Window occlusion detection method and device

Abstract

The present application provides a method and device for window occlusion detection, which belongs to the field of sensor technology and can be used for assisted driving and autonomous driving. The method includes: obtaining a first window echo intensity and a second window echo intensity, the first window echo intensity represents the reference window echo intensity, and the second window echo intensity represents the window echo intensity at the current moment; obtaining the detection parameters of the target point at the first moment and the detection parameters at the second moment, the detection parameters including the reflectivity and / or position coordinates of the target point; according to the first window echo intensity, the second window echo intensity, the detection parameters at the first moment and the detection parameters at the second moment, obtaining a window occlusion parameter, the window occlusion parameter is used to indicate the window occlusion state. This method improves the accuracy of lidar window occlusion detection, so as to improve the terminal's advanced driving assistance system (ADAS) capabilities in autonomous driving or assisted driving.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of laser radar, and in particular to a window occlusion detection method and device. Background Art

[0002] Advanced driving assistance systems (ADAS) rely on a variety of sensors to perceive their surroundings. Light detection and ranging (LIDAR) is a common sensor in ADAS, as it can quickly and accurately acquire information about the surrounding environment. LIDAR systems are used for distance detection. LIDAR emits a laser at a target, and the detector receives the target's echo signal. By measuring the round-trip time of the transmitted signal, the distance to the target can be determined. The LIDAR system expands the distance measurement results of a single point into two dimensions through scanning or multi-element array detection, forming a distance image.

[0003] LIDAR is usually installed on the vehicle body and the internal components are protected by a window. When the LIDAR window is blocked by dirt, the emitted laser will be reflected, absorbed or even refracted when passing through the dirt, and the ranging performance of the LIDAR will be affected. Figure 1 As shown, Figure 1 This diagram shows an image of a LIDAR window obscured by water droplets. When water droplets appear on the LIDAR window, LIDAR imaging will produce noise. Therefore, maintaining a clean LIDAR window is crucial for advanced driver assistance systems.

[0004] Currently, detection of LIDAR window obstruction is primarily done manually or by adding additional detection devices, which increases the cost of LIDAR use and design. Therefore, how to detect LIDAR window obstruction without increasing the cost of LIDAR use and design is an urgent problem to be solved. Summary of the Invention

[0005] The present application provides a window occlusion detection method and device, which improves the accuracy of laser radar window occlusion detection, thereby improving the accuracy of obstruction warning and classification.

[0006] In a first aspect, a window occlusion detection method is provided, including: obtaining a first window echo intensity and a second window echo intensity, the first window echo intensity representing a reference window echo intensity, and the second window echo intensity representing a window echo intensity at a current moment; obtaining detection parameters of a target point at a first moment and detection parameters at a second moment, the detection parameters including the reflectivity and / or position coordinates of the target point; obtaining a window occlusion parameter based on the first window echo intensity, the second window echo intensity, the detection parameters at the first moment, and the detection parameters at the second moment, the window occlusion parameter being used to indicate a window occlusion state.

[0007] In the technical solution of the embodiment of the present application, the window occlusion parameters are preliminarily determined by the real-time collected second window echo intensity and the reference window echo intensity. On this basis, accurate window occlusion parameters are obtained through the detection parameters at the first moment and the detection parameters at the second moment, thereby achieving accurate detection of obstructions in the front window of the lidar, so as to improve the accuracy of obstruction warning and classification.

[0008] In combination with the first aspect, in certain implementations of the first aspect, obtaining the window occlusion parameter includes: obtaining the window occlusion parameter based on a first change parameter and a second change parameter; wherein the first change parameter includes the difference between the second window echo intensity and the first window echo intensity, or the ratio of the second window echo intensity to the first window echo intensity; the second change parameter represents the change between the reflectivity of the target point at the first moment and the reflectivity at the second moment, or the change between the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment, and the estimated value of the position coordinate at the first moment is obtained based on the position coordinate at the second moment.

[0009] In the technical solution of the embodiment of the present application, the first change parameter is used to determine whether the LIDAR window is blocked. Compared with determining whether the LIDAR window is blocked by the window echo intensity, its advantage is that the window echo intensity change parameter can more clearly reflect the difference in window echo intensity before and after the window is blocked through a multiple relationship or a difference relationship, and is not affected by the size of the window echo intensity of the unobstructed window itself.

[0010] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: outputting an alarm message, the first change parameter, or at least one of the window occlusion parameters, the alarm message being used to remind the window occlusion status, and the first change parameter corresponding to the occlusion classification and / or occlusion partition.

[0011] In combination with the first aspect, in certain implementations of the first aspect, the method includes: p elements in the first change parameter that are greater than a first threshold constitute a first element, where p is a positive integer; q elements in the first change parameter that are less than or equal to the first threshold constitute a second element, and the value of the second element is 0, where q is a positive integer; obtaining the window occlusion parameter includes: determining the window occlusion parameter according to a weighted sum of the values ​​of at least one fourth change parameter, the fourth change parameter including the second element, the third element, and the fourth element; verifying the first element according to the second change parameter to obtain the third element and the fourth element; the third The element represents n times of r elements in the first element, and the r elements represent elements in the second change parameter corresponding to the r elements that are less than or equal to the second threshold, or the r elements represent elements in the second change parameter corresponding to the r elements that are less than or equal to the third threshold, wherein r is a positive integer, and n is a positive real number less than 1 or 0; the fourth element represents s elements in the first element, and the s elements represent elements in the second change parameter corresponding to the s elements that are greater than the second threshold, or the s elements represent elements in the second change parameter corresponding to the s elements that are greater than the third threshold, wherein s is a positive integer.

[0012] In the technical solution of the embodiment of the present application, the window occlusion parameter Z obtained after correcting the first change parameter by the second change parameter has more accurate window occlusion information, eliminating the influence of the inflated window echo intensity caused by other factors such as temperature, power fluctuation of the laser transmitter, etc., thereby improving the accuracy of window occlusion detection, and the window occlusion alarm based on the window occlusion parameter Z can be more accurate, and the occlusion classification result based on the corrected first change parameter can also be more accurate.

[0013] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: determining whether to output an alarm message based on the window occlusion parameter and the occlusion weights of different areas of the window.

[0014] In combination with the first aspect, in some implementations of the first aspect, the method further includes: determining occlusion weights of different areas of the window based on the relative positions of the different areas of the window and the movement direction of the mobile device.

[0015] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: determining an alarm coefficient, and the determination of whether to output an alarm message includes: if the window occlusion alarm coefficient is greater than the occlusion alarm threshold, then the alarm message is output; or if the window occlusion alarm coefficient is less than or equal to the occlusion alarm threshold, then the alarm message is not output; wherein, the occlusion alarm threshold is determined according to the application scenario.

[0016] In the technical solution of the embodiment of the present application, the window occlusion alarm coefficient is determined based on the degree of influence of the position of the window occlusion point on the LIDAR performance and the degree of occlusion of each occlusion point. The window occlusion alarm coefficient can more accurately reflect the degree of occlusion of the occlusion object on the LIDAR window, and can reflect the degree of influence of the position of the occlusion object on the LIDAR window on the LIDAR performance. Compared with the method of making alarm decisions based on the occlusion area, the alarm probability is increased when the occlusion degree of a small area is severe and the important perspective is occluded, and the alarm probability is reduced when the occlusion degree of a large area is slight. The occlusion alarm threshold changes in real time with the LIDAR usage scenario, which can well meet the requirements of LIDAR use in different scenarios.

[0017] In combination with the first aspect, in certain implementations of the first aspect, the method further includes: determining the type of lidar window obstruction based on the distribution vector of the first change parameter and the window obstruction template, or determining the type of lidar window obstruction based on the distribution vector of the fourth change parameter and the window obstruction template, wherein the lidar window obstruction type includes the type of obstruction and the degree of obstruction.

[0018] In combination with the first aspect, in certain implementations of the first aspect, the method also includes: the window occlusion template is determined based on a third variation parameter of different types of occlusion degrees, the window transmittance, and the flow velocity of the occlusion object, wherein the window transmittance represents the occlusion degree of the lidar window occlusion object type.

[0019] In conjunction with the first aspect, in certain implementations of the first aspect, determining the type of the laser radar window obstruction includes: Ι , determine the type of the laser radar window obstruction, wherein the matching score score1 is determined according to the distribution vector of the first change parameter and the window obstruction template, or according to the distribution vector of the fourth change parameter and the window obstruction template; or, according to the matching score score1 and the matching score score П , determine the type of obstruction of the laser radar window, where the matching score П It is determined based on the blocking flow rate obtained by the first change parameter and the blocking flow rate of the window blocking template, or it is determined based on the blocking flow rate obtained by the fourth change parameter and the blocking flow rate of the window blocking template.

[0020] In the technical solution of the embodiments of this application, the obstruction type obtained through window occlusion template matching and obstruction flow rate matching is more accurate. Furthermore, the degree of obstruction can be determined, represented by the transmittance corresponding to the matched window occlusion template. Accurate obstruction type and degree allow LIDAR or MDC to determine the appropriate cleaning method. For example, if the obstruction is fog, heating the front window can eliminate the obstruction.

