CN111429400A - A method, device, system and medium for detecting dirt on a lidar window - Google Patents

Abstract

The embodiment of the invention provides a method, a device, a system and a medium for detecting dirt of a laser radar window, wherein the method comprises the following steps: acquiring point cloud data obtained by scanning the laser radar; the point cloud data comprises reflected light intensity; identifying the point cloud data to determine each obstacle; determining the obstacles in a preset distance range of the window position of the laser radar in all the obstacles as suspicious obstacles; and if the intensity of at least one reflected light in each suspicious shelter is greater than the first preset light intensity, determining that the shelter exists on the window of the laser radar. The embodiment of the invention provides a method, a device, a system and a medium for detecting laser radar window dirt, so as to realize detection of the laser radar window dirt.

Description

A method, device, system and medium for detecting dirt on a lidar window

technical field

The present invention relates to laser radar technology, in particular to a method, device, system and medium for detecting dirt on a laser radar window.

Background technique

The external structure of the lidar will have a mirror structure (or called a window, a filter cover, etc.). The mirror structure is mainly used to filter the interference light in the non-radar working laser band, and also to protect the internal devices. In the harsh working environment, the mirror structure will be polluted due to various conditions, and the pollution source is usually dust, soil, water droplets or other liquid solids. Contaminated mirrors will affect the emission and reception of radar lasers, resulting in errors in the ranging of radar lasers, and the radar cannot work properly.

The traditional lidar cleaning mechanism needs to manually control the cleaning mechanism to start working to realize the cleaning of the lidar. This need to rely on manpower, which will increase the labor cost, and because the manual inspection is usually performed periodically, it will lead to the cleaning of the lidar. The work cannot be carried out in time. Alternatively, it is necessary to use corresponding hardware equipment, such as a special lidar mirror pollution degree sensing device, to sense the pollution degree of the lidar mirror surface, and then determine whether to open the cleaning mechanism. This method requires the help of other equipment, which increases the cost and has a complicated structure.

SUMMARY OF THE INVENTION

Embodiments of the present invention provide a method, device, system, and medium for detecting dirt on a lidar window, so as to detect dirt on a lidar window.

In a first aspect, an embodiment of the present invention provides a method for detecting dirt on a lidar window, including:

acquiring point cloud data obtained by scanning the lidar; the point cloud data includes reflected light intensity;

Identifying the point cloud data to determine each obstacle;

Determining the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders;

If the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, it is determined that there is a blocking object on the window of the lidar.

In a second aspect, an embodiment of the present invention provides a device for detecting dirt on a lidar window, including:

a point cloud data acquisition module for acquiring point cloud data obtained by scanning the lidar; the point cloud data includes reflected light intensity;

an obstacle identification module, used for identifying the point cloud data to determine each obstacle;

a suspicious occluder determination module, configured to determine the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders;

A blocker determination module, configured to determine that there is a blocker on the window of the lidar if the reflected light intensity of at least one of the suspicious blockers is greater than the first preset light intensity.

In a third aspect, an embodiment of the present invention provides a laser radar cleaning system, including:

A lidar cleaning mechanism for cleaning the lidar window; and

A cleaning control device is connected to the lidar cleaning mechanism, and the cleaning control device includes a processor and a memory; a computer program is stored in the memory, so that when the processor executes the computer program, the first aspect is realized. method described.

In a fourth aspect, an embodiment of the present invention provides a computer storage medium on which a computer program is stored, and when the program is executed by a processor, implements the method described in the first aspect.

In the method for detecting dirt on a lidar window provided by the embodiment of the present invention, each obstacle is acquired according to the point cloud data scanned by the lidar, and the preset distance of each obstacle located at the window position of the lidar is calculated. Obstacles within the range are determined as suspicious occluders, and then according to the intensity of reflected light to determine whether there are occluders in each suspicious occluder, when the reflected light intensity of at least one suspicious occluder is greater than the first preset light intensity, it is determined that there is an occluder , to realize the judgment of radar window pollution.

