Laser radar shielding risk detection method and detection device, laser radar and medium

Abstract

The invention provides a laser radar shielding risk detection method and device, a laser radar and a medium, the method is applied to the field of laser radar detection, and the method comprises the steps: obtaining a test point cloud; and determining the occlusion risk level of the laser radar according to the point number of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, each preset distance interval being a plurality of distance intervals which are continuous in distance value and determined by taking a preset occlusion distance threshold as an upper limit, different preset distance intervals correspond to different weight coefficients. According to the method, the shielding risk level of the laser radar can be monitored and evaluated, technical personnel can process the shielding condition of the laser radar in time, and normal work of the laser radar is ensured.

Description

Technical Field

[0001] The present application relates to the field of laser radar detection, and more specifically, to a laser radar occlusion risk detection method, a detection device, a laser radar and a medium. Background Art

[0002] At present, with the continuous development of science and technology, in the field of detection, lidar has been widely used in the fields of intelligent robots, unmanned vehicle driving, etc. due to its advantages of high resolution, good concealment and strong anti-interference ability.

[0003] The laser radar transmits a detection signal to the target object. After being reflected by the object, the laser radar receives the reflected echo signal and obtains the detection information of the object based on the detection signal and the echo signal.

[0004] During the detection process of the LiDAR, due to the influence of surrounding environmental factors and weather factors, there may be impurities such as dust, dirt, rain, condensation, etc. in the air, or there may be large obstacles in front of the LiDAR, or the window of the LiDAR may be attached with dust, dirt, rain, leaves, flying insects and other attachments, which may cause the risk of obstruction during LiDAR detection and affect the detection results of the LiDAR.

[0005] Based on this, how to effectively detect whether the LiDAR is at risk of being blocked and ensure the detection performance of the LiDAR has become an urgent problem to be solved. Summary of the invention

[0006] The present application provides a laser radar occlusion risk detection method, detection device, laser radar and medium. The method can monitor and evaluate the occlusion risk level of the laser radar, which is beneficial for technical personnel to deal with the occlusion situation of the laser radar in a timely manner and ensure that the laser radar can work normally.

[0007] In a first aspect, a laser radar occlusion risk detection method is provided, the method comprising: obtaining a test point cloud; and determining the occlusion risk level of the laser radar based on the number of points of the test point cloud in each preset distance interval and a weight coefficient corresponding to the preset distance interval, wherein each of the preset distance intervals is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals have different corresponding weight coefficients.

[0008] In the above technical solution, a laser radar occlusion risk detection method is proposed, which specifically divides the detection range with a preset occlusion distance threshold as the upper limit into multiple preset distance intervals. After obtaining the test point cloud, the occlusion risk level of the laser radar is determined by counting the points of the test point cloud in each preset distance interval and combining the weight coefficient corresponding to each preset distance interval. Among them, the weight coefficients corresponding to different preset distance intervals are different. The above process can take into account the influence of distance on occlusion risk when evaluating the occlusion risk level of the laser radar, so that the determined occlusion risk level is more accurate. For example, the greater the distance of the point in the test point cloud, the farther the point is from the laser radar, and the weaker the corresponding occlusion risk. In addition, by detecting the occlusion risk level of the laser radar, the present application can timely determine the fault state of the laser radar, so as to facilitate alarm and fault handling, and ensure that the laser radar obtains higher detection accuracy and more accurate detection results during the actual detection process.

[0009] In combination with the first aspect, in some possible implementations, the detection field of view of the laser radar includes multiple sub-detection areas arranged along a first preset direction, and the points of the test point cloud in each sub-detection area constitute a point cloud subset. The occlusion risk level of the laser radar is determined according to the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, including: determining the occlusion risk level of the laser radar in the sub-detection area according to the number of points of each point cloud subset in each of the preset distance interval and the weight coefficient corresponding to the preset distance interval; determining the occlusion risk level of the laser radar according to the occlusion risk level of the laser radar in each of the sub-detection areas.

[0010] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level of the laser radar in the sub-detection area is determined based on the number of points of each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval, including: determining a first ratio between the number of points of the point cloud subset in each preset distance interval and the theoretical total number of points of the point cloud subset; determining the occlusion risk level of the sub-detection area based on the first ratio corresponding to the point cloud subset in each preset distance interval and the corresponding weight coefficient.

[0011] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level of the sub-detection area is determined based on the first ratio corresponding to the point cloud subset in each preset distance interval and the corresponding weight coefficient, including: performing weighted summation based on multiple first ratios and corresponding multiple weight coefficients to obtain a risk reference value; determining the occlusion risk level of the sub-detection area based on the risk reference value and the multiple weight coefficients.

[0012] In combination with the first aspect and the above-mentioned implementation method, in some possible implementation methods, the number of the preset distance intervals and the weight coefficients are both four, and the four weight coefficients are, from small to large, the first weight coefficient, the second weight coefficient, the third weight coefficient, and the fourth weight coefficient. The weight coefficient corresponding to the preset distance interval with a larger upper limit value is smaller, and the occlusion risk level of the sub-detection area is determined according to the risk reference value and the multiple weight coefficients, including: if the risk reference value is less than a first preset threshold value, it is determined that the sub-detection area is not obscured, wherein the first preset threshold value is less than the first weight coefficient; if the risk reference value is greater than or equal to the first preset threshold value and is less than the first weight coefficient, then the occlusion risk level of the sub-detection area is determined. Level 1 risk; if the risk reference value is greater than or equal to the first weight coefficient and less than the second weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 2 risk; if the risk reference value is greater than or equal to the second weight coefficient and less than the third weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 3 risk; if the risk reference value is greater than or equal to the third weight coefficient and less than the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 4 risk; if the risk reference value is equal to the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 5 risk; wherein the occlusion risk levels of the level 1 risk, the level 2 risk, the level 3 risk, the level 4 risk and the level 5 risk increase in sequence.

[0013] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, if the laser radar has a risk of being blocked in the sub-detection area, the method also includes: in a continuous multi-frame test point cloud, the ratio between the first ratios corresponding to the preset distance intervals with occlusion risk in different frames of test point clouds in the sub-detection area with occlusion risk is a second ratio, and if the second ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the ratio of the risk reference values ​​corresponding to different frames of test point clouds in the sub-detection area with occlusion risk is a third ratio, and if the third ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frames of test point clouds in the sub-detection area with occlusion risk are the same, then it is determined that the laser radar is blocked in the sub-detection area.

[0014] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level of the laser radar is determined based on the occlusion risk level of the laser radar in each sub-detection area, including: determining the highest level of the occlusion risk levels of each sub-detection area as the occlusion risk level of the laser radar.

[0015] In combination with the first aspect and the above-mentioned implementation manner, in some possible implementation manners, the method further includes: if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, generating a target fault code.

[0016] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, after the step of generating a target fault code if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, the method also includes: if in a continuous multi-frame test point cloud, the first ratios of different frame test point clouds in the target sub-detection area and the target distance interval are all lower than the second preset threshold, a reset fault code of the target fault code is generated; or, if in a continuous multi-frame test point cloud, the risk reference values ​​of different frame test point clouds in the target sub-detection area are all lower than the third preset threshold, a reset fault code of the target fault code is generated; or, if in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frame test point clouds in the target sub-detection area are all lower than the first preset risk level, a reset fault code of the target fault code is generated; wherein, the target sub-detection area is a sub-detection area whose occlusion risk level was previously higher than or equal to the first preset risk level, and the target distance interval is a preset distance interval whose corresponding first ratio was previously higher than the preset threshold.

[0017] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the multiple sub-detection areas include areas of interest and areas of no interest, and the method also includes: if the occlusion risk level of the area of ​​interest is higher than or equal to the second preset risk level, generating a target fault code; and if the occlusion risk level of the area of ​​no interest is higher than or equal to the third preset risk level, generating a target fault code, wherein the risk level corresponding to the third preset risk level is higher than the risk level corresponding to the second preset risk level.

[0018] In a second aspect, a laser radar occlusion risk detection device is provided, which includes: a point cloud acquisition module for acquiring a test point cloud; an occlusion risk level determination module for determining the occlusion risk level of the laser radar based on the number of points of the test point cloud in each preset distance interval and a weight coefficient corresponding to the preset distance interval, wherein each of the preset distance intervals is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals correspond to different weight coefficients.

[0019] In combination with the second aspect, in some possible implementations, the detection field of view of the laser radar includes multiple sub-detection areas arranged along a first preset direction, and the points of the test point cloud in each sub-detection area constitute a point cloud subset. The occlusion risk level determination module is specifically used to: determine the occlusion risk level of the laser radar in the sub-detection area according to the number of points in each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval; determine the occlusion risk level of the laser radar according to the occlusion risk level of the laser radar in each sub-detection area.

[0020] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level determination module is also used to: determine a first ratio between the number of points of the point cloud subset in each of the preset distance intervals and the theoretical total number of points of the point cloud subset; determine the occlusion risk level of the sub-detection area based on the first ratio corresponding to the point cloud subset in each of the preset distance intervals and the corresponding weight coefficient.

[0021] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level determination module is also used to: obtain a risk reference value based on a weighted sum of multiple first ratios and corresponding multiple weight coefficients; and determine the occlusion risk level of the sub-detection area based on the risk reference value and the multiple weight coefficients.

[0022] In combination with the second aspect and the above-mentioned implementation method, in some possible implementation methods, the number of the preset distance intervals and the weight coefficients are both four, and the four weight coefficients are, from small to large, the first weight coefficient, the second weight coefficient, the third weight coefficient, and the fourth weight coefficient. The preset distance interval with a larger upper limit value has a smaller weight coefficient. The occlusion risk level determination module is also used to: if the risk reference value is less than a first preset threshold value, determine that the sub-detection area is not obstructed, wherein the first preset threshold value is less than the first weight coefficient; if the risk reference value is greater than or equal to the first preset threshold value and less than the first weight coefficient, determine that the occlusion risk level of the sub-detection area is a level one risk; if the risk reference value is greater than or equal to the first preset threshold value and less than the first weight coefficient, determine that the occlusion risk level of the sub-detection area is a level one risk; if the risk reference value is greater than or equal to the first preset threshold value and less than the first weight coefficient, determine that the occlusion risk level of the sub-detection area is a level one risk; If the risk reference value is greater than or equal to the first weight coefficient and less than the second weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 2 risk; if the risk reference value is greater than or equal to the second weight coefficient and less than the third weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 3 risk; if the risk reference value is greater than or equal to the third weight coefficient and less than the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 4 risk; if the risk reference value is equal to the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 5 risk; wherein the occlusion risk levels of the level 1 risk, the level 2 risk, the level 3 risk, the level 4 risk, and the level 5 risk increase in sequence.