[0021] In a second aspect, a window occlusion detection device is provided, which includes an acquisition unit and a first processing unit: the acquisition unit is used to acquire a first window echo intensity and a second window echo intensity, the first window echo intensity represents a reference window echo intensity, and the second window echo intensity represents a window echo intensity at a current moment; the acquisition unit acquires detection parameters of a target point at a first moment and detection parameters at a second moment, the detection parameters including the reflectivity and / or position coordinates of the target point; the first processing unit acquires a window occlusion parameter based on the first window echo intensity, the second window echo intensity, the detection parameters at the first moment and the detection parameters at the second moment, and the window occlusion parameter is used to indicate a window occlusion state.

[0022] In the technical solution of the embodiment of the present application, the window occlusion parameters are preliminarily determined by the real-time collected second window echo intensity and the reference window echo intensity. On this basis, accurate window occlusion parameters are obtained through the detection parameters at the first moment and the detection parameters at the second moment, thereby achieving accurate detection of obstructions in the front window of the lidar, so as to improve the accuracy of obstruction warning and classification.

[0023] In combination with the second aspect, in certain implementations of the second aspect, obtaining the window occlusion parameter includes: the first processing unit is also used to obtain the window occlusion parameter based on the first change parameter and the second change parameter; wherein the first change parameter includes the difference between the second window echo intensity and the first window echo intensity, or the ratio of the second window echo intensity to the first window echo intensity; the second change parameter represents the change in the reflectivity of the target point at the first moment and the reflectivity at the second moment, or the change in the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment, and the estimated value of the position coordinate at the first moment is obtained based on the position coordinate at the second moment.

[0024] In the technical solution of the embodiment of the present application, the first change parameter is used to determine whether the LIDAR window is blocked. Compared with determining whether the LIDAR window is blocked by the window echo intensity, its advantage is that the window echo intensity change parameter can more clearly reflect the difference in window echo intensity before and after the window is blocked through a multiple relationship or a difference relationship, and is not affected by the size of the window echo intensity of the unobstructed window itself.

[0025] In combination with the second aspect, in some implementations of the second aspect, the device also includes an output unit: the output unit is used to output at least one of an alarm message, the first change parameter or the window occlusion parameter, the alarm message is used to remind the window occlusion status, and the first change parameter corresponds to the occlusion classification and / or occlusion partition.

[0026] In combination with the second aspect, in certain implementations of the second aspect, the device is further configured to: p elements in the first change parameter that are greater than a first threshold constitute a first element, where p is a positive integer; q elements in the first change parameter that are less than or equal to the first threshold constitute a second element, and the value of the second element is 0, where q is a positive integer; obtaining the window occlusion parameter comprises: determining the window occlusion parameter according to a weighted sum of values ​​of at least one fourth change parameter, the fourth change parameter including the second element, the third element, and the fourth element; verifying the first element according to the second change parameter to obtain the third element and the fourth element; the third The element represents n times of r elements in the first element, and the r elements represent elements in the second change parameter corresponding to the r elements that are less than or equal to the second threshold, or the r elements represent elements in the second change parameter corresponding to the r elements that are less than or equal to the third threshold, wherein r is a positive integer, and n is a positive real number less than 1 or 0; the fourth element represents s elements in the first element, and the s elements represent elements in the second change parameter corresponding to the s elements that are greater than the second threshold, or the s elements represent elements in the second change parameter corresponding to the s elements that are greater than the third threshold, wherein s is a positive integer.

[0027] In the technical solution of the embodiment of the present application, the window occlusion parameter Z obtained after correcting the first change parameter by the second change parameter has more accurate window occlusion information, eliminating the influence of the inflated window echo intensity caused by other factors such as temperature, power fluctuation of the laser transmitter, etc., thereby improving the accuracy of window occlusion detection, and the window occlusion alarm based on the window occlusion parameter Z can be more accurate, and the occlusion classification result based on the corrected first change parameter can also be more accurate.

[0028] In combination with the second aspect, in some implementations of the second aspect, the first processing unit is further used to: determine whether to output warning information based on the window occlusion parameter and the occlusion weights of different areas of the window.

[0029] In combination with the second aspect, in some implementations of the second aspect, the first processing unit is further used to: determine the occlusion weights of different areas of the window based on the relative positions of the different areas of the window and the movement direction of the mobile device.

[0030] In combination with the second aspect, in certain implementations of the second aspect, the first processing unit is further used to: determine an alarm coefficient, and the determination of whether to output an alarm message includes: if the window occlusion alarm coefficient is greater than the occlusion alarm threshold, then the alarm message is output; or if the window occlusion alarm coefficient is less than or equal to the occlusion alarm threshold, then the alarm message is not output; wherein, the occlusion alarm threshold is determined according to the application scenario.

[0031] In the technical solution of the embodiment of the present application, the window occlusion alarm coefficient is determined based on the degree of influence of the position of the window occlusion point on the LIDAR performance and the degree of occlusion of each occlusion point. The window occlusion alarm coefficient can more accurately reflect the degree of occlusion of the occlusion object on the LIDAR window, and can reflect the degree of influence of the position of the occlusion object on the LIDAR window on the LIDAR performance. Compared with the method of making alarm decisions based on the occlusion area, the alarm probability is increased when the occlusion degree of a small area is severe and the important perspective is occluded, and the alarm probability is reduced when the occlusion degree of a large area is slight. The occlusion alarm threshold changes in real time with the LIDAR usage scenario, which can well meet the requirements of LIDAR use in different scenarios.

[0032] In combination with the second aspect, in some implementations of the second aspect, the device also includes a second processing unit: the second processing unit is also used to determine the lidar window obstruction type based on the distribution vector of the first change parameter and the window obstruction template, or determine the lidar window obstruction type based on the distribution vector of the fourth change parameter and the window obstruction template, wherein the lidar window obstruction type includes the obstruction type and the obstruction degree of the obstruction.

[0033] In combination with the second aspect, in certain implementations of the second aspect, the device also includes: the window occlusion template is determined based on a third variation parameter of different types of occlusion degrees, the window transmittance, and the flow velocity of the occlusion object, wherein the window transmittance represents the occlusion degree of the lidar window occlusion object type.

[0034] In conjunction with the second aspect, in some implementations of the second aspect, the second processing unit is further configured to: Ι, determine the type of the laser radar window obstruction, wherein the matching score score1 is determined according to the distribution vector of the first change parameter and the window obstruction template, or according to the distribution vector of the fourth change parameter and the window obstruction template; or I and matching score П , determine the type of obstruction of the laser radar window, where the matching score П It is determined based on the blocking flow rate obtained by the first change parameter and the blocking flow rate of the window blocking template, or it is determined based on the blocking flow rate obtained by the fourth change parameter and the blocking flow rate of the window blocking template.

[0035] In the technical solution of the embodiments of this application, the obstruction type obtained through window occlusion template matching and obstruction flow rate matching is more accurate. Furthermore, the degree of obstruction can be determined, represented by the transmittance corresponding to the matched window occlusion template. Accurate obstruction type and degree allow LIDAR or MDC to determine the appropriate cleaning method. For example, if the obstruction is fog, heating the front window can eliminate the obstruction.

[0036] In a third aspect, a computer-readable medium is provided, which stores a program code for execution by a device, wherein the program code includes a method for executing the window occlusion detection method in the first aspect or any implementation manner of the first aspect.

[0037] In a fourth aspect, a computer program product is provided, comprising: a computer program code, which, when executed on a computer, enables the computer to execute the window occlusion detection method in the first aspect or any one of the implementations of the first aspect.

[0038] In a fifth aspect, a chip is provided, which includes at least one processor and an interface circuit, and at least one processor calls a computer program through the circuit, so that the device where the chip is located executes the window occlusion detection method in the above-mentioned first aspect or any one of the implementation methods of the first aspect.

[0039] Optionally, as an implementation method, the chip may also include a memory, in which instructions are stored, and the processor is used to execute the instructions stored in the memory. When the instructions are executed, the processor is used to execute the window occlusion detection method in the first aspect or any one of the implementation methods of the first aspect.

[0040] In the sixth aspect, a device is provided, comprising: a processor and a memory, the memory being used to store the computer program code, and when the computer program code runs on the processor, the device executes the window occlusion detection method in the first aspect or any one of the implementations of the first aspect.