Description of drawings

1 is a flowchart of a method for detecting dirt on a lidar window in Embodiment 1 of the present invention;

2 is a flowchart of a method for detecting dirt on a lidar window in Embodiment 2 of the present invention;

3 is a flowchart of a method for detecting dirt on a lidar window according to Embodiment 3 of the present invention;

4 is a flowchart of a method for detecting dirt on a lidar window according to Embodiment 4 of the present invention;

5 is a schematic structural diagram of a device for detecting dirt on a lidar window according to Embodiment 5 of the present invention;

6A is a schematic structural diagram of a laser radar cleaning system according to Embodiment 6 of the present invention;

6B is a schematic structural diagram of a cleaning control device in a laser radar cleaning system according to Embodiment 6 of the present invention.

Detailed ways

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention. In addition, it should be noted that, for the convenience of description, the drawings only show some but not all structures related to the present invention.

Example 1

1 is a flowchart of a method for detecting dirt on a lidar window according to Embodiment 1 of the present invention. This embodiment is applicable to the case of detecting dirt on a lidar window from point cloud data collected by lidar. The method may be executed by the cleaning control device in the laser radar cleaning system according to the embodiment of the present invention, and the processing device may be implemented by software and/or hardware. As shown in Figure 1, the method specifically includes the following steps:

S101. Acquire point cloud data scanned by a lidar; the point cloud data includes reflected light intensity.

The point cloud data may be a collection of three-dimensional coordinate vectors recorded in the form of point clouds when the lidar scans the scene in which it is located, and each three-dimensional coordinate vector may be represented by (x, y, z). The point cloud data may include multiple data points, and the point cloud data may also include the reflected light intensity value of each data point.

S102: Identify and determine each obstacle based on the point cloud data.

Optionally, there may be many methods for identifying point cloud data, which are not limited in this embodiment. The clustering method may be used to perform clustering processing on the point cloud data of the current frame to determine at least one obstacle in the current point cloud data. It is also possible to input the point cloud data of the current frame into the target detection model based on a pre-trained deep learning model, and run the deep learning model to obtain at least one obstacle corresponding to the point cloud data of the current frame. Obstacles can be, for example, dynamic obstacles such as pedestrians, vehicles, and animals on the road, static obstacles such as street lights, signs, and trash cans on the road, or occluders on the lidar window. Among them, the obstruction on the lidar window is the dirt.

S103: Determine the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders.

Wherein, each obstacle may have different distances, and the distance here refers to the distance between the obstacle and the center point of the lidar (that is, the origin of the lidar coordinate system). The window of the lidar is fixed at the position of the lidar, so the distance of the window relative to the center of the lidar is known. Therefore, by judging whether the distance of the obstacle is the same as or close to the distance of the window, it can be roughly determined whether the obstacle is located on the window or is an environmental obstacle independent of the window. It can be understood that when the distance of the obstacle is within the preset distance range of the window position, it can also be preliminarily determined to be a suspicious obstacle, thereby avoiding the problem of missed detection caused by measurement errors and improving the accuracy of detection. For example, if the position of the viewing window is 60mm, the obstacles within a distance of ±0.5mm can be preliminarily identified as being located on the viewing window, that is, obstacles with a distance of 59.5mm~60.5mm can be preliminarily identified as being located in the viewing window. , so that it is determined to be a suspicious occluder. It can be understood that the preset distance deviation can be set according to requirements such as the size of the lidar and the required accuracy, and is not limited to the above examples.

Exemplarily, there may be many methods for obtaining the distance of the obstacle, which is not limited in this embodiment. For example, a time-of-flight method, a phase method, a frequency difference method, or a trigonometry method can be used.

S104. If the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, it is determined that there is a blocking object on the window of the lidar.

It is understandable that for the part of the lidar window that is not blocked by the obstruction, the laser beam emitted by the lidar can be emitted through the window, and the light intensity reflected by the window is small; the part of the radar window that is blocked by the obstruction, The laser beam emitted by the lidar is reflected back by the viewing window with a high intensity.

