Road surface water information prediction method, device and equipment

Abstract

The application provides a road water accumulation information prediction method, device and equipment. First, image data or point cloud data of a target vehicle in front of an autonomous vehicle in a water accumulation area is acquired. Then, information of water splashes generated by the target vehicle driving through the water accumulation area is determined according to the image data or the point cloud data. Then, a predicted water accumulation depth of the water accumulation area is determined according to the information of the water splashes generated by the target vehicle driving through the water accumulation area. Finally, a to-be-traveled route of the autonomous vehicle is determined according to the predicted water accumulation depth. In this way, the predicted water accumulation depth of the water accumulation area can be determined according to the information of the water splashes generated by the target vehicle driving through the water accumulation area, so that whether the autonomous driving function is applicable to the initial to-be-traveled route of the autonomous vehicle can be determined in time according to the predicted water accumulation depth, and the normal use of the autonomous driving function is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a method, device and equipment for predicting road surface water information. BACKGROUND

[0002] In recent years, with the continuous development and improvement of intelligent driving technology, autonomous vehicles have received widespread attention. The Society of Automotive Engineers (SAE) standard divides intelligent driving functions into six levels, L0-L5. Among them, the intelligent driving function of the L4 and below level autonomous vehicle needs to travel according to the set Operational Design Domain (ODD).

[0003] The operational design domain has certain requirements for the road surface, for example, there cannot be a large water accumulation area in the road area to be traveled. At present, due to the lack of road surface water area information in the shared high-precision map, the autonomous vehicle cannot timely identify the water accumulation area on the to-be-traveled route, thereby making the autonomous vehicle unable to judge whether the autonomous driving function is applicable to the above to-be-traveled route, limiting the use of the autonomous driving function. SUMMARY

[0004] The present application provides a method, device and equipment for predicting road surface water information to solve the technical problem that the existing technology cannot timely identify the water accumulation area on the to-be-traveled route.

[0005] In a first aspect, the present application provides a method for predicting road surface water information, applied to an autonomous vehicle, the method comprising:

[0006] obtaining image data or point cloud data of a target vehicle in front of the autonomous vehicle in a water accumulation area;

[0007] determining information of water splashes generated by the target vehicle driving through the water accumulation area according to the image data or the point cloud data;

[0008] determining a predicted water depth of the water accumulation area according to the information of water splashes generated by the target vehicle driving through the water accumulation area;

[0009] determining a to-be-traveled route of the autonomous vehicle according to the predicted water depth.

[0010] In an optional implementation, the determination of the information of water splashes generated by the target vehicle driving through the water accumulation area comprises:

[0011] input the image data or the point cloud data into a trained neural network model, and obtain information of the water splash output by the trained neural network model, the neural network model being generated according to a sample training set;

[0012] The sample training set includes historical image data and information of a measured water splash corresponding to the historical image data, or the sample training set includes historical point cloud data and information of a measured water splash corresponding to the historical point cloud data.

[0013] In an optional implementation, the determining of the predicted water depth of the water accumulation area comprises:

[0014] According to the information of the target vehicle, the information of the water splash generated by the target vehicle driving through the water accumulation area, and a mapping relationship between the information of the water splash generated by different vehicles driving through the water accumulation area and the water depth of the water accumulation area, the predicted water depth of the water accumulation area is determined.

[0015] In an optional implementation, the information of the target vehicle comprises at least one of the following: type information, speed information, and direction information.

[0016] In an optional implementation, the information of the water splash generated by the target vehicle driving through the water accumulation area comprises at least one of the following: height information of the water splash, width information of the water splash, and duration information of the water splash.

[0017] In an optional implementation, the determining of the to-be-traveled route of the autonomous vehicle comprises:

[0018] According to the image data and / or the point cloud data of the water accumulation area, a water surface texture feature of the water accumulation area is extracted;

[0019] According to the water surface texture feature, it is determined whether a to-be-avoided area exists in the water accumulation area, the to-be-avoided area being an area in which the water surface texture feature meets a preset texture feature;

[0020] If the to-be-avoided area exists in the water accumulation area, an initial to-be-traveled route of the autonomous vehicle is adjusted to a first to-be-traveled route, and the first to-be-traveled route does not include the to-be-avoided area.

[0021] In an optional implementation, if the to-be-avoided area does not exist in the water accumulation area, the method further comprises:

[0022] According to the image data and / or the point cloud data of the water accumulation area, a water accumulation contour of the water accumulation area is determined;

[0023] if the predicted water depth is greater than a preset water depth threshold and a width of the water profile is greater than a preset width threshold, the initial to-be-traveled route is adjusted to a second to-be-traveled route, and the second to-be-traveled route does not include the water area.

[0024] In an optional implementation, the determining the to-be-traveled route of the autonomous vehicle includes:

[0025] if the predicted water depth is less than or equal to a preset water depth threshold and / or the width of the water profile is less than or equal to a preset width threshold, the to-be-traveled route of the autonomous vehicle is determined as an initial to-be-traveled route.