[0041] In a seventh aspect, a terminal is provided, comprising the window occlusion detection device according to the second aspect or any one of the second aspects, or the computer-readable medium according to the third aspect, or the chip according to the fifth aspect. The terminal comprises a vehicle, a drone, or a robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a schematic diagram of the imaging after the LIDAR window of this application is blocked by water droplets;

[0043] Figure 2 This is a schematic diagram of the LIDAR window echo and target echo in this application;

[0044] Figure 3 This is a schematic diagram of the sensor distribution of the advanced driver assistance system according to an embodiment of the present application;

[0045] Figure 4 This is a schematic diagram of LIDAR mapping carried out by a drone in an embodiment of the present application;

[0046] Figure 5 This is a schematic diagram of a LIDAR window detection and classification method in the prior art;

[0047] Figure 6 This is a schematic diagram of a method for LIDAR window occlusion detection in an embodiment of the present application;

[0048] Figure 7 This is a graph showing changes in occlusion parameters obtained by LIDAR window occlusion detection in an embodiment of the present application;

[0049] Figure 8 This is a schematic diagram of the LIDAR coordinate system change in the embodiment of the present application;

[0050] Figure 9 This is a schematic diagram of a LIDAR window occlusion alarm method in an embodiment of the present application;

[0051] Figure 10 is the occlusion weight W of different regions of the LIDAR window in the embodiment of the present application j2 Schematic diagram;

[0052] Figure 11 This is a schematic diagram of a method for making a LIDAR window occlusion template in an embodiment of the present application;

[0053] Figure 12This is a schematic diagram of the LIDAR window occlusion template production process in an embodiment of the present application;

[0054] Figure 13 Schematic diagram of a method for classifying LIDAR window obstructions in an embodiment of the present application;

[0055] Figure 14 is a schematic block diagram of a window occlusion detection device in an embodiment of the present application;

[0056] Figure 15 It is a schematic block diagram of another window occlusion detection device in an embodiment of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in this application will be described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments in this application without creative effort are also within the scope of protection of this application.

[0058] For the sake of simplicity, laser radar will be referred to as LIDAR below. It should be understood that there is no essential difference between the two.

[0059] For clarity, Figure 2 This is a schematic diagram of the LIDAR window echo and target echo in this application. First, combine Figure 2 The terms used in this application are explained.

[0060] Window echo: The laser emitted by the LIDAR laser transmitter 201 is reflected by the surface of the front window 203. The LIDAR detector 202 receives the reflected laser light and generates an electrical signal, which is the LIDAR window echo. The window echo is enhanced when there is an obstruction on the window.

[0061] Target echo: The laser emitted by the LIDAR laser transmitter 201 reflects off the surface of the target 204. The LIDAR detector 202 receives the reflected laser light and generates an electrical signal, which is the LIDAR target echo. If there is an obstruction on the viewing window, the target echo will be weakened.

[0062] This application LIDAR can be applied to ADAS system, Figure 3 This is a schematic diagram of the sensor distribution of the advanced driver assistance system in an embodiment of the present application. The sensors may include an ultrasonic radar 301, a camera 302, and a LIDAR 303. Other sensors may also be included, which are not limited in this application.

[0063] The LIDAR 303 in the embodiment of the present application can be installed at the center of the front of the vehicle, or can be installed at other locations in the vehicle, such as the left side or right side of the vehicle, which is not limited in the embodiment of the present application.

[0064] This application LIDAR can also be used in surveying and mapping and remote sensing technology, Figure 3 This is a schematic diagram of a drone equipped with LIDAR for mapping in an embodiment of the present application, where 401 is a drone and 402 is a LIDAR. Drones equipped with LIDAR can be used to map terrain or urban traffic, which is not limited in this embodiment of the present application.

[0065] LIDAR has the characteristics of quickly and accurately acquiring information about the surrounding environment. If there is an obstruction on the LIDAR window, it will affect the performance of the LIDAR.

[0066] Figure 5 This is a schematic diagram of a prior art method for LIDAR window detection and classification. S501: LIDAR obtains reflected light intensity from point cloud data. S502: Identify obstacles based on the point cloud data. S503: Obstacles within the distance range of the LIDAR window are identified as suspected obstructions. S504: If at least one reflected light intensity from the suspected obstruction is greater than a first preset light intensity, then an obstacle is determined to be present on the LIDAR window. S505: Determine whether the obstruction is transparent or opaque based on the first and second preset light intensities. S506: Determine the position and size of the obstruction on the window based on the obstruction's data points and generate a cleaning instruction to clean the window.

[0067] In the existing technology, the window obstruction detection results obtained only based on the window echo intensity and a single threshold are inaccurate. The window echo intensity is also affected by other factors, such as temperature, power fluctuations of the laser generator, and fluctuations of the detection device itself. Therefore, only using the window echo intensity for obstruction detection will have a high false alarm rate.

[0068] Existing technologies issue cleaning instructions based solely on the position and size of obstructions on the window. However, when the LIDAR window requires cleaning, the LIDAR becomes unavailable. Consequently, this approach fails to consider the varying impacts of obstructions of varying degrees and locations on autonomous driving, necessitating a trade-off between degraded LIDAR performance and temporary LIDAR unavailability. In other words, the LIDAR cleaning instruction must be issued in a timely manner to minimize the impact of window obstructions and reduce the duration of LIDAR unavailability.

[0069] Classifying obstructions based on window echo intensity and preset light intensity only allows for transparent and opaque obstructions, and different obstruction types require different cleaning strategies. For example, mud obstructions require focused cleaning of the obstructed area, dust obstructions are evenly distributed, requiring uniform window cleaning, and fog obstructions require window heating. Therefore, simply classifying obstructions as transparent and opaque can lead to inappropriate cleaning strategies due to the limited number of obstruction types recognized and the lack of information on the degree of obstruction.

[0070] In the existing technology, different types of occlusion are identified based on the echo signal characteristics after window occlusion. The characteristics of the echo signal include the intensity value of the window echo, the spatial distribution of the occlusion, whether the occlusion area moves, and whether there is a target point in the occlusion area. Specifically, as shown in Table 1, Table 1 shows the signal characteristics of different types of occlusions.

[0071] Table 1

[0072]

[0073] The echo signal characteristics in Table 1 cannot distinguish between dust and moisture, two types of obstructions. In addition, the echo signal characteristics cannot distinguish between different degrees of obstruction of the same type of obstruction.

[0074] In order to solve the above technical problems, the embodiments of the present application provide a method and device for window occlusion detection, alarm and classification.

[0075] Figure 6 This is a schematic diagram of a LIDAR window occlusion detection method in an embodiment of the present application.

[0076] S601 : Acquire a first window echo intensity and a second window echo intensity, wherein the first window echo intensity represents a reference window echo intensity, and the second window echo intensity represents a current window echo intensity.

[0077] It should be noted that the echo is detected by the Lidar detector and has a corresponding intensity. As an implementation method, the first window echo intensity P C The reference window echo intensity represents the reference window echo intensity, which represents the window echo intensity collected when the LIDAR front window is not blocked by any obstruction. The reference window echo intensity can be obtained by the LIDAR offline. For example, the reference window echo intensity can be a parameter set by the LIDAR factory. For example, the reference window echo intensity can also be a parameter collected by the LIDAR online or an echo intensity obtained based on the parameter collected online. That is to say, the reference window echo intensity can be adjusted after the LIDAR leaves the factory, which is not limited in the embodiments of the present application.

[0078] In addition, when the first window echo intensity is a parameter set by the LIDAR factory, the acquisition unit can obtain the preset first window echo intensity; when the first window echo intensity is a parameter collected online by the LIDAR or an echo intensity obtained based on the parameter collected online, the acquisition unit obtains the first window echo intensity from the LIDAR detector 202 or obtains the first window echo intensity obtained based on the echo information from the Lidar detector 202.

[0079] Furthermore, the second window echo intensity may be a parameter collected online by the LIDAR or an echo intensity obtained based on the parameter collected online. The acquisition unit acquires the second window echo intensity from the LIDAR detector 202 or acquires the second window echo intensity based on the echo information from the Lidar detector 202.

[0080] Generally speaking, the echo intensity obtained in step S601 needs to be used as input for subsequent acquisition or calculation of occlusion parameters. Therefore, the echo intensity here can include the echo information fed back to the Lidar processor after the Lidar detector detects the echo signal (in this case, the echo information can be used as echo intensity for subsequent parameter acquisition or calculation), or it can also include the echo intensity obtained after the Lidar processor further processes the echo information. The specific details depend on the structure and performance of the Lidar itself.

[0081] S602 : Acquire detection parameters of the target point at the first moment and detection parameters of the target point at the second moment. The detection parameters may include reflectivity and / or position coordinates of the target point.

[0082] As a possible implementation method, inertial measurement unit IMU information can also be obtained. The IMU information obtains the LIDAR rotation angular velocity and motion linear velocity in the LIDAR coordinate system through the IMU. If the IMU is on the vehicle, the IMU information obtained in the vehicle coordinate system needs to be converted into IMU information in the LIDAR coordinate system; if the IMU is on the LIDAR, no coordinate conversion is required, and this is not limited in the embodiments of the present application.

[0083] S603 : Acquire a window occlusion parameter according to the first window echo intensity, the second window echo intensity, the detection parameter at the first moment, and the detection parameter at the second moment.

[0084] As an implementation method, the window occlusion parameter is obtained according to the first change parameter and the second change parameter, wherein the first change parameter and the second change parameter have a one-to-one correspondence.

[0085] It should be understood that the first variation parameter is used to indicate the first window echo intensity P CThe first change parameter can be the change between the second window echo intensity P and the first window echo intensity P. C The difference, PP C Alternatively, the first variation parameter may also be the second window echo intensity P and the first window echo intensity P C The ratio of P / P C , which is not limited in the embodiments of this application.

[0086] It should be understood that the second variation parameter is used to verify the elements determined as window occlusion points in the first variation parameter, thereby improving the accuracy of window occlusion detection.