In this step, if the reflected light intensity of a suspicious blocking object is greater than the first preset light intensity, it is determined that there is a blocking object on the window of the lidar. Optionally, while determining whether there are obstructions, the number of obstructions may also be determined, that is, the number of obstructions determined by statistics. For example, if the reflected light intensity of N suspicious obstructions is greater than the first preset light intensity, it may be determined that there are N obstructions on the window of the lidar. N is an integer greater than or equal to 1.

Optionally, if the reflected light intensity of any one of the suspicious obstructions is less than or equal to the first preset light intensity, it is determined that there is no obstruction on the window of the lidar.

Exemplarily, the first preset light intensity is 1uw. The first preset light intensity can be set according to the reflected light intensity formed when there is no obstruction on the viewing window. The reflected light intensity can be obtained by testing in advance. In other embodiments, possible contaminants may also be determined according to the application environment of the lidar, and then set according to the reflected light intensity range that can be formed when the contaminants are on the window.

Exemplarily, in this step, the suspicious occluder is formed by clustering a plurality of data points, and the reflected light intensity of the suspicious occluder corresponds to the average value of the reflected light intensities of the multiple data points of the suspicious occluder. In other embodiments, the maximum value, the minimum value or the median value of the reflected light intensities of a plurality of data points corresponding to the suspicious occluder may also be used as the reflected light intensity of the suspicious occluder. It can be understood that, compared with using the maximum value, the minimum value or the median value, the method of using the average value can better reflect the overall reflection situation of the suspicious occluder.

In the method for detecting dirt on a lidar window provided by the embodiment of the present invention, each obstacle is acquired according to the point cloud data scanned by the lidar, and the preset distance of each obstacle located at the window position of the lidar is calculated. Obstacles within the range are determined as suspicious occluders, and then according to the intensity of reflected light to determine whether there are occluders in each suspicious occluder, and when the reflected light intensity of at least one suspicious occluder is greater than the first preset light intensity, determine the lidar There is an obstruction on the window. The method for detecting dirt on a lidar window provided by the embodiment of the present invention directly utilizes the point cloud data collected during the working process of the lidar to perform dirt detection without manual detection or special dirt detection equipment. It can judge the dirt of the lidar window, the cost is low, and it is beneficial to realize the miniaturization of the lidar.

Optionally, after step S104, if the ratio of the data points of the occluders in the point cloud data is greater than the preset ratio, a cleaning instruction is generated to control the cleaning mechanism to clean the window. Among them, the cleaning mechanism refers to the cleaning mechanism of the lidar, and the cleaning mechanism is used to clean and remove the dirt on the lidar window. Therefore, this step can also be based on the above-mentioned implementable manner, not only can detect whether there is dirt on the lidar window, but also can automatically detect whether the ratio of the data points of the occluder in the point cloud data is greater than the preset ratio. Start the cleaning mechanism to clean the lidar window to remove the obstructions on the window, without relying on manpower for cleaning, saving manpower, and ensuring the timeliness of cleaning the lidar windows. The size of the occluder is determined by the proportion of the data points of the occluder in the entire point cloud, so that when it is larger than a certain proportion, the cleaning mechanism is controlled to clean, which can avoid the frequent triggering of the system due to the existence of clutter on the window. Clean the mechanism and interfere with the normal operation of the lidar.

Exemplarily, the proportion of the data points of the occluders in the point cloud data is greater than 30%, and there are many occluders on the lidar window, which has a great impact on the work of the lidar, and the occluders need to be cleaned up. The cleaning command is used to control the cleaning mechanism to clean the window. The proportion of the data points of the occluder in the point cloud data is less than or equal to 30%, and the occluder is relatively small on the lidar window, which has little impact on the work of the lidar. It is not necessary to clean the occluder, and no cleaning command is generated.

