically include the following steps 501 to 505.

[0101] Step 501: When the vehicle is traveling on the first road segment, obtain the road information and traffic participant information of the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants.

[0102] Step 502: Obtain a second road surface image of the first road segment using a radar sensor.

[0103] In this embodiment, the second road surface image includes a spectrum image from microwave radar, a point cloud image from lidar, and a sparse point cloud image from millimeter-wave radar. By leveraging the different functional focuses of each radar, comprehensive and high-precision detection of the puddle area is achieved.

[0104] Step 503: Determine the type of road flooding based on the road information and the second road surface image.

[0105] In this embodiment of the application, smooth and accurate contour data of puddles and potholes are output by using three-dimensional SAR images from microwave radar, point cloud images from lidar, and sparse point cloud images from millimeter-wave radar. Combined with road information, the type of road water accumulation on the first road segment is determined.

[0106] Step 503 further includes steps 513 to 523.

[0107] Step 513: Perform data fusion and pothole feature extraction on the second road surface image to obtain the pothole contour and pothole depth of the waterlogged road.

[0108] In this embodiment, the hole feature extraction mainly involves lidar edge detection, microwave radar edge enhancement, and weighted fusion of dual-source edge data.

[0109] For example, after optimizing the 3D point cloud data of the LiDAR, it includes information such as the spatial coordinates, depth, and point cloud density of the puddle area; the discrete 3D point cloud data is converted into structured voxel data to facilitate subsequent edge detection; then the gradient value of each voxel is calculated, and voxels with obvious gradient changes are selected as the edge points of the puddle detected by the LiDAR; the extracted 3D puddle edge data contains accurate edge details and depth information.

[0110] For example, after optimizing the 3D SAR image data of the microwave radar, it includes information such as the overall outline and grayscale gradient of the puddle area. By calculating the gradient changes of the pixels in the 3D SAR image using the 3D Sobel operator, the concave and convex edges in the 3D space can be captured, which corresponds to the concave outline of the puddle. Then, the gradient magnitude and direction of each pixel are filtered to transform the originally blurry puddle outline into clear edge data, which can make up for the possible local edge loss of the lidar. The extracted 3D concave edge data contains the complete overall outline of the puddle.

[0111] For example, the edge data of LiDAR and microwave radar are superimposed and calculated with a weight of 7:3. For local edge areas missing in the LiDAR point cloud, the edge data of microwave radar is used to fill in the gaps. For large-area contours detected by microwave radar, the details of LiDAR are superimposed to form an accurate overall contour.

[0112] Step 523: Determine the type of road flooding based on the road information, the contour of the pothole, and the depth of the pothole.

[0113] In this embodiment of the application, the type of road water accumulation is determined based on road information, pothole outline and pothole depth, with the water accumulation area of ​​the road as the main factor and the pothole outline and pothole depth as secondary factors.

[0114] For example, based on the fact that the water accumulation area is lower than the area threshold, the maximum depth of the crater is lower than the maximum depth threshold, and the average depth of the crater is lower than the average depth threshold, combined with the historical characteristics of no large-scale water accumulation in the road section obtained from the cloud, the road water accumulation type is determined to be small-scale water accumulation.

[0115] For example, based on the fact that the water accumulation area is higher than the area threshold, the maximum depth of the crater is higher than the maximum depth threshold, and the average depth of the crater is higher than or lower than the average depth threshold, combined with the information obtained from the cloud as historical characteristics of large-scale water accumulation on the road section, the road water accumulation type is determined to be large-scale water accumulation.

[0116] Step 504: Generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity is used to indicate the suspension height and speed of the vehicle, determined based on the road flooding type, the distance between the vehicle and the traffic participants, and the relative speed.

[0117] Step 505: Control the vehicle according to the target control quantity.

[0118] The specific steps of steps 501, 504 to 505 can be found in steps 201, 203 to 204, and will not be repeated here.

[0119] In this embodiment, the accuracy of pothole feature extraction is improved by comprehensively judging data from multiple dimensions obtained by multiple sensors, and the false positive rate is reduced by judging the water accumulation type based on road information, pothole outline and depth.

[0120] Building upon the above, this application provides an optional implementation method for generating target control quantities for vehicles based on road flooding types and information about traffic participants, such as... Figure 6 The flowchart shown is a method for obtaining a target control quantity, which may specifically include the following steps 601 to 607.

[0121] Step 601: When the vehicle is traveling on the first road segment, obtain the road information and traffic participant information of the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants.