[0087] It should be understood that if the first variation parameter is the second window echo intensity P and the first window echo intensity P C The first threshold value may be 1.2, which is not limited in the embodiment of the present application; if the first change parameter is the second window echo intensity P and the first window echo intensity P C The value of the first threshold depends on the intensity range output by the LIDAR detector and is not limited in the embodiment of the present application.

[0088] As an implementation method, there are multiple elements in the first change parameter, wherein each element of the first change parameter is the window echo change intensity of a point on the LIDAR front window. By judging whether the value of each element in the multiple elements is greater than the first threshold, it is determined whether the point corresponding to the element is blocked.

[0089] The p elements in the first change parameter that are greater than the first threshold constitute the first element, that is, the LIDAR window position corresponding to each element in the p elements is blocked by the obstruction, p is a positive integer, that is, in the embodiment of the present application, there is no limit on the number of elements in the first change parameter that are greater than the first threshold, and it can be one or more.

[0090] The q elements in the first change parameter that are less than or equal to the first threshold constitute the second element, then the LIDAR window point corresponding to the q elements is not blocked by the obstruction, the value of each element in the q elements is 0, and q is a positive integer, that is, in the embodiment of the present application, there is no limit on the number of elements in the first change parameter that are less than or equal to the first threshold, and it can be one or more.

[0091] The advantage of using the first change parameter to determine whether the LIDAR window is obstructed is compared to using the window echo intensity to determine whether the LIDAR window is obstructed. The window echo intensity change parameter can more clearly reflect the difference in window echo intensity before and after window obstruction through a multiple relationship or a difference relationship, and is not affected by the magnitude of the window echo intensity of the unobstructed window itself.

[0092] Figure 7 This is a graph showing changes in window occlusion parameters obtained by LIDAR window occlusion detection in an embodiment of the present application.

[0093] The field of view (FOV) of the LIDAR is determined by the horizontal scanning and vertical scanning of the LIDAR. For example, the range of the horizontal scanning angle of the LIDAR can be [-50°, +50°], and the range of the vertical scanning angle can be [-15°, +15°]. There is no limitation in the embodiments of the present application.

[0094] It should be understood that the first element is checked according to the second variation parameter to obtain the window occlusion parameter Z, such as Figure 7 shown.

[0095] Since the window echo intensity may change due to other factors, the occlusion point in the first element is verified to be a real occlusion point through the second change parameter, so as to correct the first element to obtain a more accurate window occlusion parameter Z. The window occlusion parameter can exist in the form of a matrix or other parameter forms, which is not limited in the embodiments of the present application.

[0096] The basis for verifying the occlusion point in the first element through the second change parameter is that if there is occlusion in the LIDAR window, the reflectivity of the target point after the occlusion will decrease, and the ranging accuracy will decrease, wherein the second change parameter includes the change in the reflectivity of the target point obtained by the LIDAR at the first moment and the reflectivity at the second moment, or the change in the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment, wherein the estimated value of the position coordinate at the first moment is obtained by transforming the position coordinate at the second moment based on IMU information.

[0097] It should be understood that whether the occlusion point in the first element is a real occlusion point can be verified only through the target point position coordinate change information or only through the target point reflectivity change information.

[0098] The following combination Figure 8 To explain in detail how to verify whether the occlusion point in the first element is a real occlusion point by using the second variation parameter. Figure 8 This is a schematic diagram of the LIDAR coordinate system changes in the embodiment of the present application.

[0099] If there is occlusion at a certain point in the Nth frame window, that is, the point corresponding to a certain element in the first element is occluded, the coordinates of the target point in the Nth frame LIDAR coordinate system are marked as X N =(x N ,y N ,z N), which is the position coordinate of the target point at the second moment, is estimated based on the IMU information and the kinematic principle to obtain the target point estimated coordinate X in the LIDAR coordinate system of the Mth frame when the window is unobstructed. M =(x M ,y M ,z M ), which is the estimated value of the position coordinate at the first moment. The actual coordinate of the target point corresponding to the angle in the LIDAR coordinate system of the Mth frame is X M '=(x M ',y M ',z M '), that is, the actual value of the position coordinate at the first moment, if X M and X M ' is greater than the second threshold, it is determined that the point is blocked, and the corresponding first element is not modified. The unmodified first element is represented as the fourth element, that is, the fourth element represents s elements in the first element, where s elements represent elements in the second change parameter corresponding to the s elements that are greater than the second threshold, where s is a positive integer; if X M and X M ' is less than or equal to the second threshold, then there is no occlusion at the point, that is, the corresponding first element is a false alarm parameter, and the first element will be corrected. The correction method can modify the value of the first element to n times the value of the first element. The modified first element is the third element, that is, the third element represents r elements in the first element, and the r elements represent the elements in the second change parameter corresponding to the r elements that are less than or equal to the second threshold, where r is a positive integer and n is a positive real number less than 1 or 0. The window occlusion parameter Z is determined according to the weighted sum of the values ​​of at least one fourth change parameter, where the fourth change parameter includes the second element, the third element and the fourth element, such as Figure 7 shown.

[0100] If there is occlusion at a point in the Nth frame window, that is, there is occlusion at a point corresponding to an element in the first element, the reflectivity of the target point in the Nth frame, that is, the reflectivity of the target point at the second moment and the reflectivity of the target point in the Mth frame when the window is unobstructed, that is, the reflectivity of the target point at the first moment, can be compared. If the difference is greater than the third threshold, it is determined that the point is occluded, and the corresponding first element is not modified. The unmodified first element is represented as the fourth element, that is, the fourth element represents s elements in the first element, where s elements represent elements in the second change parameter corresponding to the s elements that are greater than the second threshold, where s is a positive integer; if the difference is less than or equal to the third threshold, then there is no occlusion at the point, that is, the corresponding first element is a false alarm parameter, and the first element will be corrected. The correction method can be n times the value of the first element. The modified first element is the third element, that is, the third element represents r elements in the first element, and r elements represent elements in the second change parameter corresponding to the r elements that are less than or equal to the second threshold, where r is a positive integer and n is a positive real number less than 1 or 0. The window occlusion parameter Z is determined according to a weighted sum of the values ​​of at least one fourth variation parameter, wherein the fourth variation parameter includes a second element, a third element, and a fourth element, such as Figure 7 It should be understood that M and N are both positive integers, and M is smaller than N.

[0101] The LIDAR FOV can be divided into regions in a grid format, with each region corresponding to a grid. The value of the view window occlusion parameter Z for each region is determined by the weighted sum of the values ​​of at least one fourth variation parameter corresponding to all points in the region. It should be understood that when a region is unobstructed, the value of the view window occlusion parameter Z for that region is 0.

[0102] The window occlusion parameter Z obtained after correcting the first change parameter by the second change parameter has more accurate window occlusion information, eliminating the influence of inflated window echo intensity caused by other factors such as temperature and power fluctuations of the laser transmitter, thereby improving the accuracy of window occlusion detection. The window occlusion alarm based on Z can be more accurate, and the occlusion classification result based on the corrected first change parameter can also be more accurate.

[0103] S604: Output at least one of an alarm message, a first change parameter, or a window occlusion parameter, wherein the alarm message is used to remind the window occlusion state, and the first change parameter corresponds to an occlusion classification and / or an occlusion partition.

[0104] It should be understood that at least one of the warning information, the first change parameter or the window occlusion parameter is output to a multi-domain controller (MDC), which may also be referred to as an intelligent driving domain controller.

[0105] It should be understood that the window occlusion parameter is used to indicate the occlusion distribution state of the occlusion object on the LIDAR front window.

[0106] It should be understood that the first variation parameter corresponds to occlusion classification and / or occlusion partitioning, that is, the first variation parameter is used for occlusion classification and / or occlusion partitioning. Occlusion partitioning is to partition the occlusion situation of the LIDAR front window by the occlusion object. Using the first variation parameter to perform occlusion partitioning can initially obtain the occlusion situation partitioning of the window. Occlusion classification is to classify the occlusion objects on the LIDAR front window. Using the first variation parameter to perform occlusion classification can initially obtain the type of the occlusion object.

[0107] It should be understood that whether to output the warning information is determined by the window occlusion parameters and the occlusion weights of different window areas. Figure 9 Specifically, Figure 9 This is a schematic diagram of a LIDAR window occlusion alarm method in an embodiment of the present application.

[0108] Among them, the occlusion weights of different window areas are used to indicate the degree of influence of the window area where the occlusion point on the LIDAR front window is located on the LIDAR performance. If the occlusion weight of a certain window area is larger, it means that the occlusion of the area has a greater impact on the LIDAR performance; if the occlusion weight of a window area is smaller, it means that the occlusion of the area has a smaller impact on the LIDAR performance.

[0109] S901, based on the occlusion degree weight W of the window occlusion point i1 Occlusion weight W of different areas of the window j2 , get the window occlusion alarm coefficient, where i represents different points on the window, j represents different areas on the window, and i and j are positive integers.

[0110] Window occlusion point occlusion degree weight W i1 The second window echo intensity P of the i-th point in the window can be i And the first window echo intensity P at this point iC The ratio or difference is obtained, that is, W i1 =P i / P iC or W i1 =P i -P iC , which is the first change parameter of this point, is not limited in the embodiment of the present application.