[0023] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, if the laser radar has a risk of being blocked in the sub-detection area, the device also includes: an occlusion area determination module, which is used to determine that in a continuous multi-frame test point cloud, the ratio between the first ratios corresponding to the preset distance interval with occlusion risk of different frames of test point clouds in the sub-detection area with occlusion risk is a second ratio, and if the second ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the ratio of the risk reference values ​​corresponding to different frames of test point clouds in the sub-detection area with occlusion risk is a third ratio, if the third ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frames of test point clouds in the sub-detection area with occlusion risk are the same, then it is determined that the laser radar is blocked in the sub-detection area.

[0024] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the occlusion risk level determination module is also used to: determine the highest level of the occlusion risk levels of each sub-detection area as the occlusion risk level of the laser radar.

[0025] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the device also includes: a first fault code generation module, which is used to generate a target fault code if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to a first preset risk level.

[0026] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, after the step of generating a target fault code if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, the device also includes: a reset fault code generation module, which is used to generate a reset fault code for the target fault code if, in a continuous multi-frame test point cloud, the first ratios of different frame test point clouds in the target sub-detection area and the target distance interval are all lower than the second preset threshold; or, if, in a continuous multi-frame test point cloud, the risk reference values ​​of different frame test point clouds in the target sub-detection area are all lower than the third preset threshold, a reset fault code for the target fault code is generated; or, if, in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frame test point clouds in the target sub-detection area are all lower than the first preset risk level, a reset fault code for the target fault code is generated; wherein, the target sub-detection area is a sub-detection area whose occlusion risk level was previously higher than or equal to the first preset risk level, and the target distance interval is a preset distance interval whose corresponding first ratio was previously higher than the preset threshold.

[0027] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the multiple sub-detection areas include areas of interest and areas of no interest, and the device also includes: a second fault code generation module, for generating a target fault code if the occlusion risk level of the area of ​​interest is higher than or equal to the second preset risk level; and, for generating a target fault code if the occlusion risk level of the area of ​​no interest is higher than or equal to the third preset risk level, wherein the risk level corresponding to the third preset risk level is higher than the risk level corresponding to the second preset risk level.

[0028] In a third aspect, a laser radar is provided, comprising a memory and a processor. The memory is used to store executable program code, and the processor is used to call and run the executable program code from the memory, so that the laser radar executes the method in the first aspect or any possible implementation of the first aspect.

[0029] 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 method in the first aspect or any possible implementation of the first aspect.

[0030] In a fifth aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation manner of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a structural schematic diagram of a laser radar risk detection system provided in an embodiment of the present application;

[0032] Figure 2 is a schematic flow chart of a laser radar risk detection method provided in an embodiment of the present application;

[0033] Figure 3 This is a schematic diagram of a scenario in which a plurality of sub-detection areas are divided, provided in an embodiment of the present application;

[0034] Figure 4 It is a schematic diagram of a scenario for counting the number of points in a test point cloud within a plurality of preset distance intervals provided in an embodiment of the present application;

[0035] Figure 5 This is another schematic diagram of a scenario in which a plurality of sub-detection areas are divided, provided in an embodiment of the present application;

[0036] Figure 6 It is a structural schematic diagram of a laser radar risk detection device provided in an embodiment of the present application;

[0037] Figure 7 It is a structural schematic diagram of a laser radar provided in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The technical solution in the present application will be described clearly and in detail below in conjunction with the accompanying drawings. In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, "multiple" means two or more than two.

[0039] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features.

[0040] Before introducing the method of the embodiment of the present application, the application scenario of the embodiment of the present application is first introduced.

[0041] At present, in the field of detection, LiDAR is widely used in vehicle detection scenarios due to its advantages of high resolution, good concealment, and strong anti-interference ability. During the detection process of LiDAR, affected by the surrounding environmental factors and weather factors, there may be impurities such as dust, dirt, rain, condensation, etc. in the air, or there may be large obstacles in front of the LiDAR, or the window of the LiDAR may be attached with dust, dirt, rain, leaves, flying insects, etc., which may cause the risk of LiDAR being blocked during detection, making it impossible for LiDAR to effectively detect the target objects within the detection range, which may cause an accident to the vehicle equipped with the LiDAR.

[0042] Based on this, how to timely identify whether the lidar is at risk of being blocked has become the focus of researchers.

[0043] When detecting whether the laser radar is blocked, the related technology mainly detects the data collected by the sensor. For example, the related technology installs various sensors such as pressure sensors, rain sensors, humidity sensors or cameras on the laser radar window to identify whether there are obstacles blocking the laser radar window surface or in front of the window. Specifically, the laser radar can detect the pressure change on the surface of the laser radar window through a pressure sensor to determine whether there are obstacles attached to the surface of the window; or, detect the moisture on the surface of the laser radar window through a rain sensor to determine whether there are obstacles attached to the surface of the window; or, detect the relative humidity near the laser radar window through a humidity sensor to determine whether there are obstacles on the surface of the window; or, determine whether there are obstacles on the surface of the window or in front of the window through the image collected by the camera.

[0044] However, the above detection scheme requires the installation of various sensors on the laser radar window. Since the sensors themselves occupy a certain amount of space, the hardware cost of the laser radar will increase, and the light transmittance of the laser radar will also be affected. For the pressure sensor, rain sensor and humidity sensor, it is only when the obstacle is attached to the window that the laser radar can be determined to be blocked. It is impossible to predict whether the laser radar is at risk of being blocked outside the window, resulting in a lag in the laser radar's blockage detection results.

[0045] Based on the above problems, an embodiment of the present application provides a laser radar occlusion risk detection method. This method can monitor and evaluate the occlusion risk level of the laser radar in real time before or during the operation of the laser radar without increasing the hardware cost of the laser radar, which is beneficial for technicians to deal with the occlusion situation of the laser radar in time and ensure that the laser radar can work normally.

[0046] Before introducing the method of the embodiment of the present application, the hardware structure and the working principle of the hardware structure on which the method provided by the embodiment of the present application relies are first introduced in detail.

[0047] Figure 1 It is a structural schematic diagram of a laser radar occlusion risk detection system provided in an embodiment of the present application.

[0048] For example, Figure 1 As shown, a laser radar occlusion risk detection system 100 provided in an embodiment of the present application includes an object 101, a laser radar 102 and a host computer 103.

[0049] In a possible implementation, the laser radar 102 includes a photoelectric receiving device, a digital-to-analog converter, a laser device, and a field-programmable gate array (FPGA). The FPGA further includes a signal processing module such as a detection module, a distance correction module, a transmission module, and a distance segment detection module.

[0050] Based on the structure of the above system 100, the working process of the system 100 is as follows:

[0051] During the detection process, the laser radar 102 generates a trigger signal through the transmitting module in the FPGA, and the trigger signal is used to instruct the laser device to emit a laser pulse at a specific time, and accurately control the emission time. The laser device emits the laser pulse to the object 101, and after being reflected by the object 101, an echo signal is obtained.

[0052] The photoelectric receiving device (also called a photodetector) is used to capture the echo signal after the laser pulse is reflected by the object 101, and convert the signal type of the echo signal from an optical signal to an electrical signal, and then send it to the analog-to-digital converter (ADC). After receiving the electrical signal, the ADC converts the electrical signal from analog form to digital form, and then sends it to the signal processing module such as the detection module in the FPGA.

[0053] The signal processing modules such as the detection module are used to extract the information required for the actual detection process from the digital signal. For example, after determining the information of the echo signal, the signal processing modules such as the detection module can determine the distance between the object 101 and the laser radar 102, the height of the object 101, the reflectivity of the object 101, etc. according to the time difference between the laser pulse emission time and the reception time.

[0054] In one possible implementation, the signals generated by the transmitting module include two types, namely a trigger signal and a reference signal. The trigger signal is used to control the high-intensity light pulse (i.e., the above-mentioned detection signal) emitted by the laser device to illuminate the object 101. After the detection signal hits the object 101 and is reflected by the object 101, an echo signal is formed, which is received by the photoelectric receiving device, and the detection information of the object 101 is obtained after processing. The propagation path of the detection signal is laser device-object 101-photoelectric receiving device.

[0055] The reference signal is a signal propagating inside the laser radar 102, and its purpose is to provide a reference benchmark for the propagation time of the signal inside the circuit board to help correct and improve the accuracy of distance measurement. The propagation path of the reference signal is known and fixed, usually the circuit path inside the laser radar 102. Based on the above echo signal and reference signal, the FPGA can obtain the target ranging and reference ranging in a one-to-one correspondence. The target ranging refers to the distance calculated from the time when the trigger signal is generated and the time when the echo signal is received. The reference ranging refers to the distance calculated from the time when the reference signal is transmitted and the time when the reference signal is received, which is the distance that the reference signal propagates inside the laser radar 102.

[0056] Signal processing modules such as the detection module can obtain the actual distance of the object 101 relative to the laser radar 102 based on the difference between the target distance measurement and the reference distance measurement, so as to eliminate the distance error caused by the signal propagating inside the laser radar 102.

