Radar shelter detection method and device, computer device, chip and terminal

Abstract

The application discloses a radar shelter detection method and device, computer equipment, a chip and a terminal, relates to the technical field of detection, and mainly aims to solve the problem of detection efficiency of vehicle radar shelters in the existing mining area. Mainly includes: acquiring real-time point cloud frames collected by different types of radars; determining a target point cloud frame corresponding to the real-time point cloud frame based on a time threshold; counting the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame; if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than an abnormal frame number ratio threshold, it is determined that the radar is blocked, and the detection of the radar shelter is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of detection, in particular to a radar shelter detection method and device, a computer device, a chip and a terminal. BACKGROUND

[0002] In the field of automatic driving, a laser radar is one of the important sensors of an automatic driving perception system to perceive the surrounding environment. Since the radar lens is exposed to the external environment for a long time, dust, particulate matter and other objects will adhere to the mirror surface of the radar lens, affecting the laser reflection and reception of the radar, resulting in an inability to accurately detect the distance of the surrounding environment, thereby affecting the perception system of the automatic driving. Especially in more severe mine environments, intelligent vehicles need to detect whether there is a radar shelter to reduce accidents of vehicles and workers.

[0003] Currently, the existing detection of radar shelters is usually based on manual cleaning or the installation of a dirt sensor, which triggers a cleaning device when the dirt threshold is reached. However, manual cleaning increases manpower and resources, and cannot effectively remove the shelter in a timely manner. Adding a dedicated hardware device greatly increases the cost of the device, and the device structure is relatively complex, which cannot meet the effective cleaning needs of the radar of the vehicle in the mine, thereby reducing the detection efficiency of the radar shelter. SUMMARY

[0004] Therefore, the present application provides a radar shelter detection method and device, storage medium and computer device, which mainly aims to solve the problem of the detection efficiency of the radar shelter of the vehicle in the existing mine.

[0005] According to one aspect of the present application, a radar shelter detection method is provided, comprising:

[0006] S1: acquiring real-time point cloud frames collected by different types of radars;

[0007] S2: determining a target point cloud frame corresponding to the real-time point cloud frame based on a time threshold;

[0008] S3: counting the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame;

[0009] S4: if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than an abnormal frame number ratio threshold, it is determined that the radar is sheltered.

[0010] Further, before the target point cloud frame corresponding to the real-time point cloud frame is determined based on the time threshold, the method further comprises:

[0011] acquire a normal point quantity in a real-time point cloud frame collected by different types of radars, and a normal point quantity proportion threshold corresponding to the different types of radars;

[0012] If a ratio of the normal point quantity to a total point quantity in the real-time point cloud frame is greater than the normal point quantity proportion threshold, a state identifier of the real-time point cloud frame is determined as a normal point cloud frame.

[0013] If the ratio of the normal point quantity to the total point quantity in the real-time point cloud frame is less than or equal to the normal point quantity proportion threshold, the state identifier of the real-time point cloud frame is determined as an abnormal point cloud frame.

[0014] The timestamps and the state identifiers of the real-time point cloud frames are stored in a target container in a time sequence in which the real-time point cloud frames are acquired, wherein one point cloud frame corresponds to one timestamp and one state identifier.

[0015] Further, the target point cloud frame corresponding to the real-time point cloud frame is determined based on a time threshold, and the target point cloud frame corresponding to the real-time point cloud frame includes:

[0016] A target timestamp corresponding to an end point of the time threshold is queried in the target container, taking the timestamp of the real-time point cloud frame as a starting point and the time threshold as a query length.

[0017] A point cloud frame corresponding to the target timestamp is determined as the target point cloud frame.

[0018] Further, the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame are counted, and the counting includes:

[0019] The total number of point cloud frames between the real-time point cloud frame and the target point cloud frame is counted based on the number of timestamps or the number of state identifiers in the target container.

[0020] The number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame is counted based on the number of abnormal state identifiers.

[0021] Further, the normal point quantity proportion threshold corresponding to the different types of radars is acquired, and the acquiring includes:

[0022] Test point cloud frames of radars of any type of radar under a first preset shielding degree value and a second preset shielding degree value are acquired, wherein the first preset shielding degree value is less than the second preset shielding degree value.

[0023] identify the abnormal points in the test point cloud frame, count the number of abnormal points in the test point cloud frame, take the difference between the total number of points in the test point cloud frame and the number of abnormal points as the number of normal points of the test point cloud frame, and take the ratio of the number of normal points to the total number of points as the normal point number ratio threshold under the corresponding preset occlusion degree value;

[0024] The intermediate value between the normal point number ratio threshold under the first preset occlusion degree value and the normal point number ratio threshold under the second preset occlusion degree value is taken as the normal point number ratio threshold corresponding to the radar of the arbitrary type.

[0025] Further, the method further comprises:

[0026] Setting the capacity parameter threshold of the target container, the capacity parameter threshold including a time capacity parameter threshold or a frame number capacity parameter threshold;

[0027] When the capacity of the target container reaches the corresponding capacity parameter threshold, updating the target container.

[0028] Further, before the target point cloud frame corresponding to the real-time point cloud frame is determined based on the time threshold, the method further comprises:

[0029] The radar with autonomous occlusion detection function is set as a specified type radar;

[0030] The type of the radar that collects the real-time point cloud frame is identified through the real-time point cloud frame;

[0031] When the type of the radar is identified as the specified type radar, the fault code of the specified type radar is obtained, and if the fault code indicates that the specified type radar is in a non-occluded state, it is determined that the specified type radar is in a non-occluded state;

[0032] If the fault code indicates that the specified type radar is in an occluded state, steps S2-S4 are repeated to determine whether the specified type radar is occluded.

[0033] Further, after it is determined that the radar is occluded, the method further comprises:

[0034] Outputting an occlusion fault alarm information to a control safety subsystem to make the control safety subsystem control a vehicle equipped with the radar to reduce speed and notify a safety officer to perform cleaning; or control the safety subsystem to start a corresponding automatic cleaning device.

[0035] According to another aspect of the present application, a radar occlusion detection device is provided, comprising:

[0036] An acquisition module is configured to acquire real-time point cloud frames collected by different types of radars;

[0037] The first determination module is configured to determine a target point cloud frame corresponding to the real-time point cloud frame based on a time threshold value.

[0038] The statistical module is configured to count a total number of point cloud frames and a number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame.

[0039] The second determination module is configured to determine that the radar is blocked if a ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than an abnormal frame number ratio threshold value.

[0040] Further, the apparatus further comprises a third determination module, a storage module,

[0041] The acquisition module is further configured to acquire a number of normal points in a real-time point cloud frame collected by a radar of a different type and a normal point number ratio threshold value corresponding to the radar of the different type.