[0122] Step 602: Determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding or large-scale flooding.

[0123] Step 603: Based on the fact that the road flooding type is small-scale flooding, and according to the fact that the traffic participant is located to the side of the vehicle's travel direction, calculate the first control quantity of the vehicle; the first control quantity includes the lift height of the vehicle suspension; the first control quantity is determined based on the distance and relative speed between the vehicle and the traffic participant, the flood area, and the distance between the flood location and the traffic participant.

[0124] In this embodiment of the application, the road water accumulation type is small-scale water accumulation. Based on four core parameters, namely the distance and relative speed between the vehicle and traffic participants, the water accumulation area, and the distance between the water accumulation and traffic participants, the first control quantity can be calculated by multi-parameter weighted fitting. The pedestrian side suspension is raised appropriately while the vehicle speed remains unchanged or is slightly reduced to ensure that the lifting height meets the requirements for splash prevention without excessive lifting. For example, the pedestrian side suspension of the vehicle is raised by 5-10cm while the vehicle speed remains unchanged or is reduced by 5-10km / h.

[0125] Step 604: Based on the fact that the road flooding type is large-scale flooding, and according to the location of the traffic participants in front of and / or to the side of the vehicle's direction of travel, calculate the second control quantity of the vehicle; the second control quantity includes the lifting height of the vehicle suspension and the reduction of the vehicle speed; the second control quantity is determined based on the distance and relative speed between the vehicle and the traffic participants, as well as the area of ​​the flooding.

[0126] In this embodiment, the road flooding type is large-scale flooding. Based on three core parameters—the distance and relative speed between the vehicle and other road users, and the flooded area—the suspension is raised uniformly across the entire vehicle to ensure vehicle stability while balancing splash protection and driving safety. The second control parameters, suspension lift height and vehicle speed, can be calculated independently and then cross-checked to avoid over-adjustment of a single parameter. Suspension lift height is the basic splash protection parameter, while vehicle speed reduction is the core splash protection parameter. Significantly raising the pedestrian-side suspension and reducing speed ensures a more significant splash reduction effect. For example, the vehicle suspension is raised by 20-30cm, and the vehicle speed is reduced by 15-20km / h, while avoiding sudden acceleration / sharp steering.

[0127] Step 605: Determine the water spray distance based on the vehicle speed.

[0128] In this embodiment of the application, the water spray distance is calculated using formula (7).

[0129] Formula (7) Where S is the horizontal spray distance of water, in meters (m); k is the aerodynamic enhancement coefficient, ranging from 1.2 to 1.5; v is the vehicle speed; h is the initial height of the water spray off the ground (approximately the tire radius), ranging from 0.30 to 0.40 meters (m); and g is the gravitational acceleration of 9.8 m / s². 2 .

[0130] Step 606: Determine the target control quantity of the vehicle based on the water spray distance and the first or second control quantity of the vehicle; the target control quantity is used to indicate that the actual water spray distance is less than the distance between the vehicle and the traffic participant.

[0131] For example, the first control quantity and the second control quantity are the basic control quantities. Combined with the water spray distance formula, the actual water spray distance under the basic control quantity is calculated and verified with the distance R between the vehicle and the traffic participant. If the actual water spray distance is less than R, the basic control quantity is the target control quantity; if not, the basic control quantity is iteratively optimized until the zero splash condition is met.

[0132] Step 607: Control the vehicle according to the target control quantity.

[0133] In the embodiments of this application, such as Figure 7 The flowchart of another vehicle active water splash control method is shown. First, pedestrians / non-motorized vehicles on the road are identified, and it is determined whether there are pedestrians / non-motorized vehicles in the direction of travel. If they are, the road water conditions are identified. If they are not, the vehicle's suspension system is controlled according to the size of the road potholes.

[0134] Secondly, based on the identification of puddles on the road and the determination that the water area is small, the relative position and relative speed between vehicles and pedestrians can be determined. This is to facilitate the subsequent determination of driving speed over a long period of time based on the size of the puddles, as well as the control of the vehicle's suspension system.

[0135] Finally, based on the identification of puddles on the road and the determination that the water area is a large area, the relative position and relative speed between the vehicle and the pedestrian are determined. This is to facilitate the subsequent determination of the suspension lifting time based on the size of the puddles, as well as the control of the vehicle's suspension system.

[0136] The specific steps of steps 601 to 602 and 607 can be found in steps 201, 202 and 204, and will not be repeated here.