[0111] Occlusion weights W for different viewport regions j2 It can be determined by the LIDAR itself or according to the ADAS system performance requirements, and the comparison is never limited in this application.

[0112] If the occlusion weights W of different regions of the windowj2 The weight is determined by the LIDAR itself, so it can be related to the position of the LIDAR installed on the mobile device. The closer the window area is to the center of motion of the mobile device, the greater the occlusion weight W j2 The larger the occlusion weight W is, the more likely it is that the target in the environment in the direction of the mobile device's motion center can influence the subsequent mobile device motion planning or decision-making more than the target in the environment around the mobile device's motion center. Therefore, the closer the LIDAR window is to the area in the mobile device's motion center, the higher its occlusion weight W is. j2 The occlusion weights of different areas of the window can also be determined according to actual LIDAR performance requirements and are not limited in the embodiments of this application.

[0113] Taking the LIDAR installed on the vehicle as an example, the following Figure 10 For different window occlusion weights W j2 For example, Figure 10 is the occlusion weight W of different regions of the LIDAR window in the embodiment of the present application j2 Schematic diagram.

[0114] like Figure 10 As shown in (b), if the LIDAR is installed at the center of the vehicle, the center of its view window is the most important area. Therefore, the occlusion weight of the center of the view window can be set to 1. The importance of the area around the center of the view window is second only to the center of the view window, and it can be set to 0.75. And so on. Each area of ​​the view window has its own occlusion weight according to its importance. It should be understood that Figure 10 The occlusion weights W of different regions of the window in j2 This is just an example and this application does not limit its value.

[0115] like Figure 10 As shown in (a), if the LIDAR is installed on the left side of the vehicle, the middle right area of ​​its window is the most important area. Therefore, the occlusion weight of the middle right area of ​​the window can be set to 1. The importance of the middle right area of ​​the window is second only to the center area of ​​the window, and it can be set to 0.8. And so on. Each area of ​​the window has its own occlusion weight according to its importance. It should be understood that Figure 10 The occlusion weights W of different regions of the window in j2 This is just an example and this application does not limit its value.

[0116] like Figure 10As shown in (c), if the LIDAR is installed on the left side of the vehicle, the middle right area of ​​its window is the most important area. Therefore, the occlusion weight of the middle left area of ​​the window can be set to 1. The importance of the middle left area of ​​the window is second only to the center area of ​​the window, and it can be set to 0.8. And so on. Each area of ​​the window has its own occlusion weight according to its importance. It should be understood that Figure 10 The occlusion weights W of different regions of the window in j2 This is just an example and this application does not limit its value.

[0117] The occlusion degree weight W of the window occlusion point i1 And the occlusion weight W of the window area to which the occlusion point belongs j2 Perform weighted sum operation to obtain the occlusion alarm coefficient W, that is, W = ∑W i1 ×W j2 .

[0118] S902, determine whether the occlusion alarm coefficient is greater than the alarm threshold. If the occlusion alarm coefficient is greater than the occlusion alarm threshold, start the alarm instruction to execute S903. If the occlusion alarm coefficient is not greater than the occlusion alarm threshold, do not start the alarm instruction to execute S904.

[0119] It should be understood that after the occlusion alarm command is activated, the LIDAR window needs to be cleared and the LIDAR is in an unusable state. In order to meet the requirements of LIDAR use in different scenarios, the occlusion alarm threshold changes with the real-time scene changes.

[0120] For example, in the autonomous driving field, the occlusion warning threshold in an empty, deserted suburban area is lower than in a bustling urban area. The environmental targets in a bustling urban area are more complex and variable than those in an empty, deserted suburban area, requiring the LIDAR to be available for a longer period of time. In contrast, the requirement for LIDAR availability in an empty, deserted suburban area is lower than in a bustling urban area. Therefore, setting different occlusion warning thresholds for different scenarios can effectively meet the requirements of LIDAR usage in different scenarios.

[0121] The window occlusion alarm coefficient obtained in the embodiment of the present application is determined based on the degree of influence of the position of the window occlusion point on the LIDAR performance and the degree of occlusion of each occlusion point. This window occlusion alarm coefficient can more accurately reflect the degree of occlusion of the occlusion object on the LIDAR window, and can reflect the degree of influence of the position of the occlusion object on the LIDAR window on the LIDAR performance. Compared with the method of making alarm decisions based on the occlusion area in the prior art, the alarm probability is increased when the occlusion degree of a small area is severe and when the important viewing angle is occluded, and the alarm probability is reduced when the occlusion degree of a large area is minor.

[0122] In addition, the occlusion alarm threshold in the embodiment of the present application changes in real time with the LIDAR usage scenario, which can well meet the requirements of LIDAR usage in different scenarios.

[0123] When an obstruction appears in the LIDAR's front window, not only does the LIDAR need to be alerted, but the obstruction can also be classified to determine its type and degree of obstruction. The LIDAR's own processor can classify obstructions, and the MDC's processor can also classify obstructions. This is not limited to this embodiment of the application.

[0124] In addition, there is no order restriction between the alarm and classification of window obstructions. The alarm and classification can be performed simultaneously, or the alarm can be performed first and then the classification, or the classification can be performed first and the classification result can be output as an alarm information. This is not restricted in the embodiments of the present application.

[0125] Before classifying LIDAR window occlusions, it is necessary to create a window occlusion template. Figure 11 Introducing the method of making window occlusion template. Figure 11 This is a schematic diagram of a method for making a LIDAR window occlusion template in an embodiment of the present application.

[0126] It should be understood that the production of this window occlusion template should be completed before the LIDAR is put into actual use.

[0127] S1101, collecting the first window echo intensity and the first target echo intensity p y .

[0128] It should be understood that the first window echo intensity P collected by LIDAR in an open space is C It can be used for window occlusion detection, that is, the first window echo intensity preset by the LIDAR factory; the first window echo intensity p collected by the LIDAR in a fixed environment where the target is stationary and the distance is appropriate c It can be used to make a window mask template. The first target echo intensity p y It is the target echo intensity when there is no obstruction in the LIDAR front window.

[0129] The first window echo intensity includes the first window echo intensity of all points in a frame. The window echo intensity of all points in a frame can be obtained by the median or mean of the window echo intensity of corresponding points in multiple frames.

[0130] Among them, the conditions that need to be met for the target distance to be appropriate are that there are detectable targets in all directions within the LIDAR field of view, and the window echo and the target echo do not overlap.

[0131] S1102: Collect the third window echo intensity p under different types of obstructions and different degrees of obstruction. c ' and the second target echo intensity p y '.

[0132] It should be understood that in order to improve the accuracy and credibility of the experimental data, the third window echo intensity and the second target echo intensity of the same shielding type and shielding object with the same shielding degree are collected multiple times.

[0133] The type of the obstruction may be water, fog, dust, mud or white paper, which is not limited in the embodiment of the present application.

[0134] S1103, creating a third variation parameter histogram under different types and different shielding degrees and obtaining the transmittance and shielding flow velocity.

[0135] The third variation parameter can be expressed as the third window echo intensity p c ' and the first window echo intensity p c The difference between the two is expressed as p c '-p c , you can also use the third window echo intensity p c ' and the first window echo intensity p c The ratio of p c ' / p c .

[0136] The third variation parameter histogram is used to describe the variation parameters of the window echo intensity at all points after occlusion. Figure 12 Schematic diagram of the LIDAR window occlusion template production process in the embodiment of the present application, wherein Figure 12 (a) is the histogram of the third parameter change after one experiment with different types of occluders, that is, after each experiment, the following can be obtained: Figure 12 Histogram of the third variation parameter of all points in the window of (a).

[0137] The abscissa T of the third variation parameter histogram represents the boundary of the third variation parameter interval, T=[t1,...,t n+1 ], the vertical coordinate U is the third variation parameter distribution vector of all points, U=[μ1,...,μ n ],μ a Indicates the third variation parameter interval [t a ,t a+1 ], where a∈n, n is a positive integer.

[0138] Optionally, a normalized distribution vector U / |U|1 may be used to represent the third variation parameter distribution vector of all points.

[0139] When the degree of occlusion by the occluder is relatively light, for example, when only a small area of ​​the window is blocked, the third change parameter values ​​of most points are small, while the third change parameter values ​​of a small number of points are large. However, the points that have a greater impact on the occlusion classification are the points with large third change parameter values ​​but few in number. Therefore, different normalization processing can also be performed on different intervals to reduce the impact of the window occlusion area on the window occlusion classification, thereby increasing the difference between the third change parameter distribution vectors of the same occlusion type and different occlusion degrees, thereby producing a more differentiated window occlusion template. Figure 12 (b) is the histogram of changing the boundary of T, which is a preparation for performing different normalization processing on different intervals.

[0140] For example, the third change parameter distribution vector U5 = [μ1, μ2, μ3, μ4, μ5] of the third change parameter interval [t1, t2], [t2, t3], [t3, t4], [t4, t5], [t5, t6] can be normalized to obtain a distribution vector μ a = U5 / |U5|1, a∈[1,5]; the remaining third variation parameter intervals [t6,t7],[t7,t8],[t8,t9],...,[t n ,t n+1 ]The third change parameter distribution vector is normalized U n-5 =[μ6,μ7,μ8,...,μ n ], get the distribution vector μ b =U n-5 / |U n-5 |1, a∈[6,n], n is a positive integer.

