Automatic driving vehicle wading risk monitoring method, system, equipment and medium

Abstract

The invention discloses an automatic driving vehicle wading risk monitoring method, system and device and a medium, and relates to the technical field of environment perception for assisting safe driving. The method comprises the following steps: acquiring image data of a front-view camera of a vehicle, point cloud data of a vehicle gauge level laser radar and current vehicle information; performing preprocessing, target identification and region segmentation on the image data in sequence to determine a ponding region; comparing the point cloud data with the ponding area, screening out the point cloud data of the ponding area, and calculating to obtain the maximum ponding depth; and the limit water level height in the static state and the safe water depth in the current motion state are determined according to the vehicle information, and when the maximum accumulated water depth is larger than or equal to the numerical value interval of the limit water level height and the safe water depth, an early warning device is activated to conduct wading early warning. The method is helpful for the vehicle to accurately identify the water depth of the road, is combined with the self condition of the vehicle to study and judge the wading risk, and is suitable for measurement under complex weather and road surface conditions.

Description

Technical Field

[0001] The present invention relates to the field of environmental perception technology for assisting safe driving, and in particular to a method, system, equipment and medium for monitoring water wading risks of an autonomous driving vehicle. Background Art

[0002] In order to deal with the problems of road waterlogging caused by rainfall and potential vehicle failures and damage caused by wading, existing research and application work has been carried out around the monitoring of road water depth and vehicle wading risks. Relevant monitoring methods include: CNN feature extraction algorithm, YOLO target detection algorithm, ultrasonic sensor ranging, radar ranging, etc. In the existing technology, vehicle wading detection mainly uses a single sensor (such as millimeter wave radar, ultrasonic radar, camera), and some sensors are set on the side of the road, which cannot effectively judge the road waterlogging and vehicle wading risks. Summary of the invention

[0003] The purpose of the present invention is to provide a method, system, equipment and medium for monitoring the water wading risk of an autonomous driving vehicle, which can improve the vehicle's accuracy in determining the depth of water on the road and the water wading risk.

[0004] To achieve the above object, the present invention provides the following solutions:

[0005] A method for monitoring water wading risk of an autonomous driving vehicle, comprising:

[0006] Obtain image data from the vehicle's front-view camera, point cloud data from an automotive-grade laser radar, and current vehicle information; the vehicle information includes static information and dynamic information;

[0007] Preprocessing, target recognition and region segmentation are sequentially performed on the image data to determine the waterlogging area;

[0008] By comparing the point cloud data with the waterlogged area, the point cloud data of the waterlogged area is screened out, and the maximum waterlogged depth is calculated;

[0009] The extreme water level height in a stationary state and the safe water depth in a current moving state are determined respectively according to the vehicle information, and when the maximum water accumulation depth is greater than or equal to the numerical range of the extreme water level height and the safe water depth, the warning device is activated to issue a wading warning.

[0010] Optionally, the static information is built into the system to measure the extreme water level height to ensure vehicle safety in a stationary state; the dynamic information is obtained by connecting the system to the domain controller, including the vehicle's driving speed, ascent and descent slopes.

[0011] Optionally, the image data is preprocessed, target identified and region segmented in sequence to determine the waterlogged area, specifically including:

[0012] The image data is compressed, grayed, noise filtered, and enhanced in sequence to obtain a preprocessed image;

[0013] Performing target recognition on the preprocessed image, when the recognition result is that there is water accumulation, the corresponding image is saved as a waterlogged road surface image;

[0014] The flooded road surface image is input into the area segmentation module, and when the image confidence of the edge contour pixel data is not less than a preset threshold, the edge contour pixel data is judged to be qualified, and the flooded area is delineated based on all qualified data.