[0057] After obtaining the target distance measurement and the reference distance measurement (or obtaining the actual distance of the object 101), the signal processing modules such as the detection module can divide the obtained data into two paths, each of which includes data representing the distance information of the object 101, such as including the above-mentioned target distance measurement and reference distance measurement (or the actual distance of the object 101 relative to the laser radar 102); one of the data can be transmitted to the distance correction module for more precise calibration processing, so as to obtain more accurate distance information of the object 101, and obtain accurate actual point cloud data based on this data; the other data does not need to be more precisely calibrated, and it can be transmitted to the distance segmentation detection module as the test point cloud data used in this detection method, and then obtain the risk level of the laser radar 102 being blocked. Both the actual point cloud data obtained based on the distance correction module and the risk level information obtained based on the distance segmentation detection module can be sent to the host computer, such as the domain control module of the car.

[0058] After introducing the system structure of the embodiment of the present application, a laser radar occlusion risk detection method provided by the embodiment of the present application is introduced below. It should be understood that the method can be implemented by relying on a computer program and can be run on a laser radar occlusion risk detection device based on the von Neumann system. The computer program can be integrated into an application or run as an independent tool application. Among them, the laser radar occlusion risk detection device in the embodiment of the present application can be any device that adopts the laser radar occlusion risk detection method, including but not limited to: laser radar, vehicle-mounted equipment, airplanes, trains, handheld devices, wearable devices, computing devices, or other processing devices connected to wireless modems. In the following introduction of the embodiment of the present application, the device that executes the method is a laser radar as an example, and the occlusion risk detection process of the embodiment of the present application is introduced in detail.

[0059] Figure 2 It is a schematic flowchart of a laser radar occlusion risk detection method provided in an embodiment of the present application.

[0060] For example, Figure 2 As shown, the method S200 includes the following S201-S202:

[0061] S201, obtaining a test point cloud.

[0062] When performing occlusion risk detection on the laser radar, the laser radar can continuously emit multiple laser pulses within a scanning cycle based on its own scanning cycle. After each laser pulse is emitted, it will be reflected by the object and an echo signal will be obtained. During this process, the laser radar will record the emission time of each laser pulse and the reception time of the echo signal after each laser pulse is reflected by the object, so as to calculate the target distance between each point on the object irradiated by the laser pulse and the laser radar. Based on the target distance and the corresponding reference distance, the actual distance between each point on the object irradiated by the laser pulse and the laser radar can be obtained. In addition, the laser radar can also determine the three-dimensional information of the scanning point corresponding to each laser pulse in a scanning cycle by combining the actual distance and the corresponding direction angle corresponding to each laser pulse, that is, a frame of test point cloud can be obtained.

[0063] It should be noted that the "test point cloud" described in the embodiments of the present application may be different from the actual point cloud data described above. Specifically, since the "test point cloud" is intended to detect the occlusion risk level of the laser radar, it does not require excessively high distance accuracy. It can be directly derived based on the above-mentioned target ranging and reference ranging without the need for additional fine calibration processing. In contrast, the actual point cloud data, after the target ranging and the reference ranging are processed for difference, also needs to be calibrated in combination with the laser radar's own parameters to ensure the ranging accuracy of the actual point cloud data; in addition, the actual point cloud data can also be integrated with information such as the reflectivity information and color information of the target object; the actual point cloud data is the point cloud data actually used by the laser radar.

[0064] Optionally, the correction processing of the above-mentioned distance correction module includes but is not limited to compensating for temperature changes of internal components of the laser radar, compensating for fluctuations in power supply voltage, reflectivity correction, geometric correction and system calibration of the laser radar, etc.

[0065] S202, determining the occlusion risk level of the laser radar based on the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, wherein each preset distance interval is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals correspond to different weight coefficients.

[0066] In the embodiment of the present application, a segmented detection strategy is provided when determining the occlusion risk level of the laser radar. Specifically, the technician can determine the critical occlusion distance to be detected, that is, the preset occlusion distance threshold, according to the actual occlusion detection requirements. For example, if the technician believes that objects within 2m from the laser radar have an occlusion risk, the preset occlusion distance threshold can be set to 2m; for another example, if the technician believes that objects within 1m from the laser radar have an occlusion risk, the preset occlusion distance threshold can be set to 1m. Further, the laser radar can be divided into multiple preset distance intervals that are continuous in distance value based on the sensitivity of the occlusion distance, with the preset occlusion distance threshold as the upper limit. For example, if the technician sets the preset occlusion distance threshold to 65cm, 65cm can be used as the upper limit, and according to the sensitivity of the occlusion distance, multiple preset distance intervals are divided, namely: 0-15cm, 15-25cm, 25-45cm, and 45-65cm. It should be noted that the "multiple preset distance intervals that are continuous in distance value" mentioned in the present application document means that when the multiple preset distance intervals obtained by division are arranged according to the size of the distance value (from large to small, or from small to large), between two adjacent preset distance intervals, the interval upper limit value of the preset distance interval with a smaller intermediate value is equal to the interval lower limit value of the preset distance interval with a larger intermediate value.

[0067] According to the different distances, each preset distance interval occupies a different weight coefficient when the laser radar determines the occlusion risk level; generally speaking, the closer the target object is to the laser radar, the higher the occlusion risk, so the weight coefficient can be set to be higher, that is, the smaller the upper limit value of the preset distance interval, the higher the weight coefficient corresponding to. After obtaining the test point cloud, the laser radar can count the number of points in the test point cloud in each preset distance interval. The number of points in the test point cloud in each preset distance interval refers to the number of points in the test point cloud whose distance information carried by the points in the test point cloud is within each preset distance interval. Then, the laser radar can determine the occlusion risk level based on the number of points in the test point cloud in each preset distance interval and the weight coefficient corresponding to each preset distance interval.

[0068] In one possible implementation, when implementing the above-mentioned step of determining the occlusion risk level, the detection field of view of the laser radar can be divided into multiple sub-detection areas arranged along a first preset direction; for example, for a laser radar with a larger horizontal field of view, the first preset direction can be a horizontal direction; of course, in other embodiments, the first preset direction can also be a vertical direction. Correspondingly, the points of the test point cloud that fall in each sub-detection area can be regarded as a "point cloud subset". The laser radar can first determine the occlusion risk level of each sub-detection area based on the point cloud subset, and then determine the occlusion risk level of the laser radar based on the occlusion risk level of each sub-detection area. The above-mentioned step S202 specifically includes:

[0069] Determine the occlusion risk level of the laser radar in the sub-detection area according to the number of points in each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval;

[0070] The occlusion risk level of the laser radar is determined according to the occlusion risk level of the laser radar in each sub-detection area.

[0071] Figure 3 This is a scene diagram of a laser radar detection field of view division provided in an embodiment of the present application.

[0072] For example, Figure 3 As shown, the projection of the laser radar detection field of view along the depth direction is a rectangle, and the first preset direction is the horizontal direction. For the laser radar, the detection field of view is actually a three-dimensional area composed of a horizontal field of view (FOV) and a vertical FOV. When the first preset direction is the horizontal direction, the laser radar can divide the detection field of view into multiple sub-detection areas based on the preset horizontal angle. Figure 3 The sub-detection area 1, sub-detection area 2, sub-detection area 3, sub-detection area 4 and sub-detection area 5 are shown. The vertical FOV and horizontal FOV of each sub-detection area are the same.

[0073] The above-mentioned sub-detection area division method is only an illustrative example. In the actual division process, the first preset direction may also be a vertical direction, and / or the area sizes of multiple sub-detection areas may also be different. The embodiment of the present application does not limit this.

[0074] Based on the divided multiple sub-detection areas, the points in the test point cloud that fall in sub-detection area 1 constitute the point cloud subset of sub-detection area 1, recorded as "point cloud subset 1"; the points in the test point cloud that fall in sub-detection area 2 constitute the point cloud subset of sub-detection area 2, recorded as "point cloud subset 2"; the points in the test point cloud that fall in sub-detection area 3 constitute the point cloud subset of sub-detection area 3, recorded as "point cloud subset 3"; the points in the test point cloud that fall in sub-detection area 4 constitute the point cloud subset of sub-detection area 4, recorded as "point cloud subset 4"; the points in the test point cloud that fall in sub-detection area 5 constitute the point cloud subset of sub-detection area 5, recorded as "point cloud subset 5".

[0075] It should be understood that the points in the test point cloud carry not only distance information but also azimuth information, such as horizontal angle information and vertical angle information. The horizontal angle information refers to the horizontal offset angle of the point relative to the central optical axis of the laser radar; the vertical angle information refers to the vertical offset angle of the point relative to the central optical axis of the laser radar.

[0076] When the laser radar's detection field of view is divided into multiple sub-detection areas, each sub-detection area corresponds to a horizontal FOV interval and a vertical FOV interval. When the laser radar determines which sub-detection area a point in the test point cloud falls in, it can compare the horizontal angle of the point with the horizontal FOV interval of each sub-detection area, and compare the vertical angle of the point with the vertical FOV interval of each sub-detection area to determine the sub-detection area where each point is located, and obtain a point cloud subset of each sub-detection area.

[0077] When determining the number of points in each point cloud subset within each preset distance interval, for any sub-detection area, the laser radar can compare the distance information carried by each point in the point cloud subset corresponding to the sub-detection area with multiple preset distance intervals, and determine the preset distance interval in which each point in the point cloud subset corresponding to the sub-detection area is located. Through the above distance comparison, the laser radar can obtain the number of points in each preset distance interval of the point cloud subset of the sub-detection area.

[0078] For example, Figure 3 As shown, the preset blocking distance threshold is 65 cm, and the four preset distance intervals are 0-15 cm, 15-25 cm, 25-45 cm, and 45-65 cm. The sub-detection areas are sub-detection area 1, sub-detection area 2, sub-detection area 3, sub-detection area 4, and sub-detection area 5.

[0079] The total number of test point clouds is N. The lidar determines through distance comparison that the number of points of the test point cloud in sub-detection area 1 is N1, and these points constitute the “point cloud subset 1” of sub-detection area 1; the number of points of the test point cloud in sub-detection area 2 is N2, and these points constitute the “point cloud subset 2” of sub-detection area 2; the number of points of the test point cloud in sub-detection area 3 is N3, and these points constitute the “point cloud subset 3” of sub-detection area 3; the number of points of the test point cloud in sub-detection area 4 is N4, and these points constitute the “point cloud subset 4” of sub-detection area 4; the number of points of the test point cloud in sub-detection area 5 is N5, and these points constitute the “point cloud subset 5” of sub-detection area 5.