[0026] In a second aspect, the present application provides a device for predicting road water information, applied to an autonomous vehicle, and the device includes:

[0027] a obtaining module, configured to obtain image data or point cloud data of a target vehicle in a water area in front of the autonomous vehicle;

[0028] a determining module, configured to determine, according to the image data or the point cloud data, information of water splashes generated by the target vehicle driving through the water area, determine a predicted water depth of the water area according to the information of the water splashes generated by the target vehicle driving through the water area, and determine a to-be-traveled route of the autonomous vehicle according to the predicted water depth.

[0029] In an optional implementation, the determining module is specifically configured to input the image data or the point cloud data into a trained neural network model, and obtain the information of the water splashes output by the trained neural network model, the neural network model being generated according to a sample training set; wherein the sample training set includes historical image data and information of actually measured water splashes corresponding to the historical image data, or the sample training set includes historical point cloud data and information of actually measured water splashes corresponding to the historical point cloud data.

[0030] In an optional implementation, the determining module is specifically configured to determine the predicted water depth of the water area according to information of the target vehicle, the information of the water splashes generated by the target vehicle driving through the water area, and a mapping relationship between information of water splashes generated by different vehicles driving through a water area and a water depth of the water area.

[0031] In an optional implementation, the information of the target vehicle includes at least one of the following: type information, speed information, and direction information.

[0032] In an optional implementation, the information about the water splashes generated by the target vehicle driving through the water accumulation area includes at least one of the following: height information of the water splashes, width information of the water splashes, and duration information of the water splashes.

[0033] In an optional implementation, the determining module is specifically configured to: extract water surface texture features of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area; determine whether a region to be avoided exists in the water accumulation area according to the water surface texture features, the region to be avoided being a region in which the water surface texture features meet preset texture features; and if the region to be avoided exists in the water accumulation area, adjust the initial to-be-traveled route of the autonomous vehicle to a first to-be-traveled route, the first to-be-traveled route not including the region to be avoided.

[0034] In an optional implementation, if the region to be avoided does not exist in the water accumulation area, the determining module is further configured to: determine a water accumulation contour of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area; and if the predicted water accumulation depth is greater than a preset water accumulation depth threshold and a width of the water accumulation contour is greater than a preset width threshold, adjust the initial to-be-traveled route to a second to-be-traveled route, the second to-be-traveled route not including the water accumulation area.

[0035] In an optional implementation, the determining module is specifically configured to: if the predicted water accumulation depth is less than or equal to a preset water accumulation depth threshold and / or the width of the water accumulation contour is less than or equal to a preset width threshold, determine that the to-be-traveled route of the autonomous vehicle is the initial to-be-traveled route.

[0036] In a third aspect, the present application also provides a computer program product, comprising a computer program which, when executed by a processor, implements the method of any one of the first aspect.

[0037] In a fourth aspect, the present application also provides a computer storage medium, which stores a plurality of instructions, the instructions being adapted to be loaded and executed by a processor to implement the method of any one of the first aspect.

[0038] In a fifth aspect, the present application also provides an electronic device, comprising: a processor and a memory; wherein the memory stores a computer program, the computer program being adapted to be loaded and executed by the processor to implement the method of any one of the first aspect.

[0039] The application provides a road water information prediction method, device and equipment. First, image data or point cloud data of a target vehicle in front of an autonomous vehicle in a water accumulation area is acquired. Then, information of water splashes generated by the target vehicle driving through the water accumulation area is determined according to the image data or the point cloud data. Then, a predicted water depth of the water accumulation area is determined according to the information of the water splashes generated by the target vehicle driving through the water accumulation area. Finally, a to-be-traveled route of the autonomous vehicle is determined according to the predicted water depth. In this way, the predicted water depth of the water accumulation area can be determined according to the information of the water splashes generated by the target vehicle driving through the water accumulation area, so that whether the autonomous driving function is applicable to the initial to-be-traveled route of the autonomous vehicle can be determined in time according to the predicted water depth, and the normal use of the autonomous driving function is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0041] Figure 1 A system architecture diagram of a road water information prediction system provided by an embodiment of the present application;

[0042] Figure 2 A flowchart of a road water information prediction method provided by an embodiment of the present application;

[0043] Figure 3 A flowchart of another road water information prediction method provided by an embodiment of the present application;

[0044] Figure 4 A flowchart of still another road water information prediction method provided by an embodiment of the present application;

[0045] Figure 5 A structural diagram of a road water information prediction device provided by an embodiment of the present application;

[0046] Figure 6 A structural diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0047] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0048] In recent years, with the continuous development and improvement of intelligent driving technology, autonomous vehicles have received widespread attention. The Society of Automotive Engineers (SAE) standard divides intelligent driving functions into six levels, L0-L5. Among them, the intelligent driving function of the autonomous vehicle at L4 and below needs to travel according to the set Operational Design Domain (ODD).