[0042] The third determination module is configured to determine a state identifier of the real-time point cloud frame as a normal point cloud frame if a ratio of the number of normal points in the real-time point cloud frame to a total number of points is greater than the normal point number ratio threshold value, and determine the state identifier of the real-time point cloud frame as an abnormal point cloud frame if the ratio of the number of normal points in the real-time point cloud frame to the total number of points is less than or equal to the normal point number ratio threshold value.

[0043] The storage module is configured to store, in a target container, time stamps and state identifiers of the real-time point cloud frames in a time sequence in which the real-time point cloud frames are acquired, wherein one point cloud frame corresponds to one time stamp and one state identifier.

[0044] Further, the first determination module is specifically configured to query, in the target container, a target time stamp corresponding to an end point of the time threshold value, taking the time stamp of the real-time point cloud frame as a starting point and taking the time threshold value as a query length, and determine a point cloud frame corresponding to the target time stamp as the target point cloud frame.

[0045] Further, the statistical module is specifically configured to count, in the target container, a total number of point cloud frames between the real-time point cloud frame and the target point cloud frame based on a number of time stamps or a number of state identifiers, and count a number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame based on a number of abnormal state identifiers.

[0046] Further, the acquisition module comprises:

[0047] The acquisition unit is configured to acquire test point cloud frames of a radar of any type of radar at a first preset blocking degree value and a second preset blocking degree value, wherein the first preset blocking degree value is less than the second preset blocking degree value.

[0048] a statistics unit, configured to identify an abnormal point in the test point cloud frame, count a number of the abnormal points in the test point cloud frame, take a difference between a total number of points in the test point cloud frame and the number of the abnormal points as a number of normal points of the test point cloud frame, and take a ratio of the number of normal points to the total number of points as a threshold of a proportion of normal points under a preset occlusion degree value;

[0049] a determination unit, configured to take an intermediate value between the threshold of the proportion of normal points under the first preset occlusion degree value and the threshold of the proportion of normal points under the second preset occlusion degree value as a threshold of the proportion of normal points corresponding to the radar of the arbitrary type.

[0050] Further, the device further comprises:

[0051] a setting module, configured to set a capacity parameter threshold of the target container, the capacity parameter threshold comprising a time capacity parameter threshold or a frame number capacity parameter threshold;

[0052] an updating module, configured to update the target container when a capacity of the target container reaches a corresponding capacity parameter threshold.

[0053] Further, the device further comprises:

[0054] The setting module is further configured to set the radar with the autonomous occlusion detection function as a specified type radar.

[0055] The obtaining module is further configured to identify and obtain a type of the radar of the real-time point cloud frame through the real-time point cloud frame.

[0056] When the type of the radar is identified and obtained as the specified type radar, the obtaining module is further configured to obtain a fault code of the specified type radar, and determine that the specified type radar is in a non-occlusion state if the fault code indicates that the specified type radar is in the non-occlusion state, and repeat steps S2-S4 to determine whether the specified type radar is occluded if the fault code indicates that the specified type radar is in an occlusion state.

[0057] Further, the device further comprises:

[0058] an output module, configured to output an occlusion fault alarm information to a control safety subsystem, so that the control safety subsystem controls a vehicle installed with the radar to perform a speed reduction process, and notifies a safety officer to perform a cleaning process, or controls the safety subsystem to start a corresponding automatic cleaning device.

[0059] According to an aspect of the present application, a computer readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the steps of the radar occlusion detection method are implemented.

[0060] According to an aspect of the present application, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor; when the processor executes the computer program, the steps of the radar shelter detection method described above are implemented.

[0061] According to an aspect of the present application, a chip is provided, comprising at least one processor and a communication interface coupled with the at least one processor, the at least one processor being configured to execute a computer program or instructions to implement the radar shelter detection method described above.

[0062] According to an aspect of the present application, a terminal is provided, comprising the radar shelter detection device described above.

[0063] By means of the technical solutions described above, the technical solutions provided by the embodiments of the present application have at least the following advantages:

[0064] The present application provides a radar shelter detection method and device, an existing medium, and a computer device. Compared with the prior art, the embodiments of the present application acquire real-time point cloud frames collected by different types of radars, determine target point cloud frames corresponding to the real-time point cloud frames based on a time threshold, count the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frames and the target point cloud frames, and determine that the radar is sheltered if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than an abnormal frame number ratio threshold. The embodiments of the present application meet the shelter detection requirements of different radars, reduce the manpower and material resources costs of shelter detection hardware installation, improve the detection efficiency and accuracy of radar shelters, and achieve the effective cleaning requirements of mine vehicle radars.

[0065] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the embodiments of the present application can be implemented according to the content of the description, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described. BRIEF DESCRIPTION OF DRAWINGS

[0066] Various other advantages and benefits will become apparent to those of ordinary skill in the art, upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not meant to limit the present application. Furthermore, the same reference numerals are used throughout the several drawings to designate the same or similar parts. In the drawings:

[0067] Figure 1 A radar shelter detection method flowchart provided by the embodiments of the present application is shown;

[0068] Figure 2A flow chart of another radar shelter detection method provided by the embodiment of the present application is shown;

[0069] Figure 3 A schematic diagram of the timestamp and state identifier storage data provided by the embodiment of the present application is shown;

[0070] Figure 4 A flow chart of another radar shelter detection method provided by the embodiment of the present application is shown;

[0071] Figure 5 A block diagram of the radar shelter detection device provided by the embodiment of the present application is shown;

[0072] Figure 6 A structural schematic diagram of the computer readable storage medium provided by the embodiment of the present application is shown;

[0073] Figure 7 A structural schematic diagram of the computer device provided by the embodiment of the present application is shown;

[0074] Figure 8 A structural schematic diagram of the chip provided by the embodiment of the present application is shown;

[0075] Figure 9 A structural schematic diagram of the terminal provided by the embodiment of the present application is shown. DETAILED DESCRIPTION

[0076] Exemplary embodiments of the present disclosure will be described hereinafter with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be embodied in various forms without being limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the present disclosure to those skilled in the art.

[0077] It should be understood that the actual dimensions of the various parts shown in the drawings can not be to scale.

[0078] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way limiting to the scope of the application or its applications or uses.

[0079] Techniques, methods, and apparatus known to those of ordinary skill in the relevant art can not be discussed in detail herein. However, where appropriate, such techniques, methods, and apparatus can be considered part of the present disclosure.

[0080] It should be noted that like reference numerals and letters refer to like items in the drawings and thus, once an item is defined in one drawing, it is not necessary to discuss it further in subsequent drawings.