[0080] It should be understood that, since not all points included in the test point cloud are necessarily within the four preset distance intervals, N1+N2+N3+N4+N5≤N.

[0081] For sub-detection area 1, the lidar determines that the number of points in point cloud subset 1 between 0 and 15 cm is N11, the number of points in point cloud subset 1 between 15 and 25 cm is N12, the number of points in point cloud subset 1 between 25 and 45 cm is N13, and the number of points in point cloud subset 1 between 45 and 65 cm is N14.

[0082] For sub-detection area 2, the lidar determines that the number of points in point cloud subset 2 at 0-15 cm is N21, the number of points in point cloud subset 2 at 15-25 cm is N22, the number of points in point cloud subset 2 at 25-45 cm is N23, and the number of points in point cloud subset 2 at 45-65 cm is N24.

[0083] For sub-detection area 3, the lidar determines that the number of points in point cloud subset 3 between 0-15 cm is N31, the number of points in point cloud subset 3 between 15-25 cm is N32, the number of points in point cloud subset 3 between 25-45 cm is N33, and the number of points in point cloud subset 3 between 45-65 cm is N34.

[0084] For sub-detection area 4, the lidar determines that the number of points in point cloud subset 4 at 0-15 cm is N41, the number of points in point cloud subset 4 at 15-25 cm is N42, the number of points in point cloud subset 4 at 25-45 cm is N43, and the number of points in point cloud subset 4 at 45-65 cm is N44.

[0085] For sub-detection area 5, the lidar determines that the number of points in point cloud subset 5 at 0-15 cm is N51, the number of points in point cloud subset 5 at 15-25 cm is N52, the number of points in point cloud subset 5 at 25-45 cm is N53, and the number of points in point cloud subset 5 at 45-65 cm is N54.

[0086] As mentioned above, the distance directly determines the occlusion risk level of the LiDAR, so each preset distance interval occupies a different weight coefficient when the LiDAR determines the occlusion risk level. For example, the closer (or smaller) the distance, the greater the risk of the LiDAR being blocked; the farther (or larger) the distance, the smaller the risk of the LiDAR being blocked. The LiDAR can determine the occlusion risk level of the LiDAR in each sub-detection area based on the number of points in each point cloud subset in each preset distance interval and the weight coefficient corresponding to each preset distance interval, and then determine the occlusion risk level of the LiDAR based on the occlusion risk level of the LiDAR in each sub-detection area.

[0087] In a possible implementation, the occlusion risk level of the laser radar in the sub-detection area is determined according to the number of points in each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval, including:

[0088] Determine a first ratio between the number of points of the point cloud subset in each preset distance interval and the theoretical total number of points of the point cloud subset;

[0089] The occlusion risk level of the sub-detection area is determined according to the first ratio corresponding to the point cloud subset in each preset distance interval and the corresponding weight coefficient.

[0090] For lidar, the theoretical total number of points in each point cloud subset in each sub-detection area is determined by the system design parameters of the lidar, specifically, the horizontal resolution, vertical resolution, horizontal FOV, vertical FOV and scanning mode of the lidar, which are inherent parameters of the lidar.

[0091] Based on the number of points in the point cloud subset in each sub-detection area in each preset distance interval and the theoretical total number of points in the point cloud subset in each sub-detection area, the laser radar can calculate multiple first ratios.

[0092] Exemplarily, in combination with the above examples, for sub-detection area 1, the laser radar determines that the number of points in point cloud subset 1 at 0-15cm is N11, the number of points in point cloud subset 1 at 15-25cm is N12, the number of points in point cloud subset 1 at 25-45cm is N13, and the number of points in point cloud subset 1 at 45-65cm is N14. The theoretical total number of points of point cloud subset 1 in sub-detection area 1 is recorded as "M1", then the first ratio between the number of points of point cloud subset 1 in each preset distance interval and the theoretical total number of points of point cloud subset 1 M1 includes N11 / M1, N12 / M1, N13 / M1 and N14 / M1.

[0093] Similarly, the theoretical total number of points of point cloud subset 2 in sub-detection area 2 is recorded as "M2", and the first ratio between the number of points of point cloud subset 2 in each preset distance interval and the theoretical total number of points M2 of point cloud subset 2 includes N21 / M2, N22 / M2, N23 / M2 and N24 / M2.

[0094] The theoretical total number of points of point cloud subset 3 in sub-detection area 3 is recorded as "M3", and the first ratio between the number of points of point cloud subset 3 in each preset distance interval and the theoretical total number of points M3 of point cloud subset 3 includes N31 / M3, N32 / M3, N33 / M3 and N34 / M3.

[0095] The theoretical total number of points of point cloud subset 4 in sub-detection area 4 is recorded as "M4", and the first ratio between the number of points of point cloud subset 4 in each preset distance interval and the theoretical total number of points M4 of point cloud subset 4 includes N41 / M4, N42 / M4, N43 / M4 and N44 / M4.

[0096] The theoretical total number of points of point cloud subset 5 in sub-detection area 5 is recorded as "M5", and the first ratio between the number of points of point cloud subset 5 in each preset distance interval and the theoretical total number of points M5 of point cloud subset 5 includes N51 / M5, N52 / M5, N53 / M5 and N54 / M5.

[0097] After obtaining the first ratio corresponding to each point cloud subset, the laser radar can obtain the occlusion risk level of each sub-detection area based on the first ratio corresponding to each point cloud subset and the weight coefficient of each preset distance interval.

[0098] In a possible implementation, determining the occlusion risk level of the sub-detection area according to the first ratio corresponding to the point cloud subset in each preset distance interval and the corresponding weight coefficient includes:

[0099] Performing weighted summation based on the multiple first ratios and the corresponding multiple weight coefficients to obtain a risk reference value;

[0100] The occlusion risk level of the sub-detection area is determined based on the risk reference value and multiple weight coefficients.

[0101] Specifically, the process of determining the risk reference value of any sub-detection area is shown in the following formula (1).

[0102]

[0103] Wherein, in formula (1):

[0104] F i : Risk reference value of the ith sub-detection area;

[0105] N ij: The number of points in the jth preset distance interval of the point cloud subset i of the i-th sub-detection area, l is the number of preset distance intervals; when the sub-detection area is sub-detection area 1, N ij The value can be N11, N12, N13 and N14. When the sub-detection area is sub-detection area 2, N ij The value can be N21, N22, N23 and N24. When the sub-detection area is sub-detection area 3, N ij The value can be N31, N32, N33 and N34. When the sub-detection area is sub-detection area 4, N ij The value can be N41, N42, N43 and N44. When the sub-detection area is sub-detection area 5, N ij The possible values ​​are N51, N52, N53 and N54.

[0106] w j : The weight coefficient of the preset distance interval j;

[0107] M i : The theoretical total number of points in the ith sub-detection area.

[0108] The laser radar can calculate the risk reference value of each sub-detection area through the above formula (1).

[0109] Figure 4 It is a schematic diagram of a scenario for counting the number of points in a test point cloud within multiple preset distance intervals provided in an embodiment of the present application.

[0110] For example, Figure 4 As shown in the figure, each figure is a statistical graph of the number of points in each preset distance interval of a point cloud subset of a sub-detection area. The horizontal axis represents the distance, and the vertical axis represents the number of points in multiple preset distance intervals of the point cloud subset of each sub-detection area. The distance resolution of the laser radar is 0.5cm, that is, the unit length of the horizontal axis in the figure is 0.5cm.

[0111] In this scene, the preset occlusion distance threshold is 65 cm, and the preset distance intervals are (0 cm, 15 cm], (15 cm, 25 cm], (25 cm, 45 cm], and (45 cm, 65 cm], and the corresponding weight coefficients are 1.2, 1.0, 0.8, and 0.6, respectively.

[0112] Taking the third figure as an example, the theoretical total number of points of the point cloud subset of the sub-detection area corresponding to the third figure is 15750. It can be seen from the figure that the number of points of the point cloud subset of the sub-detection area in (15cm, 25cm] is 5870, the number of points of the point cloud subset of the sub-detection area in (25cm, 45cm] is 6130, and the number of points of the point cloud subset of the sub-detection area in the other two preset distance intervals is 0.

[0113] Based on formula (1), the risk reference value of the sub-detection area can be obtained as:

[0114]

[0115] After obtaining the risk reference value of each sub-detection area, the laser radar can determine the occlusion risk level of the sub-detection area according to the risk reference value of each sub-detection area and the weight coefficient of each preset distance interval.

[0116] In a possible implementation, the number of preset distance intervals and weight coefficients are both four, and the four weight coefficients are, from small to large, a first weight coefficient, a second weight coefficient, a third weight coefficient, and a fourth weight coefficient. The larger the upper limit value of the preset distance interval, the smaller the weight coefficient corresponding to the preset distance interval. According to the risk reference value and the multiple weight coefficients, the occlusion risk level of the sub-detection area is determined, including:

[0117] If the risk reference value is less than a first preset threshold, determining that the sub-detection area is not blocked, wherein the first preset threshold is less than a first weight coefficient;

[0118] If the risk reference value is greater than or equal to the first preset threshold value and less than the first weight coefficient, the occlusion risk level of the sub-detection area is determined to be a first-level risk;

[0119] If the risk reference value is greater than or equal to the first weight coefficient and less than the second weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 2 risk;

[0120] If the risk reference value is greater than or equal to the second weight coefficient and less than the third weight coefficient, the occlusion risk level of the sub-detection area is determined to be a third-level risk;

[0121] If the risk reference value is greater than or equal to the third weight coefficient and less than the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a level 4 risk;

[0122] If the risk reference value is equal to the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be level five risk;

[0123] Among them, the obstruction risk levels of first-level risk, second-level risk, third-level risk, fourth-level risk and fifth-level risk increase successively.