[0049] The operational design domain has certain requirements for the road surface, for example, there cannot be a large water accumulation area in the road area to be traveled. At present, due to the lack of road water accumulation area information in the shared high-precision map, the autonomous vehicle cannot timely identify the water accumulation area on the to-be-traveled route, thereby making the autonomous vehicle unable to judge whether the autonomous driving function is applicable to the above to-be-traveled route, and limiting the use of the autonomous driving function.

[0050] To solve the above technical problems, the embodiments of the present application provide a prediction method, device and equipment for road water accumulation information, which determines the predicted water accumulation depth of the water accumulation area according to the information of the water splash generated by the target vehicle driving through the water accumulation area, so as to timely judge whether the autonomous driving function is applicable to the initial to-be-traveled route of the autonomous vehicle according to the predicted water accumulation depth, and further ensure the normal use of the autonomous driving function.

[0051] The system architecture of a prediction system for road water accumulation information related to the present application will be described below.

[0052] Figure 1 A system architecture diagram of a prediction system for road water accumulation information provided by the embodiments of the present application is shown in FIG. 1. As shown in FIG. 1, the system architecture includes a sensor component 101 and a controller 102. Figure 1

[0053] ​The sensor assembly 101 and the controller 102 are connected. The sensor assembly 101 is used to acquire image data or point cloud data of a target vehicle in front of the autonomous vehicle in the water accumulation area, and input to the controller 102. The controller 102 is used to determine the predicted water accumulation depth of the water accumulation area according to the image data or the point cloud data; and determine the to-be-traveled route of the autonomous vehicle according to the predicted water accumulation depth of the water accumulation area.

[0054] The above sensor assembly can include but is not limited to a camera sensor, a laser radar sensor, etc. The above controller can be a single controller, or a controller group composed of multiple controllers. For example, the above controller can include an autonomous driving controller and a whole vehicle motion controller, etc., and the embodiments of the present application do not limit this.

[0055] It should be understood that the system architecture of the water accumulation information prediction system in the technical solution of the present application can be the system architecture in Figure 1 , but is not limited thereto, and can also be other types of system architectures.

[0056] It can be understood that the water accumulation information prediction method of the technical solution of the present application can be implemented by the water accumulation information prediction device provided by the embodiments of the present application. The water accumulation information prediction device can be part or all of a certain device, such as a controller.

[0057] The following describes the technical solution of the embodiments of the present application in detail with specific embodiments taking a controller integrated or installed with relevant execution code as an example. The following specific embodiments can be combined with each other, and the same or similar concepts or processes can not be described in detail in some embodiments.

[0058] Figure 2 A flowchart of a water accumulation information prediction method provided by the embodiments of the present application is shown in the figure. The present embodiment relates to the process of predicting water accumulation information. As shown in Figure 2 , the method comprises:

[0059] S201, acquiring image data or point cloud data of a target vehicle in front of the autonomous vehicle in the water accumulation area.

[0060] In the embodiments of the present application, the controller can first acquire image data or point cloud data of the target vehicle in the water accumulation area, and then determine the to-be-traveled route of the autonomous vehicle according to the acquired image data or point cloud data.

[0061] It can be understood that when the autonomous vehicle departs from the starting point A to the destination B, there can be at least one travel route. If the current position of the autonomous vehicle in the travel process is denoted as position C, the travel route from the starting point A to the position C is the traveled route, and the travel route from the position C to the destination B is the to-be-traveled route.

[0062] The embodiment of the present application does not limit the position of the water accumulation area. In some embodiments, the water accumulation area can be a water accumulation area located on the initial to-be-traveled route of the autonomous vehicle.

[0063] The target vehicle can be any vehicle, and the embodiment of the present application does not limit this. In some embodiments, the target vehicle can be a vehicle located in front of the autonomous vehicle, so that the information of the water accumulation area on the road in front of the autonomous vehicle can be obtained according to these vehicles. In other embodiments, the distance between the target vehicle and the autonomous vehicle needs to be greater than a preset distance threshold, so that the controller can obtain the information of the water accumulation area in advance and determine whether to adjust the initial to-be-traveled route of the autonomous vehicle. Exemplarily, the preset distance threshold can be 50 meters.

[0064] In some embodiments, the controller can obtain the image data through the camera sensor and obtain the point cloud data through the laser radar sensor. In other embodiments, the controller can also obtain the data or information generated when the target vehicle passes through the water accumulation area through other sensors, and the embodiment of the present application does not limit this.

[0065] S202, determining the information of the water splash generated when the target vehicle passes through the water accumulation area according to the image data or the point cloud data.

[0066] In this step, after obtaining the image data or the point cloud data, the controller can determine the information of the water splash generated when the target vehicle passes through the water accumulation area according to the image data or the point cloud data.

[0067] The information of the water splash generated when the target vehicle passes through the water accumulation area can include the height information of the water splash, the width information of the water splash, the duration of the water splash, etc.