[0081] Embodiments of the present application can be applied to a computer system / server, which can operate in connection with many other general purpose or special purpose computing system environments or configurations. Examples of well-known computing systems, environments, and / or configurations that can be suitable for use with computer system / server include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network personal computers, minicomputer systems, mainframe computer systems, and distributed cloud computing environments that include any of the above systems or the like.

[0082] The computer system / server can be described in the general context of computer system-executable instructions, such as program modules, being executed by a computer system. Generally, program modules can include routines, programs, objects, components, logic, data structures, and the like, which perform particular tasks or implement particular abstract data types. The computer system / server can be implemented in a distributed cloud computing environment where tasks are performed by remote processing devices that are linked through a communications network. In a distributed cloud computing environment, program modules can be located in local or remote computer system storage media including memory storage devices.

[0083] Embodiment one

[0084] The embodiments of the present application provide a radar shelter detection method, as shown in the method, the method comprises: Figure 1

[0085] 101, acquiring real-time point cloud frames collected by radars of different types.

[0086] As a current execution end, the radar shelter detection method provided in the embodiments of the present application can be applied to a control end of an unmanned vehicle or a service end of the unmanned vehicle, such as a vehicle control end of the unmanned vehicle or a remote service end of the unmanned vehicle, so as to detect the radars installed on the unmanned vehicle. Wherein, when performing real-time radar scanning, radars of different types will generate point cloud data in frames, that is, real-time point cloud frames, the real-time point cloud frames contain a plurality of point data, each point has three-dimensional coordinates to represent the radar scanning result.

[0087] 102, determining a target point cloud frame corresponding to the real-time point cloud frame based on a time threshold.

[0088] ​Wherein, since the real-time point cloud frame is obtained by real-time radar scanning at the current time, the current execution end can determine the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold value when detecting the generation message of the real-time point cloud frame, that is, searching for the target point cloud frame corresponding to the real-time point cloud frame according to the time length of the time threshold value. Specifically, since different radar types can scan different numbers of point cloud frames in one second when scanning, the point cloud frames obtained in the historical radar scanning time are stored according to the corresponding relationship of the time, so as to determine the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold value. For example, the time threshold value is 5 seconds, and at the time point of the real-time point cloud frame obtained by the radar scanning at the current time, 5 seconds are searched forward to determine the point cloud frame corresponding to the time point 5 seconds ago as the target point cloud frame.

[0089] It should be noted that the time threshold value of 5 seconds is only exemplary, and the time threshold value can be set according to the actual application scene, for example, it can be set to 1 second, 3 seconds, 6 seconds, 8 seconds and 10 seconds, etc., and the specific setting of the time threshold value is not limited in the present application.

[0090] 103, count the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame.

[0091] In the embodiment of the present application, since the real-time point cloud frame is obtained at the current time, and the target point cloud frame is searched according to the current time to find the point cloud frame meeting the time threshold value, there are multiple point cloud frames between the real-time point cloud frame and the target point cloud frame, so that the total number of point cloud frames and the corresponding number of abnormal point cloud frames can be counted. The total number of point cloud frames is the total number of normal point cloud frames and abnormal point cloud frames, so that the detection of the occlusion object is determined according to the ratio of the number of abnormal point cloud frames to the total number of point cloud frames and the abnormal frame number ratio threshold value. Specifically, the abnormal point cloud frame can be a point cloud frame marked as an abnormal state identifier by artificial marking, or a point cloud frame marked as an abnormal state identifier based on point cloud data, and the embodiment of the present application is not limited.

[0092] It should be noted that since different radar types have different numbers of scanning lines, the normal point number ratio is different when determining whether the state of the point cloud frame is normal based on the point cloud data, so the determination method of the abnormal point cloud frame of different radar types is different, for example, the three-dimensional coordinates xyz of the abnormal point of one type of radar are all null (NAN), and the three-dimensional coordinates xyz of the abnormal point of another type of radar are all 0, and the specific manifestation form of the abnormal point of different types of radar is not limited in the embodiment of the present application.

[0093] 104, if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than the abnormal frame number ratio threshold value, it is determined that the radar is occluded.

[0094] In the embodiment of the present application, when the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than the abnormal frame number ratio threshold, it indicates that the radar lens is blocked. The abnormal frame number ratio threshold can be pre-configured based on the radar blocking requirements, and is preferably 50%. That is, when the number of abnormal point cloud frames accounts for more than 50% of the total number of point cloud frames, it indicates that the radar is blocked or the radar lens is attached with a blocking object. It should be noted that the abnormal frame number ratio threshold corresponding to different types of radars can be the same or different. For example, the abnormal frame number ratio threshold of the first type of radar is set to 50%, the abnormal frame number ratio threshold of the second type of radar is set to 60%, and the abnormal frame number ratio threshold of the third type of radar is set to 70%. In actual application, the abnormal frame number ratio threshold can be set according to the characteristics of different types of radars, and the embodiment of the present application does not make further limitation on the abnormal frame number ratio threshold. Of course, the abnormal frame number ratio threshold of all types of radars can be set to 50%, 60%, 70%, 80%, etc.

[0095] In addition, in another embodiment of the present application, as shown in FIG. 2, after step S1 (acquiring real-time point cloud frames collected by different types of radars), before step S2 (determining target point cloud frames corresponding to the real-time point cloud frames based on a time threshold), the method further includes: Figure 2

[0096] 201, acquiring the number of normal points in the real-time point cloud frames collected by different types of radars, and a normal point number ratio threshold corresponding to the different types of radars;

[0097] 202, if the ratio of the number of normal points in the real-time point cloud frames to the total number of points is greater than the normal point number ratio threshold, determining the state identifier of the real-time point cloud frames as a normal point cloud frame;

[0098] 203, if the ratio of the number of normal points in the real-time point cloud frames to the total number of points is less than or equal to the normal point number ratio threshold, determining the state identifier of the real-time point cloud frames as an abnormal point cloud frame;

[0099] 204, storing the time stamp and the state identifier of the real-time point cloud frames in a target container in the time sequence of acquiring the real-time point cloud frames, wherein one point cloud frame corresponds to one time stamp and one state identifier.