[0124] In the embodiment of the present application, the number of preset distance intervals is specifically 4, and the weight coefficients corresponding to the preset distance intervals are also 4. The four weight coefficients are recorded in order from small to large as "the first weight coefficient (w1), the second weight coefficient (w2), the third weight coefficient (w3) and the fourth weight coefficient (w4)". Combined with the above analysis, the greater the distance, the smaller the risk of the laser radar being blocked, and the smaller the corresponding weight coefficient. Therefore, the larger the upper limit value of the preset distance interval, the smaller the corresponding weight coefficient.

[0125] Exemplarily, the four preset distance intervals are recorded as R1, R2, R3 and R4 in order from the smallest to the largest upper limit value. Then the weight coefficient corresponding to R1 is the fourth weight coefficient (w4), the weight coefficient corresponding to R2 is the third weight coefficient (w3), the weight coefficient corresponding to R3 is the second weight coefficient (w2), and the weight coefficient corresponding to R4 is the first weight coefficient (w1).

[0126] Optionally, in the embodiment of the present application, the four preset distance intervals R1, R2, R3, and R4 are 0-15 cm, 15-25 cm, 25-45 cm, and 45-65 cm, respectively. The fourth weight coefficient corresponding to R1 is 1.2, the third weight coefficient corresponding to R2 is 1.0, the second weight coefficient corresponding to R3 is 0.8, and the first weight coefficient corresponding to R4 is 0.6.

[0127] When the risk reference value F of each sub-detection area is calculated by formula (1), i Afterwards, the laser radar can calculate the risk reference value F i Compare with the four weight coefficients to determine the occlusion risk level of each sub-detection area. i Compared with the four weight coefficients, the occlusion risk level of the sub-detection area may be as follows:

[0128] (1) Risk reference value F i is less than the first preset threshold, and the first preset threshold is less than the first weight coefficient. Optionally, the first preset threshold thr1 is 0.1. Since the weight coefficient of the preset distance interval with a smaller distance is larger, combined with the above formula (1), it can be seen that when the number of points in the point cloud subset in the sub-detection area is larger and the distance is smaller, the risk reference value F obtained after multiplication is larger. i On the contrary, when the number of points in the sub-detection area where the point cloud subset is in a larger preset distance interval is greater, the risk reference value F i When the risk reference value F iWhen it is less than the first preset threshold value which is smaller than the first weight coefficient, it means that the distance of most points in the point cloud subset of the sub-detection area is greater than the upper limit distance of R4, that is, greater than the preset occlusion distance threshold. Since the distance of these points is large, there will be no occlusion risk for the laser radar, so the laser radar can determine that the sub-detection area is not blocked.

[0129] (2) When the risk reference value F i In the interval [thr1, w1), it means that most of the points in the point cloud subset of the sub-detection area are in the R4 interval, and the remaining few points fall in other preset distance intervals. The target objects corresponding to a small number of points are not likely to cause occlusion to the laser radar. Therefore, the laser radar can determine that there is a slight risk of occlusion in the sub-detection area, and the corresponding occlusion risk level is level 1 risk.

[0130] (3) When the risk reference value F i In the interval [w1, w2), it means that most of the points in the point cloud subset of the sub-detection area fall in the R3 interval, and a small number of points fall in other preset distance intervals. The target objects corresponding to a small number of points are not likely to cause occlusion to the laser radar. Therefore, the laser radar can determine that the occlusion risk level of the sub-detection area is a secondary risk. Among them, the risk level of the secondary risk is higher than the risk level of the primary risk.

[0131] (4) When the risk reference value F i In the interval [w2, w3), it means that most of the points in the point cloud subset of the sub-detection area fall in the R2 interval, and the remaining few points fall in other preset distance intervals. The target objects corresponding to a small number of points are not likely to cause occlusion to the laser radar. Therefore, the laser radar can determine that the occlusion risk level of the sub-detection area is level 3 risk. Among them, the risk level of level 3 risk is higher than that of level 2 risk.

[0132] (5) When the risk reference value F i In the interval [w3, w4), it means that most of the points in the point cloud subset of the sub-detection area fall in the R1 interval, and the remaining few points fall in other preset distance intervals. The target objects corresponding to a small number of points are not likely to cause occlusion to the laser radar. Therefore, the laser radar can determine that the occlusion risk level of the sub-detection area is level 4 risk. Among them, the risk level of level 4 risk is higher than that of level 3 risk.

[0133] (6) When the risk reference value F iWhen it is equal to w4, it means that all points in the point cloud subset of the sub-detection area fall within the R1 interval, and there are no points in the R2, R3 and R4 intervals. The laser radar is blocked by a large area in this sub-detection area. Therefore, the laser radar can determine that the occlusion risk level of the sub-detection area is level 5 risk, indicating that there is a serious risk of occlusion in the sub-detection area of ​​the laser radar. Among them, the risk level of level 5 risk is higher than the risk level of level 4 risk.

[0134] Through the above process, the LiDAR can determine the occlusion risk level of each sub-detection area. It should be understood that the above occlusion risk level only indicates the possibility of occlusion in the sub-detection area of ​​the LiDAR, which is intended to provide risk warnings to users, and does not represent whether the sub-detection area of ​​the LiDAR is actually occluded.

[0135] It should be noted that the risk level determined by a single frame of point cloud may have a certain degree of randomness, and when the obstruction in front of the LiDAR is in motion relative to the LiDAR, the occlusion risk level of the sub-detection area may also change dynamically; therefore, it is inaccurate for the LiDAR to determine whether the LiDAR is obstructed in the sub-detection area by obtaining the occlusion risk level of the sub-detection area through only one frame of test point cloud, which may lead to misjudgment. To avoid the above situation, when the LiDAR determines that there is a risk of obstruction in the sub-detection area, it can combine the information of multiple frames of test point clouds to determine whether the sub-detection area is obstructed. The specific process is as follows.

[0136] In a possible implementation, if the laser radar has a risk of being blocked in the sub-detection area, the method further includes:

[0137] In the continuous multi-frame test point cloud, the ratio between the first ratios corresponding to the preset distance intervals with occlusion risk in the sub-detection area with occlusion risk in different frames of test point clouds is a second ratio. If the second ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or,

[0138] In the continuous multi-frame test point cloud, the ratio of the risk reference values ​​corresponding to the sub-detection area with the occlusion risk in the different frames of the test point cloud is a third ratio. If the third ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or,

[0139] In the continuous multi-frame test point cloud, if the occlusion risk levels corresponding to the sub-detection area with occlusion risk in different frames of test point clouds are the same, it is determined that the laser radar is blocked in the sub-detection area.

[0140] Optionally, the multi-frame test point cloud may specifically be a three-frame test point cloud.

[0141] When the laser radar determines that the occlusion risk level of the sub-detection area is any one of the occlusion risk levels of level 1, level 2, level 3, level 4 or level 5, it can further emit laser pulses outward according to its own scanning cycle to obtain two consecutive frames of test point clouds after the current frame test point cloud.

[0142] In the embodiment of the present application, the laser radar provides three methods for determining whether the laser radar is blocked in the sub-detection area based on the multi-frame test point cloud:

[0143] Method 1: Determine whether the laser radar is blocked in the sub-detection area by comparing the first ratios corresponding to the set distance intervals with occlusion risk in the sub-detection area with occlusion risk in the three-frame test point cloud.

[0144] For example, Figure 3 As shown, taking the sub-detection area 1 with occlusion risk as an example, for the following two frames of test point clouds, the laser radar can determine the point cloud subset 1 of the two frames of test point clouds in the sub-detection area 1 respectively in the manner provided in the embodiment of the present application, and according to the number of points of the point cloud subset 1 in each preset distance interval and the theoretical total number of points of the point cloud subset 1, obtain the first ratio corresponding to the preset distance interval with occlusion risk in the point cloud subset 1 in the two frames of test point clouds.

[0145] For a frame of test point cloud, each preset distance interval corresponds to a first ratio; therefore, for the same preset distance interval, three frames of test point cloud will correspond to three first ratios. The laser radar can calculate the ratio between any two of the three first ratios corresponding to each preset distance interval to obtain a second ratio; if the first ratios of the three consecutive frames of test point cloud are close, such as the second ratio is between 0.8 and 1.2, it can be determined that the laser radar is blocked in the corresponding sub-detection area.

[0146] For example, the four preset distance intervals R1, R2, R3, and R4 are 0-15cm, 15-25cm, 25-45cm, and 45-65cm respectively; R1 is a preset distance interval with occlusion risk. The three first ratios corresponding to the three consecutive test point clouds in the sub-detection areas 1 and R1 are A R1 , B R1 , C R1 The laser radar can further calculate the ratio between any two first ratios to obtain multiple second ratios, which are B R1 / A R1 , C R1 / B R1 , C R1 / A R1 .

[0147] After obtaining multiple second ratios, the laser radar can determine whether at least one of the multiple second ratios corresponding to R1 is between 0.8-1.2. If at least one second ratio is between 0.8-1.2, it means that the risk reference value of the sub-detection area 1 obtained based on the first ratio is also stable, and it can be determined that the laser radar is blocked in the sub-detection area 1. On the contrary, when none of the multiple second ratios is between 0.8-1.2, it means that the occlusion state of the sub-detection area has been changing, so there is no need to determine that the laser radar is blocked; in this case, the laser radar cannot determine whether it is blocked in the sub-detection area 1, and can continue to provide the occlusion risk level. Of course, it can further obtain continuous multi-frame test point clouds to update the judgment result.

[0148] Similarly, for other sub-detection areas, it is also possible to determine whether the laser radar is blocked in other sub-detection areas in the same manner as described above, which will not be repeated here.

[0149] Method 2: Determine whether the laser radar is blocked in the sub-detection area by comparing the risk reference values ​​corresponding to different frames of test point clouds in the sub-detection area with occlusion risk in the three frames of test point clouds.