[0015] Optionally, the point cloud data is compared with the waterlogged area, the point cloud data of the waterlogged area is screened out, and the maximum waterlogged depth is calculated, specifically including:

[0016] The point cloud data obtained by the automotive-grade laser radar is compared with the image of the waterlogged area, the point cloud data of the waterlogged area is screened, and the distance between the laser radar and the water surface and the road surface is calculated using the point cloud data; the point cloud data of the waterlogged area includes the coordinates, color and reflection intensity of the laser point;

[0017] Calculation is performed based on the point cloud data of the waterlogged area and the radar elevation angle information to obtain a depth information set of the waterlogged area, and the maximum waterlogged depth is screened out from the depth information set of the waterlogged area. The radar elevation angle information is the angle information between the laser radar transmitting antenna and the horizontal plane.

[0018] Optionally, the judgment process of the water wading alarm is:

[0019] The smooth particle method is used to calculate the safe water depth under the current motion state according to the dynamic information, and when the maximum water accumulation depth is less than the numerical interval between the limit water level height and the safe water depth, the current state of the vehicle is maintained. When the maximum water accumulation depth is greater than or equal to the numerical interval between the limit water level height and the safe water depth, the warning device is activated to issue a wading warning.

[0020] The present invention also provides a water risk monitoring system for an autonomous driving vehicle, comprising:

[0021] The data acquisition module is connected to the vehicle central gateway through the sub-gateway, and is used to obtain the image data of the vehicle's front-view camera, the point cloud data of the automotive-grade laser radar, and the current vehicle information; the vehicle information includes static information and dynamic information;

[0022] A waterlogging area delineation module is used to perform preprocessing, target recognition and area segmentation on the image data in sequence to determine the waterlogging area;

[0023] A water accumulation depth calculation module is used to compare the point cloud data with the water accumulation area, filter out the point cloud data of the water accumulation area, and calculate the maximum water accumulation depth;

[0024] The water wading warning module is used to determine the extreme water level height in a stationary state and the safe water depth in the current moving state according to the vehicle information, and activate the warning device for water wading warning when the maximum water accumulation depth is greater than or equal to the numerical range of the extreme water level height and the safe water depth.

[0025] The present invention also provides an electronic device, including a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above-mentioned method for monitoring water wading risks of autonomous driving vehicles.

[0026] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method for monitoring water wading risks of an autonomous driving vehicle as described above.

[0027] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0028] The present invention discloses a method, system, device and medium for monitoring the risk of wading of an autonomous driving vehicle, the method comprising obtaining image data of a vehicle's front-view camera, point cloud data of an automotive-grade laser radar and current vehicle information; preprocessing, target identification and region segmentation of the image data in sequence to determine the waterlogged area; performing comparative calculations based on the point cloud data and the waterlogged area to obtain the maximum waterlogged depth; determining the extreme water level in the current static state and the safe water depth in the current moving state respectively based on the vehicle information, and activating an alarm to sound a wading alarm when the maximum waterlogged depth is greater than or equal to the numerical range of the extreme water level and the safe water depth. The present invention helps vehicles to accurately identify the depth of waterlogging on roads, and to judge the risk of wading in combination with the vehicle's own conditions, and is suitable for measurements under complex weather and road conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0030] Figure 1 It is a flow chart of the method for monitoring water wading risk of an autonomous driving vehicle of the present invention;

[0031] Figure 2 This is a schematic diagram of the water depth measurement principle based on laser radar in this embodiment;

[0032] Figure 3 2 is a diagram of the system hardware structure in this embodiment. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] The purpose of the present invention is to provide a method, system, equipment and medium for monitoring the water wading risk of an autonomous driving vehicle, which can improve the vehicle's accuracy in determining the depth of water on the road and the water wading risk.

[0035] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] like Figure 1-Figure 3 As shown, the present invention provides a method for monitoring water risk of an autonomous driving vehicle, comprising:

[0037] Step 100: Acquire image data from the vehicle's front-view camera, point cloud data from an automotive-grade laser radar, and current vehicle information; the vehicle information includes static information and dynamic information.

[0038] Step 200: Preprocessing, target recognition and region segmentation are performed on the image data in sequence to determine the waterlogged area.

[0039] Step 300: Compare the point cloud data with the waterlogged area, filter out the point cloud data of the waterlogged area, and calculate the maximum waterlogged depth.