[0100] ​In order to ensure that the search is based on the state identification of the point cloud frame between the real-time point cloud frame and the target point cloud frame, and to facilitate accurate timestamp search for accurate occlusion detection, the real-time point cloud frame collected by different radar types is obtained, and the point cloud frame is composed of a plurality of points, so that the normal point quantity of the real-time point cloud frame collected by different types of radars can be obtained, and the normal point quantity ratio threshold corresponding to different types of radars. Since the number of points in a generated point cloud frame is different when different types of radars are scanned, and the number of scanning lines of different radar types is also different, in order to meet the detection requirements of different radars, the normal point quantity ratio threshold for judging whether the point cloud frame is normal can be pre-configured or dynamically generated for different types of radars, so as to compare the ratio of the normal point quantity to the total point quantity in the real-time point cloud frame. The total point quantity is the total point quantity in the real-time point cloud frame, including the normal point quantity and the abnormal point quantity. A frame of point cloud data obtained by different radar scanning contains the three-dimensional coordinate points of each point, that is, there are a large number of three-dimensional point data in a point cloud frame, thereby forming a point cloud. For example, there are 10,000 three-dimensional points scanned in a point cloud frame, that is, 10,000 three-dimensional data points are included. At this time, for the three-dimensional points in the real-time point cloud frame obtained by normal radar scanning, if the three-dimensional coordinates x, y, z of the three-dimensional points have corresponding specific scanning values (non-0 or non-empty), it means that the three-dimensional point is a normal point, and if the three-dimensional coordinates x, y, z of the three-dimensional point are empty or 0, it means that the three-dimensional point is an abnormal point. Therefore, the total point quantity, the normal point quantity and the abnormal point quantity including the normal point and the abnormal point in the real-time point cloud frame can be determined based on the specific scanning value of the three-dimensional point, so as to compare the ratio of the normal point quantity to the total point quantity with the normal point quantity ratio threshold. If the ratio of the normal point quantity to the total point quantity in the real-time point cloud frame is greater than the normal point quantity ratio threshold, it means that most of the points in the real-time point cloud frame are normal points, so the state identification of the real-time point cloud frame is determined as a normal point cloud frame. If the ratio of the normal point quantity to the total point quantity in the real-time point cloud frame is less than or equal to the normal point quantity ratio threshold, it means that most of the points in the real-time point cloud frame are abnormal points, so the state identification of the real-time point cloud frame is determined as an abnormal point cloud frame. In addition, in the embodiment of the present application, the radar type can be distinguished based on different radar manufacturers, or based on different radar line numbers. For example, the normal point quantity ratio thresholds corresponding to the speed RS-bpearl radar, the Ouster radar and the DJI Livox radar are different. Preferably, the normal point quantity ratio threshold of the Ouster radar is 0.2, the normal point quantity ratio threshold of the speed RS-bpearl radar is 0.78, and the normal point quantity ratio threshold of the DJI livox radar is 0.5. The normal point quantity ratio threshold of different types of radars is not limited in the embodiment of the present application, and the normal point quantity ratio threshold of different types of radars is adapted to the characteristics of different types of radars.

[0101] It should be noted that, in order to accurately find the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold, when determining the abnormal point cloud frame or the normal point cloud frame corresponding to the real-time point cloud frame at different current time, the time stamp and the state identifier of the real-time point cloud frame are stored in the target container in the time sequence of obtaining the real-time point cloud frame. Wherein, one point cloud frame corresponds to one time stamp and one state identifier, that is, when storing in the target container, each point cloud frame is stored at the same time, the scanning time corresponding to the point cloud frame is stored as the time stamp, and the state identifier of the point cloud frame is stored together. In addition, the time stamp and the state identifier of the corresponding point cloud frame stored in the target container can be stored in the form of a data chain as shown in the figure, TimeStamp_n is the time stamp, Status_n is the state identifier of the point cloud frame (current real-time point cloud frame), and the target container can be a database or a storage unit with a preset storage capacity. The specific setting form of the target container is not limited by the present application. Figure 3

[0102] In another embodiment of the present application, in order to further limit and illustrate, the step of determining the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold comprises: taking the time stamp of the real-time point cloud frame as the starting point, the time threshold as the query length, and querying the target time stamp corresponding to the end point of the time threshold in the target container; determining the point cloud frame corresponding to the target time stamp as the target point cloud frame.

[0103] Since the target container has stored the time stamp corresponding to the real-time point cloud frame at different current time and the state identifier corresponding thereto, when determining the target point cloud frame, specifically, from the target container, taking the time stamp of the real-time point cloud frame as the starting point of the search, taking the time threshold as the query length, querying the target time stamp corresponding to the end point of the time threshold, at this time, the point cloud frame corresponding to the target time stamp is the target point cloud frame, therefore, the point cloud frame corresponding to the target time stamp is determined as the target point cloud frame. For example, taking the time stamp TimeStamp_n of the current frame as the starting point, taking the time threshold 5 seconds as the query length, and querying the target time stamp TimeStamp_0 corresponding to the end point of 5 seconds in the target container, then the point cloud frame corresponding to the target time stamp TimeStamp_0 is taken as the target point cloud frame of the time stamp TimeStamp_n of the current frame, and the number of abnormal point cloud frames and the total number of point cloud frames between the target point cloud frame of the target time stamp TimeStamp_0 and the real-time point cloud frame of the time stamp TimeStamp_n of the current frame are calculated.

[0104] ​It should be noted that, since the target container stores different timestamps and corresponding point cloud frame status identifiers in chronological order, when searching using a time threshold as the query length, if the data storage space within the target container can just fully store the timestamps and status identifiers within the time range corresponding to this time threshold, then the first frame timestamp in the target container can be used as the target timestamp. If the data storage space within the target container can store timestamps and status identifiers within a time range greater than the time threshold, then a timestamp other than the first frame in the target container can be used as the target timestamp, such as the third frame timestamp, the tenth frame timestamp, etc., determined solely based on the data storage space of the target container. This embodiment of the invention does not specifically limit the position of the first frame timestamp within the target container.

[0105] In another embodiment of the invention, for further definition and explanation, the steps further include: setting a capacity parameter threshold for the target container, wherein the capacity parameter threshold includes a time capacity parameter threshold or a frame count capacity parameter threshold; and updating the target container when the capacity of the target container reaches the corresponding capacity parameter threshold.

[0106] Since the data storage space of the target container can be greater than or equal to the data space storing the timestamps and status identifiers corresponding to the time threshold, a capacity parameter threshold for the target container can be preset to ensure that the total number of point cloud frames in the target container remains unchanged for statistical purposes. This capacity parameter threshold includes a time capacity parameter threshold or a frame count capacity parameter threshold, which limits the amount of point cloud frames and timestamps stored in the target container. In this embodiment of the invention, no specific limits are made on the time capacity parameter threshold or the frame count parameter threshold. When the capacity of the target container reaches the corresponding capacity parameter threshold, the target container is updated. At this time, the timestamps and status identifiers of the earliest moment (the first frame) can be deleted, and the timestamps and status identifiers of the latest moment (the current real-time point cloud frame) can be added to ensure that the total number of point cloud frames n in the target container remains constant. t The number remains unchanged. Specifically, if the earliest time frame (the first frame) is an anomalous point cloud frame, then the corresponding number of anomalous point cloud frames, n, is... a It will decrease by 1. If the latest time (the current real-time point cloud frame) is an abnormal point cloud frame, then the corresponding number of abnormal point cloud frames n a The value will be incremented by 1. If the latest time frame (the current real-time point cloud frame) is a normal point cloud frame, then the corresponding number of abnormal point cloud frames n will be incremented. a The value remains unchanged, thus based on the number of anomalous point cloud frames n. a With the total number of point cloud frames n t The ratio, i.e., based on n a / n t The radar is compared with the percentage of abnormal frames to determine whether it is blocked. This embodiment of the invention does not impose specific limitations.