[0068] It can be understood that the sensor on the autonomous vehicle can continuously track the target vehicle in front of the vehicle entering the water accumulation area. The controller can comprehensively calculate the water depth of the water accumulation area according to the information of the water splash generated when vehicles of different types and speeds pass through the water accumulation area with different water depths, and the height information of the water surface line reaching the position of the target vehicle.

[0069] The embodiment of the present application does not limit how to determine the information of the water splash generated when the target vehicle passes through the water accumulation area. In some embodiments, the controller can input the image data or the point cloud data of the target vehicle in the water accumulation area into the trained neural network model, and obtain the information of the water splash output by the trained neural network model.

[0070] The neural network model can be based on a convolutional neural network algorithm and obtained after being trained by a large amount of sample training set data to continuously improve the prediction accuracy. The sample training set can include historical image data and information of measured water splashes corresponding to the historical image data, or the sample training set can include historical point cloud data and information of measured water splashes corresponding to the historical point cloud data, and the embodiments of the present application are not limited thereto.

[0071] S203, determining the predicted water depth of the water accumulation area according to the information of the water splashes generated by the target vehicle driving through the water accumulation area.

[0072] In this step, after determining the information of the water splashes generated by the target vehicle driving through the water accumulation area, the controller can determine the predicted water depth of the water accumulation area according to the information of the water splashes generated by the target vehicle driving through the water accumulation area.

[0073] The embodiments of the present application do not limit how to determine the predicted water depth of the water accumulation area. In some embodiments, the controller can determine the predicted water depth of the water accumulation area according to the information of the target vehicle, the information of the water splashes generated by the target vehicle driving through the water accumulation area, and a mapping relationship between the information of the water splashes generated by different vehicles driving through the water accumulation area and the water depth of the water accumulation area.

[0074] The information of the target vehicle can include vehicle type, vehicle speed, vehicle acceleration, and orientation information. In actual working conditions, the controller can accurately obtain the information of the target vehicle through sensors such as cameras and laser radars.

[0075] The present application does not limit how to obtain the above mapping relationship. In some embodiments, the information of the water splashes (such as width, height from the water surface, duration, etc.) splashed by different types of vehicles (such as trucks, buses, cars, tricycles, etc.) driving through a plurality of water accumulation areas at different speeds (such as 5, 10, 15…100 km / h, etc.) can be obtained in advance, as well as the measured water depth of these water accumulation areas, to obtain a curve atlas reflecting the mapping relationship between the information of the water splashes generated by different vehicles driving through the water accumulation area and the water depth. It should be noted that when obtaining the above mapping relationship, the speed gradient of the vehicle can be set to 5 km / h, and can be further refined to 3 km / h, etc., and the embodiments of the present application are not limited thereto. Further, after obtaining the information of the water splashes generated by the target vehicle driving through the water accumulation area, the controller can combine the information of the target vehicle to find and output the predicted water depth of the water accumulation area from the above curve atlas by using the atlas interpolation method.

[0076] In some embodiments, before determining the to-be-traveled route of the autonomous vehicle, the controller can further extract water surface texture features of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area, and then determine whether there is a to-be-avoided area in the water accumulation area according to the extracted water surface texture features. The to-be-avoided area is an area with water surface texture features meeting preset texture features. For example, the visual image of the water accumulation area obtained by the camera sensor can reflect the water surface texture features of the water accumulation area. If there is vortex texture in the water surface texture features, it can be determined that there is a to-be-avoided area such as a drainage well in the water accumulation area. At this time, the controller should optimize the to-be-traveled route in advance to control the vehicle to avoid the to-be-avoided area as much as possible. In other embodiments, the controller can also determine the water accumulation contour of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area.

[0077] It can be understood that the controller can obtain the above-mentioned to-be-avoided area and / or water accumulation contour through a trained neural network model, or through other methods, which are not limited in the embodiments of the present application. For example, the reflection energy of the laser radar sensor is different for water accumulation and non-water accumulation areas, the visual image detected by the camera sensor can reflect the water surface texture features of the water accumulation area, and by training a deep learning neural network with a large amount of data, the area, edge contour, to-be-avoided area and other information of the water accumulation area can be regressed, classified and output. Further, the controller can also obtain the material information of the road surface through a neural network model, image data and / or point cloud data of the road surface. The material of the road surface can include cement, concrete, asphalt, gravel, water accumulation, snow and the like.

[0078] S204, determining the to-be-traveled route of the autonomous vehicle according to the predicted water accumulation depth.

[0079] In this step, after determining the predicted water accumulation depth of the water accumulation area, the controller can determine the to-be-traveled route of the autonomous vehicle according to the predicted water accumulation depth.

[0080] The embodiments of the present application do not limit how to determine the to-be-traveled route of the autonomous vehicle. In some embodiments, if there is a to-be-avoided area in the water accumulation area, the controller adjusts the initial to-be-traveled route of the autonomous vehicle to a first to-be-traveled route, and the first to-be-traveled route does not include the to-be-avoided area.