[0141] The degree of occlusion can be expressed by the window transmittance α, where the transmittance α at each point is i is the second target echo intensity p at this point iy ' and the first target echo intensity p iy The square root of the ratio of Window transmittance α Transmittance α at all points i The mean of .

[0142] It should be understood that each experiment will obtain the third window echo intensity p of all points in the experiment. c ' and the second target echo intensity p y ', and the window transmittance α under this experiment, the third window echo intensity of each point is the median value of the third window echo intensity of this point in multiple frames in this experiment, and the second target echo intensity of each point is the median value of the second target echo latency of this point in multiple frames in this experiment.

[0143] The flow rate of different obstructions is represented by the maximum and minimum flow rates of different obstructions, that is, v∈[vmin ,v max ], where v min and v max The flow rate of the same type of obstruction is the same. For example, the flow rate of non-flowing obstructions such as dust and mud is 0, and the flow rate of flowing obstructions such as rainwater is an empirical value, which is not limited in the embodiments of the present application.

[0144] For the same type of obstruction with the same degree of obstruction, perform l tests. For the same type of obstruction with k degrees of obstruction and m types of obstruction, repeat steps S1102 and S1103. Each test obtains the corresponding window echo intensity variation parameter distribution vector and its corresponding transmittance. The number of obstruction levels k for different types of obstructions can also vary.

[0145] S1104, selecting the third variation parameter distribution vector obtained from l experiments of the same type and the same occlusion degree to obtain a window occlusion template of the same type and the same occlusion degree. There are k window occlusion templates of the same type and different occlusion degrees.

[0146] It should be understood that when making the selection, a clustering algorithm may be used, such as K-means clustering or other clustering methods, which is not limited in the embodiments of the present application. Figure 12 (c) is the window occlusion template of different types of occluders and different occlusion degrees after normalization and clustering. The number of templates y is the m types of occluders multiplied by k occlusion degrees, that is, y=mk.

[0147] S1105 , according to the window shielding templates of different shielding types and shielding degrees, the transmittance and shielding flow velocity under the templates are obtained.

[0148] It should be understood that since each experiment will obtain a third change parameter distribution vector and its corresponding transmittance, it is necessary to select a certain transmittance from the transmittances corresponding to the third change parameter distribution vectors obtained from l experiments of the same type and the same occlusion degree as the transmittance of the clustered window occlusion template.

[0149] The specific selection basis is the matching score between the third variation parameter distribution vector obtained in the lth experiment and the window occlusion template of the same type and the same occlusion degree. x =-|o x -S y |, where o x is the third variation parameter distribution vector of the same type and the same occluder l times, x∈[1,l], S y is the window occlusion template of the same type and occlusion degree, y∈[1,mk]. Matching score xThe higher the value is, the closer the transmittance corresponding to the third variation parameter distribution vector in this test is to the transmittance of the window occlusion template. Window occlusion templates of the same type and the same occlusion degree all have a transmittance α obtained through template matching. y , where y∈[1,mk].

[0150] Transmittance α of the window occlusion template y You can select the transmittance corresponding to the third variation parameter distribution vector with the highest matching score, or calculate multiple matching scores. x The median or mean of the transmittance corresponding to the higher third variation parameter distribution vector is obtained, and multiple matching scores can also be obtained. x The maximum and minimum values ​​of the transmittance corresponding to the higher third variation parameter distribution vector are used as the transmittance interval α of the window occlusion template. y '=[α min ,α max ], which is not limited in the embodiments of the present application.

[0151] It should be understood that the flow rate of the obstruction of different templates is determined by the type of obstruction, and the flow rate of different obstructions is represented by the maximum and minimum flow rates of different obstructions, that is, v∈[v min ,v max ], where v min and v max For experience value.

[0152] Next, we will combine the window occlusion template and Figure 13 The method of classifying LIDAR window occlusions is explained. Figure 13 A schematic diagram of a method for classifying LIDAR window occluders in an embodiment of the present application.

[0153] S1301: Obtain a distribution vector of a first variation parameter or a distribution vector of a fourth variation parameter.

[0154] It should be understood that both the first variation parameter and the fourth variation parameter verified by the second variation parameter can be used as input parameters for window obstruction classification.

[0155] The method for obtaining the distribution vector of the first change parameter is consistent with the method for obtaining the distribution vector of the third change parameter, and the method for obtaining the distribution vector of the fourth change parameter is consistent with the method for obtaining the distribution vector of the third change parameter, which will not be repeated here.

[0156] S1302: Obtain the matching score between the distribution vector of the first variation parameter and the window occlusion template. Ι , or obtain the distribution vector of the fourth change parameter and the window occlusion template matching score I .

[0157] It should be understood that O is the distribution vector of the first change parameter of the current window occlusion area, S y is the window occlusion template, y∈[1,mk], and the matching score is score Ι =-|OS y |1, when matching score I The closer to 0, the closer the distribution vector of the first variation parameter of the current window occlusion area is to the window template, which means that the type of occlusion object and the degree of occlusion of the current window occlusion area are closer to the window template.

[0158] S1303, obtaining the matching score of the occlusion flow velocity of the first variation parameter and the occlusion flow velocity of the window occlusion template П , or obtain the matching score of the fourth change parameter's occlusion flow velocity and the window occlusion template's occlusion flow velocity П .

[0159] It should be understood that the blocking flow velocity v in the current window blocking area flow It can be obtained based on Z obtained in S604, that is, the window occlusion parameter obtained by the weighted sum of at least one fourth variation parameter, specifically v flow =|Z N -Z N+1 |1 / |Z N |1, where Z N is the window occlusion parameter of the Nth frame, Z N+1 is the window occlusion parameter of the N+1 frame. The occlusion flow velocity v in the front window occlusion area flow It can also be obtained based on the second window occlusion parameter obtained by the weighted sum of at least one first variation parameter, which is not limited in the embodiment of the present application. The closer the matching score is to 0, the more it means that the flow rate of the occluder in the current window occlusion area is within the flow rate range of the occluder in the template, which means that the occluder type and occlusion degree in the current window occlusion area are closer to the occluder type and occlusion degree in the template.

[0160] S1304, based on the matching score I , determine the type of occluder in the window occlusion area and its occlusion degree, or according to the matching score I and matching score П , determine the type of occluder in the window occlusion area and its occlusion degree.

[0161] As an implementation method, in the embodiment of the present application, it is possible to only use the matching score IDetermine the type of the occluder in the window occlusion area and its occlusion degree, wherein the type of the occluder is obtained by the occluder type of the matched window occlusion template, and the occlusion degree of the occluder is represented by the transmittance corresponding to the matched window occlusion template.

[0162] As an implementation method, in the embodiment of the present application, the matching score can also be used to I and matching score П Determine the type of obstruction in the window occlusion area and its degree of occlusion, wherein the type of obstruction is obtained by the type of obstruction corresponding to the matching window occlusion template and the matching obstruction flow rate, and the degree of occlusion is represented by the transmittance corresponding to the matching window occlusion template.

[0163] The obstruction type obtained through window occlusion template matching and obstruction flow rate matching is more accurate. Furthermore, the degree of obstruction can be determined, represented by the transmittance corresponding to the matched window occlusion template. Accurate obstruction type and degree allow LIDAR or MDC to determine the appropriate cleaning method. For example, if the obstruction is fog, heating the front window can eliminate the obstruction.

[0164] Combination of the above Figures 1 to 13 The embodiment of the method for detecting window occlusion of the present application is described in detail. Figures 14 and 15 The present invention introduces the device embodiment of the present application. For details not described in detail, please refer to the method embodiment above.

[0165] Figure 14 It is a schematic block diagram of a window occlusion detection device in an embodiment of the present application.

[0166] Figure 14 The window occlusion detection device 1400 can be used to implement Figures 1 to 13 To avoid repetition, the corresponding functions in each embodiment will not be described in detail. Figure 14 The window occlusion detection apparatus 1400 may include an acquisition unit 1401 , a first processing unit 1402 , and a second processing unit 1403 . Optionally, the window occlusion detection apparatus 1400 further includes an output unit 1404 .

[0167] Acquisition unit 1401 is configured to acquire a first window echo intensity and a second window echo intensity, where the first window echo intensity represents the reference window echo intensity, and the second window echo intensity represents the window echo intensity at the current moment. Acquisition unit 1401 is also configured to acquire detection parameters of a target point at the first moment and detection parameters at the second moment, where the detection parameters include the reflectivity and / or position coordinates of the target point.

[0168] The first processing unit 1402 is configured to obtain a window occlusion parameter according to the first window echo intensity and the second window echo intensity, as well as the detection parameters at the first moment and the detection parameters at the second moment, where the window occlusion parameter is used to indicate a window occlusion state.

[0169] Optionally, as an embodiment, the first processing unit 1402 is further configured to: obtain the window occlusion parameter based on a first variation parameter and a second variation parameter; wherein the first variation parameter includes a difference between the second window echo intensity and the first window echo intensity, or a ratio between the second window echo intensity and the first window echo intensity. The second variation parameter represents a change between the reflectivity of the target point at the first moment and the reflectivity at the second moment, or a change between the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment, wherein the estimated value of the position coordinate at the first moment is obtained based on the position coordinate at the second moment.