[0107] It should be noted that, since the data storage space of the target container can be set, if the first frame timestamp in the target container is less than the time threshold from the current time range during the process of searching for the target timestamp based on the time threshold, it indicates that the number of state identifiers and timestamps stored in the target container does not reach the statistical requirement, therefore, the next time radar scanning is performed to obtain the next time timestamp and the state identifier of the real-time point cloud frame for storage, until the first frame timestamp is equal to the time threshold from the current time range, and then the statistics can be performed, and the embodiment of the present application is not limited in this regard.

[0108] For example, only the timestamps and the corresponding state identifiers in the last 5 seconds are stored in the target container, the time threshold is configured as 5 seconds, the vehicle is driven for 20 seconds, and the radar scans the point cloud for 10 frames per second, that is, 200 frames of point cloud frames are obtained. At the initial time, the timestamp frame1 is stored in the target container, and over time, the timestamps frame50 are successively stored. All the timestamps and the state identifiers between frame1 and frame50 are stored in the target container. If frame1 is the first frame, frame50 is 5 seconds away from frame1 in time, and the state identifiers of the 50 frames are counted. When the state identifier corresponding to the timestamp frame51 needs to be stored, the target container is updated, that is, the timestamp frame1 is deleted, and the target container stores frame2-frame51 at this time. The first frame is the timestamp frame2, and the timestamps between frame2 and frame51 are also 5 seconds, so as to continue the counting. Similarly, until the last frame is stored, that is, the timestamps frame151-frame200 are stored, the first frame timestamp in the target container is frame151, that is, the latest timestamp and the state identifier are stored, and then the first frame timestamp and the state identifier are deleted, so that the target container can store 50 frames of timestamps and state identifiers.

[0109] In another embodiment of the present application, in order to further limit and illustrate, the step of counting the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame comprises:

[0110] In the target container, the total number of point cloud frames between the real-time point cloud frame and the target point cloud frame is counted based on the number of timestamps or the number of state identifiers.

[0111] The number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame is counted based on the number of abnormal state identifiers.

[0112] Since different time stamps and corresponding state identifiers are stored in the target container, and different point cloud frames correspond to a time stamp uniquely, in order to accurately count the total number of point cloud frames and the number of abnormal point cloud frames, specifically, in the target container, after determining the target point cloud frame, the number of point cloud frames is counted based on the number of time stamps between the target time stamp of the target point cloud frame and the current time, or based on the number of identification states of the target point cloud frame and the real-time point cloud frame at the current time, to determine the total number of point cloud frames. At the same time, the number of abnormal point cloud frames in the target container can be directly counted based on the number of abnormal state identifiers, and the embodiments of the present application do not make specific limitations.

[0113] In another embodiment of the present application, in order to further limit and illustrate, as shown in Figure 4 The step further comprises:

[0114] 301. Obtain the test point cloud frame of the radar under the first preset shielding degree value and the second preset shielding degree value of any type of radar;

[0115] 302. Identify the abnormal points in the test point cloud frame, count the number of abnormal points in the test point cloud frame, take the difference between the total number of points in the test point cloud frame and the number of abnormal points as the number of normal points in the test point cloud frame, and take the ratio of the number of normal points to the total number of points as the normal point number ratio threshold under the corresponding preset shielding degree value.

[0116] 303. Take the intermediate value between the normal point number ratio threshold under the first preset shielding degree value and the normal point number ratio threshold under the second preset shielding degree value as the corresponding normal point number ratio threshold of the any type of radar.

[0117] Since the number of lines of different types of radars is different, the number of points in the obtained point cloud frame is different. In order to accurately determine the state of the point cloud frame, in addition to manually configuring the normal point number proportion threshold, the normal point number proportion threshold can also be configured based on the test mode, thereby improving the detection accuracy of the occluder. Specifically, first, the test point cloud frame of the radar of any type of radar under the first preset occlusion degree value and the second preset occlusion degree value is obtained. At this time, the first preset occlusion degree value and the second preset occlusion degree value are respectively used to represent different degrees of occlusion of the radar lens by the occluder, such as slight occlusion and serious occlusion, which correspond to different occlusion degree values. Among them, the first preset occlusion degree value and the second preset occlusion degree value obtained based on the test scene can be based on the occlusion of the radar lens by the occluder such as sludge and dirty water, as the degree values corresponding to slight occlusion and serious occlusion respectively, wherein the first preset occlusion degree value is less than the second preset occlusion degree value. In the test scene, for different occlusion degrees, the abnormal points in the test point cloud frame collected by the radar of any type are identified, the number of abnormal points is counted, and the number of normal points is obtained, that is, the difference between the total number of points in the test point cloud frame and the number of abnormal points. At this time, the ratio of the obtained number of normal points to the total number of points is taken as the normal point number proportion threshold under this occlusion degree, for example, for the first preset occlusion degree value representing the slight occlusion degree, the normal point number proportion threshold corresponding to the test point cloud frame collected is taken as the normal point number proportion threshold corresponding to the slight occlusion degree.

[0118] After obtaining the second preset occlusion degree value in the same way, in order to configure the normal point number proportion threshold for any type of radar, the intermediate value between the normal point number proportion threshold under the first preset occlusion degree value and the normal point number proportion threshold under the second preset occlusion degree value is taken as the normal point number proportion threshold corresponding to any type of radar. For example, the radar of type a obtains the normal point number proportion threshold a2 under the first occlusion degree value representing slight occlusion and the normal point number proportion threshold a3 under the second occlusion degree value representing serious occlusion in the test scene, and the intermediate value between a2 and a3 is taken as the normal point number proportion threshold of the radar of type a. Moreover, since the first preset occlusion degree value representing slight occlusion is less than the second preset occlusion degree value representing serious occlusion, and the more serious the occlusion degree in the point cloud frame, the more the number of abnormal points, the normal point number proportion threshold a2 corresponding to the first preset occlusion degree value is greater than the normal point number proportion threshold a3 corresponding to the second preset occlusion degree value.