[0081] In some embodiments, if the water accumulation region does not exist the region to be avoided, and the predicted water accumulation depth is less than or equal to a preset water accumulation depth threshold and / or the width of the water accumulation profile is less than or equal to a preset width threshold, the controller determines the initial route to be traveled by the autonomous vehicle as the initial route to be traveled. The preset water accumulation depth threshold can be determined according to the requirements of the operational design domain (ODD) for water accumulation depth, and the preset width threshold can be determined according to the distance between the two front wheels or the two rear wheels of the autonomous vehicle, which is not limited in the embodiments of the present application. In some other embodiments, if the water accumulation region does not exist the region to be avoided, but the predicted water accumulation depth is greater than the preset water accumulation depth threshold and the width of the water accumulation profile is greater than the preset width threshold, the controller adjusts the initial route to be traveled to a second route to be traveled, and the second route to be traveled does not include the water accumulation region.

[0082] For example, when the predicted water accumulation depth is greater than the preset water accumulation depth threshold, if the width of the water accumulation profile is less than or equal to the distance between the two front wheels of the autonomous vehicle, the autonomous vehicle can directly pass through the water accumulation region; if the width of the water accumulation profile is greater than the distance between the two front wheels of the autonomous vehicle, the controller can detect whether there is a drivable region in the adjacent lane, and plan to change lanes in advance to avoid the water accumulation region; if there is no drivable region in the adjacent lane, the controller can control the autonomous vehicle to slow down in advance and issue a prompt message to warn the passengers in the vehicle that there may be a danger in the water accumulation region ahead.

[0083] In some other embodiments, when the autonomous vehicle travels through the water accumulation region according to the initial route to be traveled or the first route to be traveled, the controller can also obtain the measured water accumulation depth of the water accumulation region. For example, when the autonomous vehicle approaches the water accumulation region or has entered the water accumulation region, the controller can accurately obtain the measured water accumulation depth of the water accumulation region through the wading sensor assembly. The wading sensor assembly can include a millimeter wave radar sensor, an ultrasonic radar sensor, a sonar detector, etc. Further, the controller can determine a correction coefficient of the water accumulation depth according to the measured water accumulation depth and the predicted water accumulation depth of the water accumulation region, correct the subsequent predicted water accumulation depth according to the correction coefficient, and then determine the route to be traveled by the autonomous vehicle according to the corrected predicted water accumulation depth, and optimize the vehicle kinematics control model (such as longitudinal control, lateral control, etc.).

[0084] In some other embodiments, the controller can also pack and upload the water accumulation profile, the measured water accumulation depth, the high-precision positioning information, etc. of the water accumulation region to the cloud database server. The cloud database server can update the water accumulation information of the water accumulation region in the high-definition map, and automatically distribute the updated high-definition map to the vehicles traveling on the road segment where the water accumulation region is located. The controllers of these vehicles directly read the water accumulation information of the water accumulation region from the high-definition map, without the need for the vehicle to make predictions, thereby improving the accuracy, efficiency and speed of the vehicle in obtaining the water accumulation information of the road.

[0085] The high-precision positioning information can include a Global Positioning System (GPS), a Real-time kinematic (RTK) carrier phase differential technology, or the like. The controller can upload the packaged information to a cloud server through a vehicle communication module (Telematics BOX, T-BOX). It can be understood that if multiple vehicles simultaneously upload the packaged information to the cloud server, the road water information of the road section can be quickly and efficiently created.

[0086] It should be noted that the method for predicting road water information provided in the embodiments of the present application is not only applicable to autonomous vehicles, but also applicable to vehicles without or without an autonomous driving function, so that these vehicles can timely obtain road water information and reflect the road water information to the driver.

[0087] It should be noted that in the prior art, the information of the water accumulation area is measured by using the vehicle-mounted sensor when the vehicle approaches or enters the water accumulation area, but the area, contour, water depth, and the like of the water accumulation area cannot be obtained in advance. Therefore, the prior art has certain limitations and cannot be applied to the path planning and optimization process of the autonomous driving function. For example, in the driving scene of the road in the rain or after the rain, the autonomous vehicle with the intelligent driving function may mistakenly enter the deep water accumulation area or be trapped in the manhole cover drainage area, etc., which affects the safety of the vehicle and the passengers. In the embodiments of the present application, the contour and water depth of the water accumulation area can be predicted by using the trained neural network model, and the to-be-traveled route of the autonomous vehicle can be optimized in advance according to the predicted information, thereby effectively ensuring the safety of the autonomous vehicle and the passengers.