[0150] For example, Figure 3 As shown, taking the sub-detection area 1 as having an occlusion risk as an example, for the two frames of test point clouds subsequently acquired, the laser radar can calculate the risk reference value corresponding to the sub-detection area 1 respectively in the manner provided in the embodiment of the present application. The laser radar further calculates the ratio between any two risk reference values ​​of the three consecutive frames of test point clouds in the sub-detection area 1, and obtains a third ratio; if the risk reference values ​​of the three consecutive frames of test point clouds are close, such as the third ratio is between 0.8 and 1.2, it can be determined that the laser radar is blocked in the corresponding sub-detection area.

[0151] For example, the risk reference values ​​of sub-detection area 1 corresponding to three consecutive frames of test point clouds are P1, Q1, and S1. The laser radar can calculate the ratio between any two risk reference values ​​to obtain multiple third ratios, namely Q1 / P1, S1 / Q1, and S1 / P1.

[0152] After obtaining multiple third ratios, the laser radar can determine whether at least one of the multiple third ratios is between 0.8-1.2. If at least one third ratio is between 0.8-1.2, it means that the risk reference value of the sub-detection area 1 tends to be stable, and it can be determined that the laser radar is blocked in the sub-detection area 1. On the contrary, when none of the multiple third ratios is between 0.8-1.2, it means that the blocking state of the sub-detection area has been changing, so there is no need to determine that the laser radar is blocked; in this case, the laser radar cannot determine whether it is blocked in the sub-detection area 1, and can continue to provide the blocking risk level. Of course, it can further obtain continuous multi-frame test point clouds to update the judgment result.

[0153] Similarly, for other sub-detection areas, it is also possible to determine whether the laser radar is blocked in other sub-detection areas in the same manner as described above, which will not be repeated here.

[0154] Method 3: Determine whether the LiDAR is blocked in the sub-detection area by using the occlusion risk levels corresponding to the different frames of test point clouds in the sub-detection area with occlusion risk in the three frames of test point clouds.

[0155] For example, Figure 3 As shown, taking the occlusion risk in sub-detection area 1 as an example, for the two frames of test point clouds subsequently acquired, the laser radar can obtain the occlusion risk level corresponding to sub-detection risk area 1 in the three frames of test point clouds in accordance with the method provided in the embodiment of the present application.

[0156] For example, the occlusion risk levels of sub-detection area 1 are L1 risk, M1 risk, and N1 risk, respectively, where the value range of L1, M1, and N1 is [1, 5].

[0157] When at least two of L1, M1 and N1 are equal, it means that the occlusion risk level of sub-detection area 1 tends to be stable, and the laser radar determines that it is blocked in sub-detection area 1. On the contrary, when L1, M1 and N1 are not equal, it means that the occlusion state of the sub-detection area has been changing, so there is no need to determine whether the laser radar is blocked; in this case, the laser radar cannot determine whether it is blocked in sub-detection area 1, and the occlusion risk level can be continuously provided. Of course, it can further obtain continuous multi-frame test point clouds to update the judgment result.

[0158] Similarly, for other sub-detection areas, it is also possible to determine whether the laser radar is blocked in other sub-detection areas in the same manner as described above, which will not be repeated here.

[0159] In addition, in addition to judging whether the laser radar is blocked in each sub-detection area according to the blockage risk level of each sub-detection area, the laser radar can also determine the blockage risk level of the laser radar based on the blockage risk level of each sub-detection area.

[0160] In a possible implementation, the occlusion risk level of the laser radar is determined according to the occlusion risk level of the laser radar in each sub-detection area, including:

[0161] The highest level of the occlusion risk levels of each sub-detection area is determined as the occlusion risk level of the laser radar.

[0162] Exemplarily, the occlusion risk levels of the four sub-detection areas determined by the laser radar are level 5 risk, level 4 risk, level 1 risk and level 2 risk respectively, then the occlusion risk level of the laser radar is level 5 risk.

[0163] In another scenario, in addition to determining the occlusion risk level of the laser radar based on the occlusion risk level of the sub-detection area, the embodiment of the present application can also determine the occlusion risk level of the laser radar based on the risk reference value of the sub-detection area, specifically including the following steps: determining the average value of the risk reference value of each sub-detection area; determining the weight coefficient interval or target weight coefficient corresponding to the risk reference value average value, and the occlusion risk level corresponding to the weight coefficient interval or the occlusion risk level corresponding to the target weight coefficient; determining the occlusion risk level corresponding to the weight coefficient interval as the occlusion risk level of the laser radar, or determining the occlusion risk level corresponding to the target weight coefficient as the occlusion level of the laser radar.

[0164] For example, the weighted risk intervals provided in the embodiment of the present application include the aforementioned [thr1, w1), [w1, w2), [w2, w3) and [w3, w4), and the target weight coefficient is w4. i After that, the risk reference values ​​of all sub-detection areas can be averaged; further, the laser radar determines the weight coefficient interval or the corresponding target weight coefficient to which the average value of the risk reference value belongs, and determines the occlusion risk level of the laser radar as the occlusion risk level corresponding to the weight coefficient interval or the occlusion risk level corresponding to the target weight coefficient. For example, when the weight coefficient interval is [w2, w3), the occlusion risk level of the laser radar is level 3 risk; when the target weight coefficient is w4, the occlusion risk level of the laser radar is level 5 risk.

[0165] Therefore, through the above process, the laser radar can further obtain the occlusion risk level of the laser radar after obtaining the occlusion risk level of each sub-detection area.

[0166] In an embodiment of the present application, in addition to determining the occlusion risk level of the laser radar based on the occlusion risk level of each sub-detection area, the laser radar can also perform a fault alarm if the occlusion risk level of each sub-detection area is too high, indicating that there is a high possibility that the laser radar is obscured in the sub-detection area.

[0167] In a possible implementation manner, the method further includes:

[0168] If the occlusion risk level of at least one sub-detection area is higher than or equal to a first preset risk level, a target fault code is generated.

[0169] Optionally, the first preset risk level is level three risk.

[0170] When the occlusion risk level of at least one sub-detection area is higher than or equal to level 3 risk, the laser radar can generate a target fault code, where the target fault code is used to identify a scenario where at least one sub-detection area in the laser radar has a high occlusion risk. The laser radar can report the target fault code to the host computer, which determines the fault type of the laser radar through the target fault code and notifies the fault type to the technician's terminal device in the form of a sound or text prompt, so that the technician can quickly handle the fault.

[0171] In the above technical solution, when the occlusion risk level of the sub-detection area of ​​the laser radar is high, by generating a target fault code and reporting it to the host computer, the host computer can respond to the fault of the laser radar in time to ensure that the laser radar can work normally.

[0172] After generating the target fault code, the laser radar can also determine whether the occlusion risk level of the laser radar has changed through the next multiple frames of continuous test point cloud to determine whether it is necessary to generate a reset fault code for the target fault code.

[0173] In a possible implementation, if the shielding risk level of at least one sub-detection area is higher than or equal to the first preset risk level, after the step of generating a target fault code, the method further includes:

[0174] If, in the continuous multi-frame test point cloud, the first ratios of the different frame test point clouds in the target sub-detection area and the target distance interval are all lower than the second preset threshold, a reset fault code of the target fault code is generated; or,

[0175] If, in the continuous multi-frame test point cloud, the risk reference values ​​of different frame test point clouds in the target sub-detection area are all lower than the third preset threshold, a reset fault code of the target fault code is generated; or,

[0176] In the continuous multi-frame test point cloud, if the occlusion risk levels corresponding to the target sub-detection area of ​​the different frame test point clouds are all lower than the first preset risk level, a reset fault code of the target fault code is generated;

[0177] Among them, the target sub-detection area is a sub-detection area whose occlusion risk level is higher than or equal to the first preset risk level, and the target distance interval is the preset distance interval corresponding to the occlusion risk level; for example, if the occlusion risk level of the target sub-detection area is level three risk, then the target distance interval is R2; for another example, if the occlusion risk level of the target sub-detection area is level four risk or level five risk, then the target distance interval is R1.

[0178] When the laser radar determines that the occlusion risk level of at least one sub-detection area is higher than or equal to the first preset risk level, the laser radar generates a target fault code. In the embodiment of the present application, the laser radar can continue to send laser pulses based on its own scanning cycle, thereby obtaining a continuous multi-frame test point cloud, and can determine whether the risk level has changed based on the difference between the multi-frame test point clouds to decide whether to generate a reset fault code; optionally, the continuous multi-frame test point cloud is specifically a three-frame test point cloud.

[0179] Similar to the above method of determining whether the laser radar is blocked in the sub-detection area, the laser radar also includes the following three methods when determining whether a reset fault code of the target fault code needs to be generated:

[0180] Method 1: Determine whether to generate a reset fault code for the target fault code based on a first ratio of a multi-frame test point cloud in a target sub-detection area and a target distance interval.

[0181] The target distance interval is a preset distance interval in which the first ratio determined based on the previous test point cloud is higher than a preset threshold.

[0182] For example, Figure 3 As shown, the target sub-detection area is sub-detection area 1, and the target distance interval is 0-15cm. Based on the three frames of test point clouds obtained continuously, the laser radar can calculate the point cloud subsets of the three frames of test point clouds in sub-detection area 1, and obtain the number of points of the point cloud subset in the target distance interval according to the point cloud subset of each frame of test point cloud in sub-detection area 1. Finally, according to the number of points of the point cloud subset of each frame of test point cloud in sub-detection area 1 in the target distance interval and the theoretical total number of points of the point cloud subset of sub-detection area 1, the three first ratios of the three frames of test point clouds in sub-detection area 1 and the target distance interval are obtained.

[0183] When the three first ratios are all lower than the second preset threshold, it means that the number of points in the point cloud subset of sub-detection area 1 in the target distance interval has obviously decreased or disappeared, and the occlusion risk level of sub-detection area 1 finally determined will inevitably decrease significantly. Therefore, the laser radar can generate a reset fault code for the target fault code and send it to the host computer so that the host computer can clear or reset the fault state of the laser radar. Optionally, the second preset threshold can be a threshold value less than or equal to the lower limit value of the weight coefficient interval corresponding to the above-mentioned first preset risk level; for example, in some embodiments, the second preset threshold can be configured as the above-mentioned first preset threshold, that is, 0.1.