[0119] In another embodiment of the present application, in order to further limit and illustrate, before the step of determining the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold, the method further comprises:

[0120] The radar with autonomous occlusion detection function is set as a specified type of radar;

[0121] identifying the type of radar acquiring the real-time point cloud frame;

[0122] If the type of the radar is identified as a specified type of radar, acquiring a fault code of the specified type of radar, and if the fault code indicates that the specified type of radar is in a non-occluded state, determining that the specified type of radar is in a non-occluded state.

[0123] If the fault code indicates that the specified type of radar is in an occluded state, repeating steps S2-S4 to determine whether the specified type of radar is occluded.

[0124] Since a small part of different types of radars already have the function of autonomous occlusion detection, for such radars, in order to further simplify the detection steps of the occlusion and improve the efficiency of the occlusion detection, and to meet the effectiveness of the radar detection of the occlusion for different types of radars, the radars with the function of autonomous occlusion detection are set as the specified type of radar, for example, the DJI Livox has the function of autonomous occlusion detection, and the DJI Livox is set as the specified type of radar. At this time, before performing step S2 in the embodiment of the present application, first, the type of the radar acquiring the real-time point cloud frame is identified based on the message name, message identification, etc. of the generation message of the real-time point cloud frame to determine whether the radar at this time is a radar with the function of autonomous occlusion detection. The type of the radar can be determined based on the message name, message identification, etc. of the generation message of the real-time point cloud frame, and the type of the radar includes but is not limited to Speedten RS-bpearl, Ouster and DJI Livox. When the type of the radar is identified as the specified type of radar, it means that the radar can detect the occlusion by itself, therefore, the fault code of the specified type of radar is acquired to determine whether the radar is occluded based on the fault code. In the embodiment of the present application, for the specified type of radar, if the fault code indicates that the specified type of radar is in a non-occluded state, it means that the radar is not occluded, and since the identification of the specified type of radar in a non-occluded state is usually accurate, at this time, it is determined that the specified type of radar is in a non-occluded state. However, when the fault code indicates that the specified type of radar is in an occluded state, in order to avoid the misjudgment of the occlusion detection caused by the design defects of the radar itself, light irradiation or scratches on the surface glass of the radar, steps S2-S4 in the embodiment of the present application are performed to determine whether the specified type of radar is occluded, thereby improving the detection accuracy of the occlusion.

[0125] In a specific application scenario, the specified type of radar can be DJI Livox, and the fault code of the DJI Lidar is acquired to determine that the radar is in a non-occluded state, and the occlusion detection in the embodiment of the present application can not be performed. If it is determined to be in an occluded state based on the fault code, steps S2-S4 in the embodiment of the present application are performed to improve the detection accuracy based on the occlusion detection of the radar in the embodiment of the present application.

[0126] In another embodiment of the invention, for further definition and explanation, after determining that the radar is obstructed, the method further includes: outputting obstruction fault alarm information to the control and safety subsystem, so that the control and safety subsystem controls the vehicle equipped with the radar to reduce its speed and notifies the safety officer to remove the obstruction; or controlling the safety subsystem to activate the corresponding automatic cleaning device.

[0127] To ensure timely removal of detected obstructions, once an obstruction is detected on the radar, an obstruction malfunction alarm is output to the control and safety subsystem. This allows the control and safety subsystem to control the vehicle equipped with the radar to slow down, preventing operation when the radar scan is unclear. Alternatively, the control and safety subsystem can activate a corresponding automatic cleaning device. This subsystem can be a subsystem with an automatic cleaning device installed on the vehicle, or a terminal that remotely controls the automatic cleaning device installed on the vehicle; this embodiment of the invention does not impose specific limitations. Furthermore, while outputting the obstruction malfunction alarm, to ensure safe vehicle operation, a safety operator can also be notified to remove the obstruction, thus ensuring vehicle safety.

[0128] This invention provides a method for detecting radar obstructions. Compared with existing technologies, this invention acquires real-time point cloud frames collected by different types of radars; determines the target point cloud frame corresponding to the real-time point cloud frame based on a time threshold; counts the total number of point cloud frames between the real-time point cloud frame and the target point cloud frame, as well as the number of abnormal point cloud frames; if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than the abnormal frame percentage threshold, then it is determined that the radar is obstructed. This method meets the obstruction detection requirements of different radars, reduces the material cost of installing obstruction detection hardware, thereby improving the detection efficiency and accuracy of radar obstructions and effectively addressing the cleaning requirements of vehicle radars in mining areas.

[0129] Example 2

[0130] Furthermore, as a response to the above Figure 1 The implementation of the method shown in this invention provides a radar obstruction detection device, such as... Figure 5 As shown, the device includes:

[0131] The acquisition module 41 is used to acquire real-time point cloud frames collected by different types of radar.

[0132] The first determining module 42 is used to determine the target point cloud frame corresponding to the real-time point cloud frame based on a time threshold.

[0133] a statistical module 43 configured to count a total number of point cloud frames and a number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame;

[0134] a second determination module 44 configured to determine that the radar is blocked if a ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than a threshold of an abnormal frame number ratio.

[0135] Further, the apparatus further comprises a third determination module, a storage module,

[0136] The acquisition module 41 is further configured to acquire a number of normal points in the real-time point cloud frame collected by the radar of the different type and a threshold of a normal point number ratio corresponding to the radar of the different type.

[0137] The third determination module is configured to determine a state identifier of the real-time point cloud frame as a normal point cloud frame if a ratio of the number of normal points in the real-time point cloud frame to a total number of points is greater than the threshold of the normal point number ratio, and determine the state identifier of the real-time point cloud frame as an abnormal point cloud frame if the ratio of the number of normal points in the real-time point cloud frame to the total number of points is less than or equal to the threshold of the normal point number ratio.

[0138] The storage module is configured to store the timestamps and the state identifiers of the real-time point cloud frames in a target container in a time sequence in which the real-time point cloud frames are acquired, wherein one point cloud frame corresponds to one timestamp and one state identifier.

[0139] Further, the first determination module 42 is specifically configured to query a target timestamp corresponding to an end point of the time threshold in the target container, taking the timestamp of the real-time point cloud frame as a starting point and the time threshold as a query length, and determine a point cloud frame corresponding to the target timestamp as a target point cloud frame.

[0140] Further, the statistical module 43 is specifically configured to count a total number of point cloud frames between the real-time point cloud frame and the target point cloud frame based on a number of timestamps or a number of state identifiers in the target container, and count a number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame based on a number of abnormal state identifiers.

[0141] Further, the acquisition module 41 comprises:

[0142] The acquisition unit is configured to acquire test point cloud frames of a radar of any type of radar at a first preset blocking degree value and a second preset blocking degree value, wherein the first preset blocking degree value is less than the second preset blocking degree value.