[0040] Step 400: Determine the extreme water level in a stationary state and the safe water depth in a current moving state according to the vehicle information, and activate the warning device for wading warning when the maximum water depth is greater than or equal to the numerical range of the extreme water level and the safe water depth.

[0041] Based on the above solution, the detailed processing procedures of each step are provided as shown below.

[0042] As a specific implementation of step 200, specifically:

[0043] A.1 Image preprocessing. Preprocessing includes image compression, grayscale, noise filtering, and image enhancement. Image compression and grayscale can reduce the computing power required for subsequent target recognition. Noise filtering uses the Retinex algorithm to remove the interference of raindrop noise on the image. Image enhancement uses an impact filter for sharpening, highlighting the boundaries of objects on the image and enhancing the processing effect. The system generates image data that meets the processing requirements of the target detection model through image preprocessing.

[0044] A.2 Road water accumulation target recognition. The system uses the yolov8 target recognition model to perform target recognition on the preprocessed image to determine whether there is water accumulation on the road surface. If there is water accumulation, the water accumulation image is saved as a water accumulation road surface image and the next step is performed; if there is no water accumulation, return to step A.1 to preprocess the new image data.

[0045] A.3 Semantic segmentation of waterlogged road areas. The waterlogged road image obtained in A.2 is used as input, and the area segmentation module segments the image data to delineate the waterlogged area. By presetting a threshold, when the image confidence of the edge contour pixel data is not less than the threshold, the edge contour pixel data is judged to be qualified, otherwise the image confidence is re-compared; finally, the waterlogged area is delineated from the image.

[0046] As a specific implementation of step 300, specifically:

[0047] B.1 Extraction of point cloud data of waterlogged area: The system extracts the laser radar scanning point cloud data as raw data, and obtains the point cloud data of the waterlogged area by comparing it with the image of the demarcated waterlogged area obtained in step A.3. The point cloud data of the waterlogged area obtained includes the coordinates of the laser point, color, and reflection intensity, etc., and then calculates the distance L1 between the laser radar and the water surface and the distance L2 between the laser radar and the road surface.

[0048] B.2 Obtaining radar pitch angle information: The system obtains the angle information between the laser radar transmitting antenna and the horizontal plane, namely the pitch angle α, by analyzing the raw data.

[0049] B.3 Calculate the water depth information: The system calculates the water depth information H based on the distance information L1, L2 obtained in B.1 and the angle information α obtained in B.2; the maximum water depth value H is selected from the water depth data set of the waterlogged area. max .

[0050] As a specific implementation of step 400, specifically:

[0051] C.1 Extract vehicle information: Vehicle information includes static information and dynamic information. The static information is built into the system to measure the maximum water level height h that ensures vehicle safety in a stationary state. limit-s The dynamic information is obtained by connecting the system to the domain controller, including the vehicle speed, ascent and descent slope, etc. The system calculates the safe water depth h under the motion state through the SPH (Smoothed Particle Hydrodynamics) method. limit-d .

[0052] C.2 Determine the risk of water wading: The system compares the maximum water depth H obtained in step B.3 max The safe water depth [h limit-s ,h limit-d ], determine the risk of wading in the current state (stationary or moving). max< [h limit-s ,h limit-d ], the vehicle continues to maintain the current state and restarts step A. Otherwise, go to C.3.

[0053] C.3 System warning: When the system determines that the vehicle is at risk of wading, the alarm is activated to sound an alarm.

[0054] Therefore, this embodiment has the following beneficial effects:

[0055] (1) The system can more accurately predict vehicle wading conditions under complex environmental conditions;

[0056] (2) The components used in the system design are all self-driving car configuration equipment, which has extremely high promotion value and does not require the installation of other equipment, saving costs;

[0057] (3) In complex road environments, the system has the ability to detect and discover potholes on the road, which has great value for redevelopment.