[0184] Method 2: Determine whether to generate a reset fault code for the target fault code based on the risk reference value of the multi-frame test point cloud in the target sub-detection area.

[0185] Based on the obtained multi-frame test point cloud, the laser radar can calculate the risk reference value of the multi-frame test point cloud in the target sub-detection area according to the method provided in the embodiment of the present application.

[0186] For example, Figure 3 As shown, the target sub-detection area is sub-detection area 1. By further acquiring three frames of test point clouds, the laser radar can calculate three risk reference values ​​corresponding to the three frames of test point clouds in sub-detection area 1 respectively.

[0187] When the three risk reference values ​​corresponding to sub-detection area 1 are all less than the third preset threshold, it means that the occlusion risk of the target sub-detection area decreases or disappears, and the laser radar generates a reset fault code of the target fault code and sends it to the host computer so that the host computer clears or resets the fault state of the laser radar. Optionally, the third preset threshold is a threshold value that is less than or equal to the lower limit value of the weight coefficient interval corresponding to the above-mentioned first preset risk level; for example, in some embodiments, the third preset threshold value can be configured as the above-mentioned second weight coefficient w2; for example, in some embodiments, the second preset threshold value can be configured as the above-mentioned first weight coefficient w1; for example, in other embodiments, the second preset threshold value can be configured as the above-mentioned first preset threshold value thr1.

[0188] When the three risk reference values ​​corresponding to sub-detection area 1 are not all less than the third preset threshold, it means that the occlusion risk of the target sub-detection area has not decreased or disappeared, and the laser radar does not generate a reset fault code for the target fault code.

[0189] Method 3: Determine whether to generate a reset fault code for the target fault code based on the occlusion risk level corresponding to the multi-frame test point cloud in the target sub-detection area.

[0190] Based on the obtained three frames of test point clouds, the laser radar can calculate the occlusion risk level of the three frames of test point clouds in the target sub-detection area according to the method provided in the embodiment of the present application.

[0191] For example, Figure 3 As shown, the target sub-detection area is sub-detection area 1. By further acquiring three frames of test point clouds, the laser radar can calculate three occlusion risk levels corresponding to the three frames of test point clouds in sub-detection area 1 respectively.

[0192] When the three occlusion risk levels corresponding to sub-detection area 1 are all lower than the first preset risk level (level 3 risk), it means that the occlusion risk of the target sub-detection area has decreased or disappeared, and the laser radar generates a reset fault code for the target fault code and sends it to the host computer so that the host computer can clear or reset the fault state of the laser radar.

[0193] When the three occlusion risk levels corresponding to sub-detection area 1 are not all lower than the preset risk level, it means that the occlusion risk of the target sub-detection area has not decreased or disappeared, and the laser radar does not generate a reset fault code for the target fault code.

[0194] In the above technical solution, after the laser radar generates a target fault code, the laser radar combines continuous multi-frame test point clouds to calculate the first ratio of the target sub-detection area and the target distance interval, or calculates the risk reference value of the target sub-detection area, or calculates the occlusion risk level of the target sub-detection area to determine whether the laser radar still has a high occlusion risk in the target sub-detection area. When the high occlusion risk of the laser radar in the target sub-detection area is lifted, the laser radar can also generate a reset fault code for the target fault code and report it. The above process makes the generation method of the reset fault code flexible and diverse, and can also timely update the occlusion risk level of the laser radar in the target sub-detection area and clear the laser radar fault.

[0195] It is worth mentioning that even though the above is explained using the "multi-frame test point cloud" as a three-frame test point cloud as an example, the present application is not limited to this. As long as there are more than two frames of test point clouds, the above-mentioned "determining that the laser radar is blocked in the sub-detection area" step can be achieved, and the above-mentioned "generating a reset fault code for the target fault code" step can also be achieved; for example, in other embodiments, the above process can also be performed based on two consecutive frames of test point clouds.

[0196] It should be understood that in the embodiment of the present application, when dividing the detection field of view of the laser radar, the detection field of view can be divided into a region of interest (ROI) and a non-region of interest (Non-ROI) according to the importance of the detection area, and the region of interest and the non-region of interest constitute multiple sub-detection areas.

[0197] Regions of interest refer to specific areas that require special attention during data processing and analysis. These areas usually contain important information or key targets.

[0198] Regions of non-interest refer to areas that do not require special attention during data processing and analysis. These areas usually do not contain important information or key targets.

[0199] According to the different attention levels of the areas of interest and non-areas of interest, the laser radar in the embodiment of the present application can adopt different alarm strategies according to the different sub-detection areas when a fault alarm occurs.

[0200] In a possible implementation manner, the method further includes:

[0201] If the occlusion risk level of the region of interest is higher than or equal to a second preset risk level, generating a target fault code; and

[0202] If the occlusion risk level of the non-interested area is higher than or equal to a third preset risk level, a target fault code is generated, wherein the risk level corresponding to the third preset risk level is higher than the risk level corresponding to the second preset risk level.

[0203] Figure 5 This is another schematic diagram of a scenario in which a plurality of sub-detection areas are divided provided in an embodiment of the present application.

[0204] For example, Figure 5 As shown, the middle area of ​​the detection field of view is divided into the region of interest, and the area other than the region of interest in the detection field of view is divided into non-interest region 1 and non-interest region 2. The multiple sub-detection regions include the region of interest, non-interest region 1 and non-interest region 2.

[0205] Since the data in the area of ​​interest is of high importance, when the LiDAR issues an occlusion risk level alarm, the reporting threshold of the area of ​​interest can be set lower, that is, the target fault code is triggered when the occlusion risk level is low. On the contrary, since the data in the area of ​​interest is of low importance, when the LiDAR issues an occlusion risk level alarm, the reporting threshold of the non-interested area can be set higher, that is, the target fault code is triggered when the occlusion risk level is slightly higher.

[0206] Specifically, after the laser radar determines the occlusion risk level of the area of ​​interest, the occlusion risk level of the non-interest area 1, and the occlusion risk level of the non-interest area 2, if the occlusion risk level of the area of ​​interest is higher than or equal to the second preset risk level, the laser radar generates a target fault code based on the occlusion risk level of the area of ​​interest. Optionally, the second preset risk level is a secondary risk.

[0207] If the occlusion risk level of the non-interest area 1 or the non-interest area 2 is higher than or equal to the third preset risk level, a target fault code is generated. Optionally, the third preset risk level is a third level risk.

[0208] In the above technical solution, a lower second preset risk level is set for the area of ​​interest, and a higher third preset risk level is set for the area of ​​non-interest, which can ensure that the occlusion risk of the area of ​​interest can be reported as early as possible, focus more attention on the area of ​​interest, and enable the occlusion risk of the area of ​​interest to achieve a faster response.

[0209] Figure 6 It is a structural schematic diagram of a laser radar occlusion risk detection device provided in an embodiment of the present application.

[0210] For example, Figure 6 As shown, the device 600 includes:

[0211] Point cloud acquisition module 601, used to acquire test point cloud;

[0212] The occlusion risk level determination module 602 is used to determine the occlusion risk level of the laser radar based on the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, wherein each of the preset distance intervals is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals correspond to different weight coefficients.

[0213] In one possible implementation, the detection field of view of the laser radar includes multiple sub-detection areas arranged along a first preset direction, and the points of the test point cloud in each sub-detection area constitute a point cloud subset. The occlusion risk level determination module 602 is specifically used to: determine the occlusion risk level of the laser radar in the sub-detection area according to the number of points of each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval; determine the occlusion risk level of the laser radar according to the occlusion risk level of the laser radar in each sub-detection area.

[0214] In one possible implementation, the occlusion risk level determination module 602 is also used to: determine a first ratio between the number of points of the point cloud subset in each of the preset distance intervals and the theoretical total number of points of the point cloud subset; and determine the occlusion risk level of the sub-detection area based on the first ratio corresponding to the point cloud subset in each of the preset distance intervals and the corresponding weight coefficient.

[0215] In one possible implementation, the occlusion risk level determination module 602 is also used to: obtain a risk reference value based on a weighted sum of multiple first ratios and corresponding multiple weight coefficients; and determine the occlusion risk level of the sub-detection area based on the risk reference value and the multiple weight coefficients.

[0216] In a possible implementation, the number of the preset distance intervals and the weight coefficients are both four, and the four weight coefficients are, from small to large, the first weight coefficient, the second weight coefficient, the third weight coefficient, and the fourth weight coefficient. The larger the upper limit value of the preset distance interval, the smaller the weight coefficient corresponding to the preset distance interval. The occlusion risk level determination module 602 is also used to: if the risk reference value is less than a first preset threshold, determine that the sub-detection area is not obstructed, wherein the first preset threshold is less than the first weight coefficient; if the risk reference value is greater than or equal to the first preset threshold and less than the first weight coefficient, determine that the occlusion risk level of the sub-detection area is a first-level risk; if the risk reference value is greater than or equal to If the first weight coefficient is greater than or equal to the second weight coefficient and is less than the second weight coefficient, the occlusion risk level of the sub-detection area is determined to be a second-level risk; if the risk reference value is greater than or equal to the second weight coefficient and is less than the third weight coefficient, the occlusion risk level of the sub-detection area is determined to be a third-level risk; if the risk reference value is greater than or equal to the third weight coefficient and is less than the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a fourth-level risk; if the risk reference value is equal to the fourth weight coefficient, the occlusion risk level of the sub-detection area is determined to be a fifth-level risk; wherein the occlusion risk levels of the first-level risk, the second-level risk, the third-level risk, the fourth-level risk, and the fifth-level risk increase in sequence.

[0217] Optionally, if the laser radar has a risk of being blocked in the sub-detection area, the device also includes: an occlusion area determination module, which is used to determine that in a continuous multi-frame test point cloud, the ratio of the first ratios corresponding to the preset distance intervals with occlusion risk of different frame test point clouds in the sub-detection area with occlusion risk is a second ratio, and if the second ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the ratio of the risk reference values ​​corresponding to different frame test point clouds in the sub-detection area with occlusion risk is a third ratio, and if the third ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frame test point clouds in the sub-detection area with occlusion risk are the same, then it is determined that the laser radar is blocked in the sub-detection area.