[0137] In this embodiment, only one side of the suspension is adjusted when there is a small amount of water accumulation, while the suspension and vehicle speed are adjusted when there is a large amount of water accumulation, thus reducing the interference with the driving experience and balancing water splash prevention with the driving experience.

[0138] To achieve the functions of the above embodiments, the vehicle active water splash control method includes hardware structures and / or software modules corresponding to each function. Those skilled in the art should readily recognize that, based on the units and method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed through hardware or computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0139] exist Figures 2 to 7 Based on the vehicle active water splash control method shown, this application further describes a vehicle active water splash control device, such as... Figure 8 The diagram shows a structural schematic of a vehicle active water splash control device 800, which includes: an acquisition module 810, a determination module 820, a generation module 830, and a control module 840.

[0140] The acquisition module 810 is used to acquire road information and traffic participant information for the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants; wherein, the acquisition module 810 may include, for example, Figure 1 The sensor module 112 and cloud 120 in the vehicle 110 shown.

[0141] The determining module 820 is used to determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding and large-scale flooding; wherein, the determining module 820 may include, for example, Figure 1 The control unit 111 in the vehicle 110 shown.

[0142] Generation module 830 is configured to generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity indicates the suspension height and speed of the vehicle, determined based on the road flooding type, the distance between the vehicle and the traffic participants, and their relative speed; wherein, generation module 830 may include, for example... Figure 1 The control unit 111 and chassis system 113 in the vehicle 110 shown.

[0143] Control module 840 is configured to control the vehicle according to the target control quantity; wherein, control module 840 may include, for example, Figure 1 The control unit 111 and chassis system 113 in the vehicle 110 shown.

[0144] In some embodiments, the acquisition module 810 includes: acquiring an initial image of a traffic participant on the first road segment via a vehicle-mounted camera; performing target feature extraction and target feature fusion processing on the initial image to obtain a target image; and determining information about the traffic participant based on the target image.

[0145] In some embodiments, the acquisition module 810 further includes: acquiring historical road information of the first road segment from the cloud; acquiring a first road surface image of the first road segment through a vehicle-mounted camera; and determining the road information based on the historical road information and the first road surface image.

[0146] In other embodiments, the acquisition module 820 includes: acquiring a second road surface image of the first road segment via a radar sensor; and determining the type of road flooding based on the road information and the second road surface image.

[0147] In some embodiments, the determining module 820 further includes: performing data fusion and pothole feature extraction on the second road surface image to obtain the pothole contour and pothole depth of the flooded road; and determining the type of road flooding based on the road information, the pothole contour and pothole depth.

[0148] In some embodiments, the generation module 830 includes: calculating a first control quantity for the vehicle based on the road flooding type being small-scale flooding and according to the traffic participant being located to the side of the vehicle's direction of travel; the first control quantity includes the lift height of the vehicle suspension; the first control quantity is determined based on the distance and relative speed between the vehicle and the traffic participant, the flooded area, and the distance between the flooded location and the traffic participant.

[0149] In some embodiments, the generation module 830 further includes: calculating a second control quantity for the vehicle based on the road flooding type being large-scale flooding and according to the location of the traffic participant in front of and / or to the side of the vehicle's direction of travel; the second control quantity includes the lift height of the vehicle suspension and the reduction in vehicle speed; the second control quantity is determined based on the distance and relative speed between the vehicle and the traffic participant, as well as the flooding area.

[0150] In some embodiments, the generation module 830 further includes: determining a water spray distance based on the vehicle speed; determining a target control quantity for the vehicle based on the water spray distance and a first control quantity or a second control quantity for the vehicle; the target control quantity being used to indicate that the actual water spray distance is less than the distance between the vehicle and a traffic participant.

[0151] According to one aspect of the embodiments of this application, Figure 9 This is a structural schematic diagram of a vehicle provided in an embodiment of this application. Figure 9 As shown, the vehicle 900 includes a control unit 910 and one or more memory units 920. The one or more memory units 920 are used to store program instructions executed by the control unit 910. When the control unit 910 executes the program instructions, it implements the above-mentioned vehicle active water splash control method.

[0152] Furthermore, the control unit 910 may include one or more processing cores. The control unit 910 runs or executes instructions, programs, code sets, or instruction sets stored in the memory 920, and calls data stored in the memory 920. Optionally, the control unit 910 may be implemented using at least one hardware form selected from Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), and Programmable Logic Array (PLA). The control unit 910 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and Modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor and may be implemented using a separate communication chip.