[0058] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0059] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only used to help understand the core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for monitoring water risk of an autonomous driving vehicle, characterized in that: include: Obtain image data from the vehicle's front-view camera, point cloud data from the automotive-grade LiDAR, and current vehicle information; The vehicle information includes static information and dynamic information; Preprocessing, target recognition and region segmentation are sequentially performed on the image data to determine the waterlogging area; By comparing the point cloud data with the waterlogged area, the point cloud data of the waterlogged area is screened out, and the maximum waterlogged depth is calculated; The extreme water level height in a stationary state and the safe water depth in a current moving state are determined respectively according to the vehicle information, and when the maximum water accumulation depth is greater than or equal to the numerical range of the extreme water level height and the safe water depth, the warning device is activated to issue a wading warning.

2. The method for monitoring water wading risk of an autonomous driving vehicle according to claim 1, characterized in that: The static information is built into the system to measure the extreme water level height to ensure vehicle safety in a stationary state; the dynamic information is obtained by connecting the system with the domain controller, including the vehicle's driving speed, ascent and descent slopes.

3. The method for monitoring water wading risk of an autonomous driving vehicle according to claim 1, characterized in that: The image data is sequentially preprocessed, target identified and region segmented to determine the waterlogged area, specifically including: The image data is compressed, grayed, noise filtered, and enhanced in sequence to obtain a preprocessed image; Performing target recognition on the preprocessed image, when the recognition result is that there is water accumulation, the corresponding image is saved as a waterlogged road surface image; The flooded road surface image is input into the area segmentation module, and when the image confidence of the edge contour pixel data is not less than a preset threshold, the edge contour pixel data is judged to be qualified, and the flooded area is delineated based on all qualified data.

4. The method for monitoring water wading risk of an autonomous driving vehicle according to claim 1, characterized in that: According to the comparison between the point cloud data and the waterlogged area, the point cloud data of the waterlogged area is screened out, and the maximum waterlogged depth is calculated, specifically including: The point cloud data obtained by the automotive-grade laser radar is compared with the image of the waterlogged area, the point cloud data of the waterlogged area is screened, and the distance between the laser radar and the water surface and the road surface is calculated using the point cloud data; the point cloud data of the waterlogged area includes the coordinates, color and reflection intensity of the laser point; Calculation is performed based on the point cloud data of the waterlogged area and the radar elevation angle information to obtain a depth information set of the waterlogged area, and the maximum waterlogged depth is screened out from the depth information set of the waterlogged area. The radar elevation angle information is the angle information between the laser radar transmitting antenna and the horizontal plane.

5. The method for monitoring water wading risk of an autonomous driving vehicle according to claim 1, characterized in that: The judgment process of the water wading alarm is as follows: The smooth particle method is used to calculate the safe water depth under the current motion state according to the dynamic information, and when the maximum water accumulation depth is less than the numerical interval between the limit water level height and the safe water depth, the current state of the vehicle is maintained. When the maximum water accumulation depth is greater than or equal to the numerical interval between the limit water level height and the safe water depth, the warning device is activated to issue a wading warning.

6. A water risk monitoring system for an autonomous driving vehicle, characterized in that: include: The data acquisition module is connected to the vehicle central gateway through the sub-gateway, and is used to obtain the image data of the vehicle's front-view camera, the point cloud data of the automotive-grade laser radar, and the current vehicle information; the vehicle information includes static information and dynamic information; A waterlogging area delineation module is used to perform preprocessing, target recognition and area segmentation on the image data in sequence to determine the waterlogging area; A water accumulation depth calculation module is used to compare the point cloud data with the water accumulation area, filter out the point cloud data of the water accumulation area, and calculate the maximum water accumulation depth; The water wading warning module is used to determine the extreme water level height in a stationary state and the safe water depth in the current moving state according to the vehicle information, and activate the warning device for water wading warning when the maximum water accumulation depth is greater than or equal to the numerical range of the extreme water level height and the safe water depth.

7. An electronic device, characterized in that: It includes a memory and a processor, the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the water wading risk monitoring method for an autonomous driving vehicle according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that: It stores a computer program, which, when executed by a processor, implements the method for monitoring water wading risks of an autonomous driving vehicle as described in any one of claims 1-5.

Images