[0143] a statistics unit, configured to identify an abnormal point in the test point cloud frame, count a number of the abnormal points in the test point cloud frame, take a difference between a total number of points in the test point cloud frame and the number of the abnormal points as a number of normal points of the test point cloud frame, and take a ratio of the number of normal points to the total number of points as a normal point number ratio threshold under a corresponding preset occlusion degree value;

[0144] a determination unit, configured to take an intermediate value between the normal point number ratio threshold under the first preset occlusion degree value and the normal point number ratio threshold under the second preset occlusion degree value as a normal point number ratio threshold corresponding to the radar of the arbitrary type.

[0145] Further, the device further comprises:

[0146] a setting module, configured to set a capacity parameter threshold of the target container, the capacity parameter threshold including a time capacity parameter threshold or a frame number capacity parameter threshold;

[0147] an updating module, configured to update the target container when a capacity of the target container reaches a corresponding capacity parameter threshold.

[0148] Further, the device further comprises:

[0149] The setting module is further configured to set a radar with an autonomous occlusion detection function as a specified type radar.

[0150] The obtaining module is further configured to identify and obtain a radar type of the real-time point cloud frame through the real-time point cloud frame.

[0151] When the radar type is identified as the specified type radar, the obtaining module is further configured to obtain a fault code of the specified type radar, and determine that the specified type radar is in a non-occlusion state if the fault code indicates that the specified type radar is in the non-occlusion state, or repeat steps S2-S4 to determine whether the specified type radar is occluded if the fault code indicates that the specified type radar is in an occlusion state.

[0152] Further, the device further comprises:

[0153] an output module, configured to output an occlusion fault alarm information to a control safety subsystem, so that the control safety subsystem controls a vehicle installed with the radar to perform a speed reduction process and informs a safety officer to perform a clearing process.

[0154] It should be noted that the structure, function implementation and technical effects of the radar occlusion object detection device correspond one-to-one to the implementation steps of the radar occlusion object detection method, and the same content will not be described again.

[0155] The radar shelter detection device provided by the embodiment of the present application can meet the shelter detection requirements of different radars, reduce the manpower and material resources cost of shelter detection hardware installation, improve the detection efficiency and accuracy of the radar shelter, and achieve the effective cleaning requirement of the mine vehicle radar.

[0156] Embodiment three

[0157] Figure 6 A structural schematic diagram of a computer readable storage medium provided by the embodiment of the present application is shown in FIG. 5, which is a computer readable storage medium 500 storing a computer program 510. When the computer program 510 is executed by a processor, the radar shelter detection method described in the embodiment one is implemented. The radar shelter detection method has been described in detail in the embodiment one, and will not be described here. Figure 6 The method described in the above embodiments can be implemented by software, hardware, firmware or any combination thereof, in whole or in part. The computer readable medium 500 can include computer storage medium and communication medium, and can also include any medium that can transfer a computer program from one place to another. The storage medium can be any target medium that can be accessed by a computer.

[0158] As a possible design, the computer readable medium 500 can include a compact disc read-only memory (CD-ROM), RAM, ROM, EEPROM or other optical disk storage; the computer readable medium can include a magnetic disk storage or other magnetic disk storage device. Moreover, any connection line can also be appropriately referred to as a computer readable medium. For example, if software is transmitted from a website, server or other remote source using a coaxial cable, optical fiber cable, twisted pair, DSL or wireless technology (such as infrared, radio and microwave), the coaxial cable, optical fiber cable, twisted pair, DSL or wireless technology (such as infrared, radio and microwave) is included in the definition of the medium. As used herein, the disk and the optical disk include a compact disc (CD), a laser disc, an optical disc, a digital versatile disc (DVD), a floppy disk and a Blu-ray disc, wherein the disk is usually reproduced by magnetism, and the optical disk is optically reproduced by laser.

[0159]

[0160] ​Embodiment Four

[0161] Figure 7 A structural schematic diagram of a computer device provided for an embodiment of the present application is shown in FIG. 6. The computer device 600 includes a memory 610, a processor 620, and a computer program stored in the memory 610 and executable by the processor 620. When the processor 620 executes the computer program 640, it performs the steps of the method of the present application, and the occlusion on the radar lens can be accurately detected. It should be noted that the computer program 440 in this embodiment is the same as the computer program 310. Figure 7

[0162] The memory 610 can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. The memory 620 has a storage space 630 for storing the computer program 640 for executing any of the steps of the above-mentioned methods. The computer program 640 can be read from or written to one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disc (CD), a memory card, or a floppy disk. Such a computer program product is typically a computer-readable storage medium such as that described above. The computer device can include multiple processors, each of which can be a single-CPU processor or a multi-CPU processor. The processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions). Figure 6

[0163] Embodiment Five

[0164] Figure 8 A structural schematic diagram of a chip provided for an embodiment of the present application is shown in FIG. 8. The chip 800 includes one or more (including two) processors 810 and a communication interface 830. The communication interface 830 and the at least one processor 810 are coupled, and the at least one processor 810 is configured to run a computer program or instructions to implement the method for detecting the radar occlusion as described in the above-mentioned method embodiments. Figure 8

[0165] Preferably, the memory 840 stores the following elements: executable modules or data structures, or a subset thereof, or an extended set thereof.

[0166] ​​​The memory 840 can include read-only memory and random access memory, and provide instructions and data to the processor 810. A portion of the memory 840 can also include non-volatile random access memory (NVRAM).

[0167] In the embodiments of the present application, the memory 840, the communication interface 830 and the memory 840 are coupled together through the bus system 820. Among them, the bus system 820 can include not only a data bus, but also a power bus, a control bus and a status signal bus, etc. For the convenience of description, all kinds of buses are marked as the bus system 820. Figure 8

[0168] The method described in the above embodiments of the present application can be applied in the processor 810 or implemented by the processor 810. The processor 810 can be an integrated circuit chip with processing capability. In the implementation process, each step of the above method can be completed by the integrated logic circuit or the instruction in the form of software in the processor 810. The processor 810 described above can be a general processor (for example, a microprocessor or a conventional processor), a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices or discrete hardware components, and the processor 810 can implement or execute the methods, steps and logical block diagrams disclosed in the embodiments of the present application.

[0169] Embodiment six

[0170] Figure 9 A structural schematic diagram of a terminal provided by the embodiments of the present application is shown in FIG. 9. As shown in FIG. 9, the terminal 900 includes the radar shelter detection device 910 as described. Figure 9

[0171] The terminal 900 described above can execute the method described in the above embodiments through the radar shelter detection device 910. It can be understood that the implementation manner of the terminal 900 controlling the radar shelter detection device 910 can be set according to the actual application scene, and the embodiments of the present application are not limited specifically.