[0218] In a possible implementation, the occlusion risk level determination module 602 is further used to: determine the highest level of the occlusion risk levels of each of the sub-detection areas as the occlusion risk level of the laser radar.

[0219] Optionally, the device further includes: a first fault code generating module, configured to generate a target fault code if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to a first preset risk level.

[0220] Optionally, after the step of generating a target fault code if the occlusion risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, the device also includes: a reset fault code generation module, which is used to generate a reset fault code for the target fault code if, in a continuous multi-frame test point cloud, a first ratio of different frame test point clouds in the target sub-detection area and the target distance interval is lower than a second preset threshold; or, if, in a continuous multi-frame test point cloud, the risk reference values ​​of different frame test point clouds in the target sub-detection area are lower than a third preset threshold, a reset fault code for the target fault code is generated; or, if, in a continuous multi-frame test point cloud, the occlusion risk levels corresponding to different frame test point clouds in the target sub-detection area are lower than the first preset risk level, a reset fault code for the target fault code is generated; wherein, the target sub-detection area is a sub-detection area whose previous occlusion risk level is higher than or equal to the first preset risk level, and the target distance interval is the preset distance interval corresponding to the occlusion risk level.

[0221] Optionally, the multiple sub-detection areas include areas of interest and areas of no interest, and the device also includes: a second fault code generation module, used to generate a target fault code if the occlusion risk level of the area of ​​interest is higher than or equal to a second preset risk level; and to generate a target fault code if the occlusion risk level of the area of ​​no interest is higher than or equal to a third preset risk level, wherein the risk level corresponding to the third preset risk level is higher than the risk level corresponding to the second preset risk level.

[0222] Figure 7 It is a structural schematic diagram of a laser radar provided in an embodiment of the present application.

[0223] For example, Figure 7 As shown, the laser radar 102 includes: a memory 701 and a processor 702, wherein the memory 701 stores an executable program code 7011, and the processor 702 is used to call and execute the executable program code 7011 to perform a laser radar occlusion risk detection method.

[0224] In this embodiment, the laser radar can be divided into functional modules according to the above method example. For example, each functional module can be corresponded, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0225] In the case of dividing each functional module according to each function, the laser radar may include: a point cloud acquisition module and an occlusion risk level determination module, etc. It should be noted that all relevant contents of each step involved in the above method embodiment can be referred to the functional description of the corresponding functional module, which will not be repeated here.

[0226] The laser radar provided in this embodiment is used to execute the above-mentioned laser radar occlusion risk detection method, and thus can achieve the same effect as the above-mentioned implementation method.

[0227] In the case of an integrated unit, the laser radar may include a processing module and a storage module. The processing module may be used to control and manage the actions of the laser radar. The storage module may be used to support the laser radar in executing related program codes and data.

[0228] The processing module may be a processor or a controller, which may implement or execute various exemplary logic blocks, modules and circuits shown in conjunction with the disclosure of the present application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module may be a memory.

[0229] This embodiment also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement a laser radar occlusion risk detection method in the above-mentioned embodiment.

[0230] This embodiment also provides a computer program product. When the computer program product runs on a computer, it enables the computer to execute the above-mentioned related steps to implement a laser radar occlusion risk detection method in the above-mentioned embodiment.

[0231] In addition, the laser radar provided in the embodiments of the present application can specifically be a chip, component or module, and the laser radar may include a connected processor and memory; wherein the memory is used to store instructions, and when the laser radar is running, the processor can call and execute instructions so that the chip executes a laser radar occlusion risk detection method in the above-mentioned embodiment.

[0232] Among them, the laser radar, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0233] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0234] In the embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, 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.

[0235] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A laser radar occlusion risk detection method, characterized in that: The method comprises: Obtaining a test point cloud; and The occlusion risk level of the laser radar is determined based on the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, wherein each of the preset distance intervals is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals have different corresponding weight coefficients.

2. The method according to claim 1, characterized in that The detection field of view of the laser radar includes a plurality of sub-detection areas arranged along a first preset direction, the points of the test point cloud in each sub-detection area constitute a point cloud subset, and the occlusion risk level of the laser radar is determined according to the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, including: Determine the occlusion risk level of the laser radar in the sub-detection area according to the number of points of each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval; The occlusion risk level of the laser radar is determined according to the occlusion risk level of the laser radar in each of the sub-detection areas.

3. The method according to claim 2, characterized in that Determining the occlusion risk level of the laser radar in the sub-detection area according to the number of points of each point cloud subset in each preset distance interval and the weight coefficient corresponding to the preset distance interval includes: Determine a first ratio between the number of points of the point cloud subset in each of the preset distance intervals and the theoretical total number of points of the point cloud subset; The occlusion risk level of the sub-detection area is determined according to the first ratio corresponding to the point cloud subset in each of the preset distance intervals and the corresponding weight coefficient.

4. The method according to claim 3, characterized in that Determining the occlusion risk level of the sub-detection area according to the first ratio corresponding to the point cloud subset in each of the preset distance intervals and the corresponding weight coefficient includes: Obtaining a risk reference value by performing weighted summation based on the plurality of first ratios and the corresponding plurality of weight coefficients; The shielding risk level of the sub-detection area is determined according to the risk reference value and the multiple weight coefficients.

5. The method according to claim 4, characterized in that The number of the preset distance intervals and the weight coefficients are both four, and the four weight coefficients are, from small to large, a first weight coefficient, a second weight coefficient, a third weight coefficient, and a fourth weight coefficient. The larger the upper limit value of the preset distance interval, the smaller the weight coefficient corresponding to the preset distance interval. The determining of the occlusion risk level of the sub-detection area according to the risk reference value and the multiple weight coefficients includes: If the risk reference value is less than a first preset threshold, determining that the sub-detection area is not blocked, wherein the first preset threshold is less than the first weight coefficient; If the risk reference value is greater than or equal to the first preset threshold value and less than the first weight coefficient, determining that the occlusion risk level of the sub-detection area is a level one risk; If the risk reference value is greater than or equal to the first weight coefficient and less than the second weight coefficient, determining that the occlusion risk level of the sub-detection area is a level 2 risk; If the risk reference value is greater than or equal to the second weight coefficient and less than the third weight coefficient, determining that the occlusion risk level of the sub-detection area is a third-level risk; If the risk reference value is greater than or equal to the third weight coefficient and less than the fourth weight coefficient, determining that the occlusion risk level of the sub-detection area is a level 4 risk; If the risk reference value is equal to the fourth weight coefficient, determining that the occlusion risk level of the sub-detection area is a level 5 risk; Among them, the occlusion risk levels of the first-level risk, the second-level risk, the third-level risk, the fourth-level risk, and the fifth-level risk increase in sequence.

6. The method according to claim 5, characterized in that If the laser radar has a risk of being blocked in the sub-detection area, the method further includes: In the continuous multi-frame test point cloud, the ratio between the first ratios corresponding to the preset distance intervals with occlusion risk in the sub-detection area with occlusion risk in different frames of test point clouds is a second ratio. If the second ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or, In the continuous multi-frame test point cloud, the ratio of the risk reference values ​​corresponding to the sub-detection area with the risk of occlusion in different frames of test point cloud is a third ratio. If the third ratio is between 0.8 and 1.2, it is determined that the laser radar is blocked in the sub-detection area; or In the continuous multi-frame test point cloud, if the occlusion risk levels corresponding to the sub-detection area with occlusion risk in different frames of test point clouds are the same, it is determined that the laser radar is blocked in the sub-detection area.

7. The method according to claim 2, characterized in that: Determining the occlusion risk level of the laser radar according to the occlusion risk level of the laser radar in each of the sub-detection areas includes: The highest level among the occlusion risk levels of the sub-detection areas is determined as the occlusion risk level of the laser radar.

8. The method according to claim 2, characterized in that: The method further comprises: If the shielding risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, a target fault code is generated.

9. The method according to claim 8, characterized in that After the step of generating a target fault code if the shielding risk level of at least one of the sub-detection areas is higher than or equal to the first preset risk level, the method further includes: If, in the continuous multi-frame test point cloud, the first ratios of the different frame test point clouds in the target sub-detection area and the target distance interval are all lower than the second preset threshold, a reset fault code of the target fault code is generated; or, If, in the continuous multi-frame test point cloud, the risk reference values ​​of different frame test point clouds in the target sub-detection area are all lower than the third preset threshold, a reset fault code of the target fault code is generated; or, If, in the continuous multi-frame test point cloud, the occlusion risk levels corresponding to the different frame test point clouds in the target sub-detection area are all lower than the first preset risk level, a reset fault code of the target fault code is generated; The target sub-detection area is a sub-detection area whose previous shielding risk level is higher than or equal to the first preset risk level, and the target distance interval is a preset distance interval corresponding to the shielding risk level.

10. The method according to claim 2, characterized in that The plurality of sub-detection areas include an area of ​​interest and an area of ​​non-interest, and the method further includes: If the occlusion risk level of the region of interest is higher than or equal to a second preset risk level, generating a target fault code; and, If the occlusion risk level of the non-interest area is higher than or equal to a third preset risk level, a target fault code is generated, wherein the risk level corresponding to the third preset risk level is higher than the risk level corresponding to the second preset risk level.

11. A laser radar occlusion risk detection device, characterized in that: The device comprises: A point cloud acquisition module, used to acquire a test point cloud; and An occlusion risk level determination module is used to determine the occlusion risk level of the laser radar based on the number of points of the test point cloud in each preset distance interval and the weight coefficient corresponding to the preset distance interval, wherein each of the preset distance intervals is a plurality of distance intervals continuous in distance value determined with a preset occlusion distance threshold as an upper limit, and different preset distance intervals correspond to different weight coefficients.

12. A laser radar, characterized in that: The laser radar comprises: A memory for storing executable program codes; A processor, configured to call and run the executable program code from the memory so that the laser radar executes the method as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method according to any one of claims 1 to 10 is implemented.

Images