[0170] Optionally, as an embodiment, the window occlusion detection apparatus 1400 further includes an output unit 1404. The output unit 1404 is configured to output at least one of an alarm message, a first change parameter, and a window occlusion parameter. The alarm message is used to indicate the window occlusion status, and the first change parameter corresponds to an occlusion classification and / or occlusion partition.

[0171] Optionally, as an embodiment, the first variation parameter includes a first element and a second element. The first element indicates that p elements in the first variation parameter are greater than a first threshold value, where p is a positive integer. The second element indicates that q elements in the first variation parameter are less than or equal to the first threshold value. Optionally, the value of the second element can be 0, where q is a positive integer.

[0172] When acquiring a window occlusion parameter, the acquisition unit 1401 may determine the window occlusion parameter based on a weighted sum of values ​​of at least one fourth variation parameter. The fourth variation parameter includes the second element, the third element, and the fourth element; the first element is checked according to the second variation parameter to obtain the third element and the fourth element; the third element represents n times r elements in the first element, the r elements representing that the value of the second variation parameter corresponding to the r elements is less than or equal to the second threshold, or the r elements representing that the value of the second variation parameter corresponding to the r elements is less than or equal to the third threshold, wherein r is a positive integer and n is a positive real number less than 1 or 0; the fourth element represents s elements in the first element, the s elements representing that the value of the second variation parameter corresponding to the s elements is greater than the second threshold, or the s elements representing that the value of the second variation parameter corresponding to the s elements is greater than the third threshold, wherein s is a positive integer.

[0173] Optionally, as an embodiment, the first processing unit 1402 is further configured to determine whether to output warning information according to the window occlusion parameter and occlusion weights of different areas of the window.

[0174] Optionally, as an embodiment, the occlusion weights of different areas of the window include: the first processing unit determines the occlusion weights of different areas of the window according to relative positions of different areas of the window and a movement direction of the mobile device.

[0175] Optionally, as an embodiment, the first processing unit 1402 is also used to determine an alarm coefficient, and the determination of whether to output the alarm information includes: if the window occlusion alarm coefficient is greater than the occlusion alarm threshold, the alarm instruction is executed; or if the window occlusion alarm coefficient is less than or equal to the occlusion alarm threshold, the alarm instruction is not executed; wherein, the occlusion alarm threshold is determined according to the application scenario.

[0176] Optionally, as an embodiment, the second processing unit is used to: determine the lidar window obstruction type based on the distribution vector of the first change parameter and the window occlusion template, or determine the lidar window obstruction type based on the distribution vector of the fourth change parameter and the window occlusion template, wherein the lidar window obstruction type includes the obstruction type and the obstruction degree of the obstruction.

[0177] It should be understood that the second processing unit 1403 can represent the first processing unit 1402 of the laser radar internal window detection method device, and can also represent the processing unit in the MDC. The embodiment of the present application does not limit this.

[0178] Optionally, as an embodiment, the window occlusion template is determined based on a third variation parameter of different types of occlusion degrees, the window transmittance and the flow velocity of the occlusion object, wherein the window transmittance represents the occlusion degree of the lidar window occlusion object type.

[0179] Optionally, as an embodiment, the determining of the type of obstruction of the laser radar window includes: the second processing unit 1403 is further configured to: Ι , determine the type of obstruction of the laser radar window, where the matching score I Determined based on the distribution vector of the first variation parameter and the window occlusion template, or based on the distribution vector of the fourth variation parameter and the window occlusion template; or based on the matching score I and matching score П , determine the type of obstruction of the laser radar window, where the matching score ПIt is determined based on the blocking flow rate obtained by the first change parameter and the blocking flow rate of the window blocking template, or it is determined based on the blocking flow rate obtained by the fourth change parameter and the blocking flow rate of the window blocking template.

[0180] The window occlusion detection device 1400 of the embodiment of the present application can be implemented in any suitable form. In one embodiment, the window occlusion detection device 1400 can be the LIDAR device itself or a component inside the LIDAR device. Optionally, the window occlusion detection device can also be set independently of the LIDAR, for example, the function of the window occlusion detection device is implemented by the MDC. For example, the acquisition unit 1401, the first processing unit 1402 and the second processing unit 1403 can be implemented by the processor of the LIDAR device; the output unit 1404 can be implemented as a communication interface of the LIDAR device. Specifically, for example, when the LIDAR device obtains the first window echo intensity offline, the processor in the LIDAR device can obtain the preset first window echo intensity; or, when the LIDAR device obtains the first window echo intensity online, the processor in the LIDAR device can obtain the first window echo intensity from the LIDAR device (for example Figure 2 The first window echo intensity of the laser detector 202 in the image.

[0181] In another embodiment, the window occlusion detection device 1400 can be implemented as a combination of a LIDAR device and an MDC processor. For example, the first processing unit 1402 can be implemented by a processor in the LIDAR device, and the second processing unit 1403 can be implemented by an MDC processor. The acquisition unit 1401 can also be implemented by a processor in the LIDAR device. Specifically, for example, the above-mentioned occlusion classification step can be performed by the MDC processor, and the above-mentioned occlusion detection step and occlusion alarm step can be performed by the LIDAR processor. The output unit 1404 can be Figure 15 The communication interface 1501 in Figure 15 shown.

[0182] Figure 15 It is a schematic block diagram of a window occlusion detection device in an embodiment of the present application. Figure 15 The illustrated window occlusion detection device may include a communication interface 1501, a processor 1502, and a memory 1503. The memory 1503 is used to store instructions, the processor 1502 is used to execute the instructions stored in the memory 1503, and the communication interface 1501 is used to transmit information. Optionally, the memory 1503 may be coupled to the first processor 1502 via an interface or may be integrated with the first processor 1502.

[0183] During implementation, each step of the above method can be completed by an integrated logic circuit of the hardware in the first processor 1502 or by instructions in the form of software. The method disclosed in conjunction with the embodiments of the present application can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor. The software module can be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 1504, and the first processor 1502 reads the information in the memory 1504 and completes the steps of the above method in combination with its hardware. To avoid repetition, it will not be described in detail here.

[0184] It should be understood that in the embodiments of the present application, the memory may include read-only memory and random access memory, and provide instructions and data to the processor. A portion of the processor may also include non-volatile random access memory. For example, the processor may also store device type information.

[0185] The window occlusion detection device can be a vehicle with a window occlusion detection function, or other components with a window occlusion detection function. The window occlusion detection device includes, but is not limited to, an onboard terminal, an onboard controller, an onboard module, an onboard component, an onboard chip, an onboard unit, an onboard radar, or an onboard camera, and other sensors. The vehicle can implement the method provided in this application through the onboard terminal, onboard controller, onboard module, onboard component, onboard chip, onboard unit, onboard radar, or camera.

[0186] The window obstruction detection device can also be a smart terminal other than a vehicle with a window obstruction detection function, or be installed in a smart terminal other than a vehicle with a window obstruction detection function, or be installed in a component of such a smart terminal. The smart terminal can be other terminal devices such as smart transportation equipment, smart home appliances, robots, etc. The window obstruction detection device includes but is not limited to the smart terminal or its controller, chip, other sensors such as radar or cameras, and other components.

[0187] The window obstruction detection device can be a general-purpose device or a dedicated device. In a specific implementation, the device can also be a desktop computer, a portable computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, an embedded device, or other device with processing capabilities. The embodiments of the present application do not limit the type of the window obstruction detection device.

[0188] The window occlusion detection device may also be a chip or processor with processing capabilities, and may include multiple processors. The processor may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. The chip or processor with processing capabilities may be located within the sensor, or may be located outside the sensor but at the receiving end of the sensor's output signal.

[0189] An embodiment of the present application also provides a system for use in unmanned driving or intelligent driving, which includes at least one of the window occlusion detection device, camera, lidar and other sensors mentioned in the above embodiments of the present application. At least one device in the system can be integrated into a complete machine or equipment, or at least one device in the system can also be independently set as a component or device.

[0190] Furthermore, any of the above systems may interact with the MDC of the vehicle to provide detection and / or fusion information for decision-making or control of the vehicle driving.

[0191] The present application also provides a terminal comprising at least one of the window occlusion detection devices or any of the aforementioned systems described in the above embodiments of the present application. Furthermore, the terminal may be a vehicle, drone, surveying and mapping device, or robot, etc., on which a lidar can be installed.

[0192] It should be understood that the term "and / or" in this document simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document generally indicates that the related objects are in an "or" relationship.

[0193] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0194] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0195] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0196] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0197] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0198] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0199] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0200] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A window occlusion detection method, characterized in that: include: Acquire a first window echo intensity and a second window echo intensity, wherein the first window echo intensity represents a reference window echo intensity, and the second window echo intensity represents a current window echo intensity; Acquire detection parameters of a target point at a first moment and detection parameters of a target point at a second moment, wherein the detection parameters include a reflectivity and / or a position coordinate of the target point; A window occlusion parameter is acquired according to the first window echo intensity, the second window echo intensity, the detection parameter at the first moment, and the detection parameter at the second moment, where the window occlusion parameter is used to indicate a window occlusion state.