[0172] ​​The terminal 900 includes, but is not limited to, a server, a vehicle, a vehicle terminal, a vehicle controller, a vehicle module, a vehicle module, a vehicle component, a vehicle chip, a vehicle unit, a vehicle radar, or a vehicle camera and other sensors. The vehicle can implement the method provided by the present application through the vehicle terminal, vehicle controller, vehicle module, vehicle module, vehicle component, vehicle chip, vehicle unit, vehicle radar, or camera.

[0173] The vehicle in the embodiments of the present application includes a passenger car and a commercial vehicle. Common models of the commercial vehicle include, but are not limited to, a pickup truck, a micro truck, a light truck, a minibus, a self-loading truck, a truck, a tractor, a trailer, a special-purpose vehicle, and a mining vehicle. The mining vehicle includes, but is not limited to, a mining truck, a wide-body vehicle, an articulated vehicle, a shovel, an electric shovel, and a bulldozer. The embodiments of the present application do not further limit the type of the vehicle, and any type of vehicle is within the protection scope of the embodiments of the present application.

[0174] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present application can be realized by a general computing device, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices. Alternatively, they can be realized by program codes executable by a computing device, so that they can be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described can be executed in an order different from that described herein, or they can be manufactured into individual integrated circuit modules, or multiple modules or steps thereof can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

[0175] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting radar obstructions, characterized in that, include: S1: Acquire real-time point cloud frames collected by different types of radar; S2: Determine the target point cloud frame corresponding to the real-time point cloud frame based on a time threshold, wherein the target point cloud frame is determined according to the query length of the time threshold; S3: Count the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame. The total number of point cloud frames and the number of abnormal point cloud frames are counted based on the point cloud frames between the real-time point cloud frame and the target point cloud frame. S4: If the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than the threshold for the proportion of abnormal frames, then it is determined that the radar is blocked.

2. The method according to claim 1, characterized in that, Before determining the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold, the method further includes: The number of normal points in real-time point cloud frames collected by different types of radars is obtained, as well as the threshold for the percentage of normal points corresponding to different types of radars. If the ratio of the number of normal points to the total number of points in the real-time point cloud frame is greater than the threshold for the proportion of normal points, then the status identifier of the real-time point cloud frame is determined to be a normal point cloud frame. If the ratio of the number of normal points to the total number of points in the real-time point cloud frame is less than or equal to the normal point count percentage threshold, then the status identifier of the real-time point cloud frame is determined to be an abnormal point cloud frame. The timestamps and status identifiers of the real-time point cloud frames are stored in the target container according to the time sequence of acquisition of the real-time point cloud frames, wherein one point cloud frame corresponds to one timestamp and one status identifier.

3. The method according to claim 2, characterized in that, The step of determining the target point cloud frame corresponding to the real-time point cloud frame based on a time threshold includes: Starting from the timestamp of the real-time point cloud frame and using the time threshold as the query length, the target timestamp corresponding to the end point of the time threshold is queried in the target container. The point cloud frame corresponding to the target timestamp is determined as the target point cloud frame.

4. The method according to claim 3, characterized in that, The statistics on the total number of point cloud frames between the real-time point cloud frame and the target point cloud frame, as well as the number of abnormal point cloud frames, include: In the target container, based on the number of timestamps or the number of status identifiers, the total number of point cloud frames between the real-time point cloud frame and the target point cloud frame is counted. Based on the number of abnormal state identifiers, the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame is counted.

5. The method according to claim 2, characterized in that, The threshold for obtaining the percentage of normal points corresponding to different types of radar includes: Obtain test point cloud frames of any type of radar under a first preset occlusion degree value and a second preset occlusion degree value, wherein the first preset occlusion degree value is less than the second preset occlusion degree value; Identify abnormal points in the test point cloud frame, count the number of abnormal points in the test point cloud frame, take the difference between the total number of points in the test point cloud frame and the number of abnormal points as the number of normal points in the test point cloud frame, and take the ratio of the number of normal points to the total number of points as the threshold of the proportion of normal points under the corresponding preset occlusion degree value. The midpoint between the normal point percentage threshold under the first preset occlusion level and the normal point percentage threshold under the second preset occlusion level is used as the normal point percentage threshold for any type of radar.

6. The method according to claim 2, characterized in that, The method further includes: Set the capacity parameter threshold of the target container, which includes a time capacity parameter threshold or a frame count capacity parameter threshold; When the capacity of the target container reaches the corresponding capacity parameter threshold, the target container is updated.

7. The method according to any one of claims 1-6, characterized in that, Before determining the target point cloud frame corresponding to the real-time point cloud frame based on the time threshold, the method further includes: Set the radar with autonomous obstruction detection function as a designated type of radar; The radar type of the real-time point cloud frame is obtained by identifying the real-time point cloud frame. When the radar type is identified as a specified type radar, the fault code of the specified type radar is obtained. If the fault code indicates that the specified type radar is in an unobstructed state, then the specified type radar is determined to be in an unobstructed state. If the fault code indicates that the specified type of radar is blocked, repeat steps S2-S4 to determine whether the specified type of radar is blocked.

8. The method according to claim 1, characterized in that, After determining that the radar is blocked, the method further includes: The system outputs an obstruction fault alarm message to the control and safety subsystem, so that the control and safety subsystem controls the vehicle equipped with the radar to reduce its speed and notifies the safety officer to clear the obstruction; or it controls the safety subsystem to activate the corresponding automatic cleaning device.

9. A radar obstruction detection device, characterized in that, include: The acquisition module is used to acquire real-time point cloud frames collected by different types of radar. The first determining module is used to determine the target point cloud frame corresponding to the real-time point cloud frame based on a time threshold, wherein the target point cloud frame is determined according to the query length of the time threshold; The statistics module is used to count the total number of point cloud frames and the number of abnormal point cloud frames between the real-time point cloud frame and the target point cloud frame. The total number of point cloud frames and the number of abnormal point cloud frames are counted based on the point cloud frames between the real-time point cloud frame and the target point cloud frame. The second determining module is used to determine that the radar is blocked if the ratio of the number of abnormal point cloud frames to the total number of point cloud frames is greater than the threshold for the proportion of abnormal frame count.

10. A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method as claimed in any one of claims 1-8.

11. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor; wherein, when the processor executes the computer program, it implements the steps of the method as claimed in any one of claims 1-8.

12. A chip, characterized in that, The chip includes at least one processor and a communication interface, the communication interface being coupled to the at least one processor, the at least one processor being used to run computer programs or instructions to implement the radar obstruction detection method as described in any one of claims 1-8.

13. A terminal, characterized in that, The terminal includes the radar obstruction detection device as described in claim 9.

Images