[0088] The method for predicting road water information provided in the present application first obtains image data or point cloud data of a target vehicle in front of an autonomous vehicle in a water accumulation area; then, according to the image data or point cloud data, information of water splashes generated when the target vehicle drives through the water accumulation area is determined; then, according to the information of the water splashes generated when the target vehicle drives through the water accumulation area, a predicted water depth of the water accumulation area is determined; and finally, according to the predicted water depth, a to-be-traveled route of the autonomous vehicle is determined. In this way, since the predicted water depth of the water accumulation area can be determined according to the information of the water splashes generated when the target vehicle drives through the water accumulation area, it can be determined in a timely manner whether the autonomous driving function is applicable to the initial to-be-traveled route of the autonomous vehicle according to the predicted water depth, thereby ensuring the normal use of the autonomous driving function.

[0089] On the basis of the above embodiments, how to determine the information of the water splashes generated when the target vehicle drives through the water accumulation area will be described below. Figure 3Another flowchart of a method for predicting road water information provided by an embodiment of the present application is shown in FIG. 6, and the method comprises the following steps. Figure 3

[0090] S301, obtaining image data or point cloud data of a target vehicle in front of an autonomous vehicle in a water accumulation area.

[0091] S302, inputting the image data or point cloud data into a trained neural network model, and obtaining water splash information output by the trained neural network model.

[0092] S303, determining a predicted water depth of the water accumulation area according to the water splash information generated by the target vehicle driving through the water accumulation area.

[0093] S304, correcting the predicted water depth according to a correction coefficient of the water depth.

[0094] S305, determining a to-be-traveled route of the autonomous vehicle according to the corrected predicted water depth.

[0095] The technical terms, technical effects, technical features, and optional embodiments of S301-S305 can be understood with reference to S201-S204 shown in FIG. 5, and repeated descriptions are not given here. Figure 2

[0096] On the basis of the above embodiments, how to determine the to-be-traveled route of the autonomous vehicle is described below. Figure 4 Another flowchart of a method for predicting road water information provided by an embodiment of the present application is shown in FIG. 6, and the method comprises the following steps. Figure 4

[0097] S401, obtaining image data or point cloud data of a target vehicle in front of an autonomous vehicle in a water accumulation area.

[0098] S402, determining water splash information generated by the target vehicle driving through the water accumulation area according to the image data or point cloud data.

[0099] S403, determining a predicted water depth of the water accumulation area according to the water splash information generated by the target vehicle driving through the water accumulation area.

[0100] S404, determining whether there is a to-be-avoided area in the water accumulation area.

[0101] If yes, step S405 is performed; if no, step S406 is performed.

[0102] S405, adjusting an initial to-be-traveled route of the autonomous vehicle to a first to-be-traveled route.

[0103] ​​​S406, determine whether the predicted water depth is greater than a preset water depth threshold and the width of the water profile is greater than a preset width threshold.

[0104] If the predicted water depth is greater than the preset water depth threshold and the width of the water profile is greater than the preset width threshold, step S407 is performed; if the predicted water depth is less than or equal to the preset water depth threshold and / or the width of the water profile is less than or equal to the preset width threshold, step S408 is performed.

[0105] S407, adjust the initial to-be-traveled route to a second to-be-traveled route.

[0106] S408, determine that the to-be-traveled route of the autonomous vehicle is the initial to-be-traveled route.

[0107] The technical terms, technical effects, technical features, and optional embodiments of S401-S408 can be understood with reference to Figure 2 The repeated content is not repeated here.

[0108] The method for predicting road water information provided in the present application first acquires image data or point cloud data of a target vehicle in front of the autonomous vehicle in a water accumulation area; then, according to the image data or point cloud data, information of water splashes generated by the target vehicle driving through the water accumulation area is determined; then, according to the information of water splashes generated by the target vehicle driving through the water accumulation area, a predicted water depth of the water accumulation area is determined; finally, according to the predicted water depth, a to-be-traveled route of the autonomous vehicle is determined. In this way, since the predicted water depth of the water accumulation area can be determined according to the information of water splashes generated by the target vehicle driving through the water accumulation area, it can be determined in a timely manner whether the autonomous driving function is applicable to the initial to-be-traveled route of the autonomous vehicle according to the predicted water depth, thereby ensuring the normal use of the autonomous driving function.

[0109] Those skilled in the art can understand that all or part of the steps of the above-mentioned method embodiments can be completed by program instruction related hardware. The aforementioned program can be stored in a computer readable storage medium, and the program executes the steps of the above-mentioned method embodiments when executed; and the aforementioned storage medium includes ROM, RAM, magnetic disc or optical disc and various storage medium that can store program code.

[0110] Figure 5 A structure diagram of a road water information prediction device provided in an embodiment of the present application is shown. The road water information prediction device can be realized by software, hardware or a combination of the two, and can be, for example, a controller in the above-mentioned embodiments to execute the road water information prediction method in the above-mentioned embodiments. As Figure 5 shown, the road water information prediction device 500 includes:

[0111] The acquisition module 501 is configured to acquire image data or point cloud data of a target vehicle in front of the autonomous vehicle in a water accumulation area.