[0153] According to one aspect of this application, a computer-readable storage medium is also provided, which may be included in the vehicle described in the above embodiments; or it may exist independently and not installed in the vehicle. The computer-readable storage medium carries computer-readable instructions that, when executed by a processor, implement the methods in any of the above embodiments.

[0154] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. Computer-readable storage media can be, for example, but not limited to: electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium can include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such transmitted data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0155] The units described in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the specific unit itself.

[0156] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0157] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the embodiments disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein.

[0158] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A method for active water splash prevention control of vehicles, characterized in that, Control units used in vehicles include: When the vehicle is traveling on the first road segment, the road information and traffic participant information of the first road segment are acquired; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants. Based on the road information, the type of road flooding is determined; the type of road flooding includes small-scale flooding or large-scale flooding. Based on the road flooding type and information about traffic participants, a target control quantity for the vehicle is generated; the target control quantity is used to indicate the suspension height and speed of the vehicle, determined based on the road flooding type, the distance and relative speed between the vehicle and the traffic participants. The vehicle is controlled according to the target control quantity.

2. The method according to claim 1, characterized in that, The acquisition of road information and traffic participant information for the first road segment includes: Initial images of traffic participants on the first road segment are obtained using vehicle-mounted cameras; The initial image is subjected to target feature extraction and target feature fusion processing to obtain the target image; Based on the target image, the information of the traffic participants is determined.

3. The method according to claim 1, characterized in that, The vehicle communicates with the cloud, and obtaining the road information of the first road segment includes: Historical road information of the first road segment is obtained from the cloud; The first road surface image of the first road section is obtained through the vehicle-mounted camera; The road information is determined based on the historical road information and the first road surface image.

4. The method according to claim 1, characterized in that, The method for determining the type of road flooding based on the road information includes: The second road surface image of the first road section is obtained using radar sensors; The type of road flooding is determined based on the road information and the second road surface image.

5. The method according to claim 4, characterized in that, The method for determining the type of road flooding based on the road information and the second road surface image includes: Data fusion and pothole feature extraction are performed on the second road surface image to obtain the pothole contours and pothole depths of the waterlogged road. The type of road flooding is determined based on the road information, the contour of the pothole, and the depth of the pothole.

6. The method according to claim 1, characterized in that, The method for generating the target control quantity for the vehicle based on the type of road flooding and information on traffic participants includes: Based on the fact that the road flooding is of a small-scale nature, and considering that the traffic participant is located to the side of the vehicle's direction of travel, a first control quantity for the vehicle is calculated. The first control quantity includes the lift height of the vehicle's suspension. The first control quantity is determined based on the distance and relative speed between the vehicle and the traffic participant, the area of ​​the flooding, and the distance between the flooding location and the traffic participant.

7. The method according to claim 1, characterized in that, The method for generating the target control quantity for the vehicle based on the road flooding type and information on traffic participants further includes: Based on the fact that the road flooding is of a large scale, a second control quantity for the vehicle is calculated according to the location of the traffic participants in front of and / or to the side of the vehicle's direction of travel. The second control quantity includes the lifting height of the vehicle suspension and the reduction in vehicle speed. The second control quantity is determined based on the distance and relative speed between the vehicle and the traffic participants, as well as the area of ​​the flooding.

8. The method according to any one of claims 1-7, characterized in that, The method for generating the target control quantity for the vehicle based on the type of road flooding and information on traffic participants includes: The water spray distance is determined based on the vehicle's speed. Based on the water spray distance and the vehicle's first or second control quantity, a target control quantity for the vehicle is determined; the target control quantity is used to indicate that the actual water spray distance is less than the distance between the vehicle and the traffic participant.

9. A vehicle active water splash control device, characterized in that, A control unit for a vehicle traveling on a first road segment, the device comprising: The acquisition module is used to acquire road information and traffic participant information for the first road segment; the road information includes the location and area of ​​water accumulation on the road; the traffic participant information includes the location and movement status of the traffic participants. The determination module is used to determine the type of road flooding based on the road information; the type of road flooding includes small-scale flooding and large-scale flooding. A generation module is used to generate a target control quantity for the vehicle based on the road flooding type and information about traffic participants; the target control quantity is used to indicate the suspension height and speed of the vehicle determined based on the road flooding type, the distance between the vehicle and the traffic participants, and the relative speed. A control module is used to control the vehicle according to the target control quantity.

10. A vehicle, characterized in that, The vehicles include: Control unit; A memory storing computer-readable instructions that, when executed by the control unit, implement the method as described in any one of claims 1 to 8.

Images