2. The method according to claim 1, characterized in that The obtaining of the window occlusion parameters includes: Obtaining the window occlusion parameter according to the first change parameter and the second change parameter; The first change parameter includes the difference between the second window echo intensity and the first window echo intensity, or the ratio of the second window echo intensity to the first window echo intensity; the second change parameter represents the change between the reflectivity of the target point at the first moment and the reflectivity at the second moment, or the change between the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment. The estimated value of the position coordinate at the first moment is obtained based on the position coordinate at the second moment.

3. The method according to claim 2, characterized in that The method further comprises: At least one of an alarm message, the first change parameter, or the window occlusion parameter is output, the alarm message is used to remind the window occlusion state, and the first change parameter corresponds to an occlusion classification and / or an occlusion partition.

4. The method according to claim 3, wherein: p elements of the first variation parameter that are greater than a first threshold value constitute a first element, where p is a positive integer; q elements of the first variation parameter that are less than or equal to the first threshold value constitute a second element, where the value of the second element is 0, where q is a positive integer; The obtaining of the window occlusion parameters includes: determining the viewport occlusion parameter according to a weighted sum of values ​​of at least one fourth variation parameter, the fourth variation parameter including the second element, the third element, and the fourth element; verifying the first element according to the second variation parameter to obtain the third element and the fourth element; The third element represents n times r elements in the first element, and the r elements represent elements in the second variation parameter corresponding to the r elements that are less than or equal to a second threshold, or the r elements represent elements in the second variation parameter corresponding to the r elements that are less than or equal to a third threshold, where r is a positive integer, and n is a positive real number less than 1 or 0; The fourth element represents s elements in the first elements, and the s elements represent elements in the second change parameters corresponding to the s elements that are greater than the second threshold, or the s elements represent elements in the second change parameters corresponding to the s elements that are greater than the third threshold, where s is a positive integer.

5. The method according to claim 4, characterized in that Also includes: According to the window occlusion parameter and the occlusion weights of different areas of the window, a window occlusion alarm coefficient is obtained and it is determined whether to output alarm information.

6. The method according to claim 5, characterized in that The occlusion weights of different areas of the window include: The occlusion weights of the different areas of the window are determined according to the relative positions of the different areas of the window and the movement direction of the mobile device.

7. The method according to claim 6, characterized in that The method further includes: determining an alarm coefficient, wherein the determining whether to output the alarm information includes: If the window occlusion alarm coefficient is greater than the occlusion alarm threshold, output the alarm information; or If the window occlusion alarm coefficient is less than or equal to the occlusion alarm threshold, the alarm information is not output; wherein the occlusion alarm threshold is determined according to the application scenario.

8. The method according to claim 4, characterized in that Also includes: The laser radar window obstruction type is determined according to the distribution vector of the first change parameter and the window occlusion template, or the laser radar window obstruction type is determined according to the distribution vector of the fourth change parameter and the window occlusion template, wherein the laser radar window obstruction type includes the obstruction type and the obstruction degree.

9. The method according to claim 8, characterized in that The window occlusion template is determined based on the third variation parameter of different types of occlusions and different occlusion degrees, the window transmittance and the flow speed of the occlusion object, wherein the window transmittance represents the occlusion degree of the laser radar window occlusion object type.

10. The method according to claim 8 or 9, characterized in that Determining the type of obstruction of the laser radar window includes: According to the matching score Ι , determine the type of obstruction of the laser radar window, where the matching score I Determined according to the distribution vector of the first variation parameter and the viewport occlusion template, or determined according to the distribution vector of the fourth variation parameter and the viewport occlusion template; Or, according to the matching score I and matching score П , determine the type of obstruction of the laser radar window, where the matching score П It is determined based on the blocking flow rate obtained by the first change parameter and the blocking flow rate of the window blocking template, or it is determined based on the blocking flow rate obtained by the fourth change parameter and the blocking flow rate of the window blocking template.

11. A window occlusion detection device, characterized in that: include: an acquiring unit, configured to acquire a first window echo intensity and a second window echo intensity, wherein the first window echo intensity represents a reference window echo intensity, and the second window echo intensity represents a current window echo intensity; The acquisition unit is configured to acquire detection parameters of a target point at a first moment and a second moment, wherein the detection parameters include a reflectivity and / or a position coordinate of the target point; The first processing unit is configured to obtain a window occlusion parameter according to the first window echo intensity and the second window echo intensity, and the first moment detection parameter and the second moment detection parameter, wherein the window occlusion parameter is used to indicate a window occlusion state.

12. The device according to claim 11, characterized in that The first processing unit is further configured to: Obtaining the window occlusion parameter according to the first change parameter and the second change parameter; The first change parameter includes the difference between the second window echo intensity and the first window echo intensity, or the ratio of the second window echo intensity to the first window echo intensity; the second change parameter represents the change between the reflectivity of the target point at the first moment and the reflectivity at the second moment, or the change between the actual value of the position coordinate of the target point at the first moment and the estimated value of the position coordinate at the first moment. The estimated value of the position coordinate at the first moment is obtained based on the position coordinate at the second moment.

13. The device according to claim 12, characterized in that Also includes: Output unit, The output unit is used to output at least one of an alarm message, the first change parameter and the window occlusion parameter, the alarm message is used to remind the window occlusion state, and the first change parameter corresponds to the occlusion classification and / or occlusion partition.

14. The device according to claim 13, characterized in that p elements of the first variation parameter that are greater than a first threshold value constitute a first element, where p is a positive integer; q elements of the first variation parameter that are less than or equal to the first threshold value constitute a second element, where the value of the second element is 0, where q is a positive integer; The acquisition unit is specifically used to: determining the viewport occlusion parameter according to a weighted sum of values ​​of at least one fourth variation parameter, the fourth variation parameter including the second element, the third element, and the fourth element; verifying the first element according to the second variation parameter to obtain the third element and the fourth element; The third element represents n times r elements in the first element, and the r elements represent elements in the second variation parameter corresponding to the r elements that are less than or equal to the second threshold, or the r elements represent elements in the second variation parameter corresponding to the r elements that are less than or equal to the third threshold, where r is a positive integer, and n is a positive real number less than 1 or 0; The fourth element represents s elements in the first elements, and the s elements represent elements in the second change parameters corresponding to the s elements that are greater than the second threshold, or the s elements represent elements in the second change parameters corresponding to the s elements that are greater than the third threshold, where s is a positive integer.

15. The device according to claim 14, characterized in that The first processing unit is further configured to: According to the window occlusion parameter and the occlusion weights of different areas of the window, a window occlusion alarm coefficient is obtained and it is determined whether to output alarm information.

16. The device according to claim 15, characterized in that The first processing unit is specifically configured to determine occlusion weights of different areas of the window according to relative positions of the different areas of the window and a movement direction of the mobile device.

17. The device according to claim 16, characterized in that The first processing unit is further configured to determine an alarm coefficient. If the window occlusion warning coefficient is greater than the occlusion warning threshold, the first processing unit determines to output the warning information; or If the window occlusion alarm coefficient is less than or equal to the occlusion alarm threshold, the first processing unit determines not to output the alarm information; wherein the occlusion alarm threshold is determined according to an application scenario.

18. The device according to claim 14, characterized in that The device further comprises a second processing unit, The second processing unit is configured to: The laser radar window obstruction type is determined according to the distribution vector of the first change parameter and the window occlusion template, or the laser radar window obstruction type is determined according to the distribution vector of the fourth change parameter and the window occlusion template, wherein the laser radar window obstruction type includes the obstruction type and the obstruction degree.

19. The device according to claim 18, characterized in that The window occlusion template is determined based on the third variation parameter of different types of occlusions and different occlusion degrees, the window transmittance and the flow speed of the occlusion object, wherein the window transmittance represents the occlusion degree of the laser radar window occlusion object type.

20. The device according to claim 18 or 19, characterized in that The second processing unit is further configured to: According to the matching score Ι , determining the laser radar window obstruction type, wherein the matching score score1 is determined according to the distribution vector of the first change parameter and the window obstruction template, or according to the distribution vector of the fourth change parameter and the window obstruction template; Or, according to the matching score score1 and matching score score П , determine the type of obstruction of the laser radar window, where the matching score П It is determined based on the blocking flow rate obtained by the first variation parameter and the blocking flow rate of the window blocking template, or it is determined based on the blocking flow rate obtained by the fourth variation parameter and the blocking flow rate of the window blocking template.

21. A computer-readable medium, characterized in that The method comprises computer instructions, which, when executed on a device, cause the device to perform the method according to any one of claims 1 to 10.

22. A chip, characterized in that: include: At least one processor and an interface circuit, wherein the at least one processor calls a computer program through the interface circuit, so that the device where the chip is located executes the method according to any one of claims 1 to 10.

23. A terminal, characterized in that: The terminal includes the window occlusion detection device according to any one of claims 11 to 20, or includes the computer-readable medium according to claim 21, or includes the chip according to claim 22.

24. The terminal according to claim 23, wherein The terminal is a vehicle, a drone or a robot.

Images