[0112] The determination module 502 is configured to determine, according to the image data or the point cloud data, information of water splashes generated by the target vehicle driving through the water accumulation area, determine a predicted water accumulation depth of the water accumulation area according to the information of the water splashes generated by the target vehicle driving through the water accumulation area, and determine a to-be-traveled route of the autonomous vehicle according to the predicted water accumulation depth.

[0113] In an optional implementation, the determination module 502 is specifically configured to input the image data or the point cloud data into a trained neural network model, and acquire information of water splashes output by the trained neural network model, the neural network model being generated according to a sample training set; and the sample training set includes historical image data and information of measured water splashes corresponding to the historical image data, or the sample training set includes historical point cloud data and information of measured water splashes corresponding to the historical point cloud data.

[0114] In an optional implementation, the determination module 502 is specifically configured to determine the predicted water accumulation depth of the water accumulation area according to the information of the target vehicle, the information of the water splashes generated by the target vehicle driving through the water accumulation area, and a mapping relationship between information of water splashes generated by different vehicles driving through the water accumulation area and water accumulation depths of the water accumulation area.

[0115] In an optional implementation, the information of the target vehicle includes at least one of the following: type information, speed information, and direction information.

[0116] In an optional implementation, the information of the water splashes generated by the target vehicle driving through the water accumulation area includes at least one of the following: height information of the water splashes, width information of the water splashes, and duration information of the water splashes.

[0117] In an optional implementation, the determination module 502 is specifically configured to extract a water surface texture feature of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area, determine whether a to-be-avoided area exists in the water accumulation area according to the water surface texture feature, the to-be-avoided area being an area in which the water surface texture feature meets a preset texture feature, and adjust an initial to-be-traveled route of the autonomous vehicle to a first to-be-traveled route if the to-be-avoided area exists in the water accumulation area, the first to-be-traveled route not including the to-be-avoided area.

[0118] In an optional implementation, if the water accumulation area does not exist the area to be avoided, the determining module 502 is further configured to determine a water accumulation contour of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area; and if the predicted water accumulation depth is greater than the preset water accumulation depth threshold and the width of the water accumulation contour is greater than the preset width threshold, the initial driving route is adjusted to a second driving route, and the second driving route does not include the water accumulation area.

[0119] In an optional implementation, the determining module 502 is specifically configured to determine that the driving route of the autonomous vehicle is the initial driving route if the predicted water accumulation depth is less than or equal to the preset water accumulation depth threshold and / or the width of the water accumulation contour is less than or equal to the preset width threshold.

[0120] It should be noted that, Figure 5 The prediction device for water accumulation information on a road provided in the embodiments can be used to execute the prediction method for water accumulation information on a road provided in any of the above embodiments, and the specific implementation and technical effects are similar, which will not be described here.

[0121] Figure 6 A structural schematic diagram of an electronic device is provided in the embodiments of the present application. As shown in the figure, Figure 6 The electronic device 600 can include at least one processor 601 and a memory 602. Figure 6 As shown, the electronic device is an example of an electronic device with one processor.

[0122] The memory 602 is used to store programs. Specifically, the programs can include program codes, and the program codes include computer operation instructions.

[0123] The memory 602 can include a high-speed RAM memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0124] The processor 601 is used to execute the computer execution instructions stored in the memory 602 to implement the prediction method for water accumulation information on a road described above; wherein the processor 601 can be a central processing unit (CPU), or a specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0125] Optionally, in a specific implementation, if the communication interface, the memory 602 and the processor 601 are implemented independently, the communication interface, the memory 602 and the processor 601 can be connected to each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc., but it does not mean that there is only one bus or one type of bus.

[0126] Optionally, in a specific implementation, if the communication interface, the memory 602 and the processor 601 are integrated on a chip, the communication interface, the memory 602 and the processor 601 can complete communication through an internal interface.

[0127] The embodiment of the application further provides a chip, including a processor and an interface. The interface is used for inputting and outputting data or instructions processed by the processor. The processor is used for executing the method provided in the above method embodiment. The chip can be applied to the prediction device of road water information.

[0128] The embodiment of the application further provides a program, which is used for executing the prediction method of road water information provided in the above method embodiment when executed by a processor.

[0129] The embodiment of the application further provides a program product, for example, a computer readable storage medium, in which instructions are stored, which, when executed on a computer, cause the computer to execute the prediction method of road water information provided in the above method embodiment.

[0130] The application further provides a computer readable storage medium, which can include: a U disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk and various media that can store program codes. Specifically, the computer readable storage medium stores program information, which is used for the prediction method of road water information.

[0131] In the above embodiments, the implementation can be wholly or partially realized by software, hardware, firmware or any combination thereof. When realized by software, the implementation can be wholly or partially realized in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the whole or part of the process or function according to the embodiments of the present application is produced. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions can be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center through wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.) manner. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. integrated with one or more available media sets. The available media can be a magnetic medium (for example, floppy disk, hard disk, magnetic tape), an optical medium (for example, DVD), or a semiconductor medium (for example, solid state disk (SSD)) and the like.

[0132] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method of predicting road surface water information, characterized by, The method is applied to road water information, and the method comprises: acquiring image data or point cloud data of a target vehicle in front of the road water information in a water accumulation area; determining water splash information generated when the target vehicle drives through the water accumulation area according to the image data or the point cloud data; determining a predicted water accumulation depth of the water accumulation area according to the water splash information generated when the target vehicle drives through the water accumulation area; determining a to-be-traveled route of the road water information according to the predicted water accumulation depth; the determination of the to-be-traveled route of the road water information comprises: extracting water surface texture features of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area; determining whether a to-be-avoided area exists in the water accumulation area according to the water surface texture features, the to-be-avoided area being an area in which the water surface texture features meet preset texture features; if the to-be-avoided area exists in the water accumulation area, adjusting an initial to-be-traveled route of the road water information to a first to-be-traveled route, the first to-be-traveled route not including the to-be-avoided area; if the to-be-avoided area does not exist in the water accumulation area, the method further comprises: determining a water accumulation contour of the water accumulation area according to the image data and / or the point cloud data of the water accumulation area; if the predicted water accumulation depth is greater than a preset water accumulation depth threshold and a width of the water accumulation contour is greater than a preset width threshold, adjusting the initial to-be-traveled route to a second to-be-traveled route, the second to-be-traveled route not including the water accumulation area; packing the water accumulation contour, a measured water accumulation depth, and high-precision positioning information of the water accumulation area and uploading them to a cloud database server, the cloud database server updating water accumulation information of the water accumulation area in a high-precision map and automatically distributing the updated high-precision map to vehicles driving on a road segment on which the water accumulation area is located.

2. The method of claim 1, wherein, the determination of the water splash information generated when the target vehicle drives through the water accumulation area comprises: inputting the image data or the point cloud data into a trained neural network model and acquiring the water splash information output by the trained neural network model, the neural network model being generated according to a sample training set; wherein the sample training set comprises historical image data and measured water splash information corresponding to the historical image data, or the sample training set comprises historical point cloud data and measured water splash information corresponding to the historical point cloud data.

3. The method of claim 2, wherein, the determination of the predicted water accumulation depth of the water accumulation area comprises: determining the predicted water accumulation depth of the water accumulation area according to information of the target vehicle, the water splash information generated when the target vehicle drives through the water accumulation area, and a mapping relationship between water splash information generated when different vehicles drive through a water accumulation area and a water accumulation depth of the water accumulation area.

4. The method of claim 3, wherein, The information of the target vehicle comprises at least one of the following: type information, speed information, and direction information.

5. The method according to any one of claims 1-4, characterized in that, The water splash information generated when the target vehicle drives through the water accumulation area comprises at least one of the following: height information of the water splash, width information of the water splash, and duration information of the water splash.

6. The method of claim 1, wherein, the determination of the to-be-traveled route of the road water information comprises: If the predicted water depth is less than or equal to a preset water depth threshold and / or the width of the water profile is less than or equal to a preset width threshold, the to-be-traveled route of the road water information is determined as an initial to-be-traveled route.

7. A device for predicting road surface water information, characterized by comprising: a road surface water information prediction unit that predicts road surface water information based on a road surface water information map and a vehicle speed. The device is applied to road water information, and the device comprises: An acquisition module is configured to acquire image data or point cloud data of a target vehicle in front of the road water information in a water area; A determination module is configured to determine information of water splashes generated when the target vehicle drives through the water area according to the image data or the point cloud data, determine a predicted water depth of the water area according to the information of the water splashes generated when the target vehicle drives through the water area, and determine a to-be-traveled route of the road water information according to the predicted water depth. The determination of the to-be-traveled route of the road water information comprises: extracting water surface texture features of the water area according to the image data and / or point cloud data of the water area; determining whether a to-be-avoided area exists in the water area according to the water surface texture features, the to-be-avoided area being an area in which the water surface texture features meet preset texture features; if the to-be-avoided area exists in the water area, adjusting an initial to-be-traveled route of the road water information to a first to-be-traveled route, the first to-be-traveled route not including the to-be-avoided area; if the to-be-avoided area does not exist in the water area, determining a water profile of the water area according to the image data and / or point cloud data of the water area; if the predicted water depth is greater than a preset water depth threshold and the width of the water profile is greater than a preset width threshold, adjusting the initial to-be-traveled route to a second to-be-traveled route, the second to-be-traveled route not including the water area; packing the water profile of the water area, a measured water depth, and high-precision positioning information and uploading them to a cloud database server, the cloud database server updating water information of the water area in a high-precision map and automatically distributing the updated high-precision map to vehicles driving on a road segment on which the water area is located.

8. An electronic device, comprising: The device comprises: a processor and a memory; the memory stores a computer program, the computer program is adapted to be loaded and executed by the processor, and the method in any one of claims 1 to 6 is executed.

Images