Optionally, in the embodiment of the present invention, there are many methods for determining obstacles located within a preset distance range of the window position of the lidar as suspicious occluders, which are not limited in this embodiment.

A possible implementation manner may be: acquiring multi-frame point cloud data collected by the lidar scanning multiple times. Obstacles with similar position distributions in at least two frames of adjacent point cloud data within the preset distance range of the lidar window position among the obstacles are determined as suspicious occluders. Since the position of the occluder on the lidar window is fixed, on the basis of the preset distance range of each obstacle located in the lidar window position, the same obstacle is combined with adjacent points in at least two frames. The cloud data has similar location distribution features, which can improve the accuracy of suspicious occluder judgment, and thus can improve the accuracy of occluder judgment. In this way, it is also possible to avoid the problem of triggering the cleaning mechanism to perform cleaning by determining those obstacles that are momentarily on the window and then fall off as suspicious obstructions, which are then confirmed as obstructions.

Exemplarily, in the first frame of point cloud data, the obstacle A is located within the preset distance range of the window position of the lidar, and the obstacle A is located at position 1 in the upper left corner of the point cloud. In the second frame of point cloud data, obstacle A is located within the preset distance range of the lidar window position, and obstacle A is located at position 1 in the upper left corner of the point cloud. It can be understood that the so-called position 1 can be completely the same as the position 1, or a certain deviation can be allowed. As long as the deviation is within the allowable range, it can be determined that the obstacle is located at position 1, then the obstacle A is determined as Suspicious obstruction. In the first frame of point cloud data, obstacle B is located within the preset distance range of the lidar window position, and obstacle B is located at position 1 in the upper left corner of the point cloud. In the second frame of point cloud data, the obstacle B is beyond the preset distance range of the window position of the lidar, and the obstacle B is located at position 1 in the upper left corner of the point cloud. It is understandable that the obstacle B may originally be an obstacle on the window but it falls off automatically, or the obstacle B is a dynamic obstacle, so the obstacle B is not a suspicious obstacle. In the first frame of point cloud data, the obstacle C is located in the preset distance range of the window position of the lidar, and the obstacle C is located at position 1 in the upper left corner of the point cloud. In the second frame of point cloud data, obstacle C is located within the preset distance range of the lidar window position, and obstacle C is located at position 2 in the upper right corner of the point cloud, then obstacle C is not a suspicious occluder.

Another possible implementation may be: if each obstacle is located within a preset distance range of the window position of the lidar and the distance deviation in at least two consecutive frames of point cloud data is less than the first distance, the obstacle is determined to be suspicious cover. Since the occluder on the lidar window is on the window and its position is fixed, based on the preset distance range of each obstacle located on the lidar window position, combine the same obstacle in at least two adjacent frames. The feature that the distance deviation in the point cloud data is smaller than the first distance can improve the accuracy of judging suspicious occluders, and therefore can improve the accuracy of judging occluders.

Exemplarily, in the first frame of point cloud data, the obstacle D is located within a preset distance range of the window position of the lidar, and the obstacle D is located at a distance 1 from the window position. In the second frame of point cloud data, obstacle A is located within the preset distance range of the window position of the lidar, and obstacle D is located at distance 2 from the window position. The difference between distance 1 and distance 2 is the distance deviation of obstacle D in the first frame of point cloud data and the second frame of point cloud data. When the distance deviation is less than the first distance, the obstacle is determined as a suspicious occluder. . When the distance deviation is greater than or equal to the first distance, the obstacle is not a suspicious occluder. Exemplarily, the first distance is 1 cm.

Optionally, in step S103, the methods described in the above two possible embodiments may be used simultaneously to further improve the accuracy of the judgment.

Embodiment 2

FIG. 2 is a flowchart of a method for detecting dirt on a lidar window in Embodiment 2 of the present invention. This embodiment is based on the above-mentioned embodiment, and further optimization is carried out, and it is specifically given how to use the reflection of the occluder. The introduction of light intensity to judge the type of occluder. As shown in Figure 2, the operation process includes the following steps:

S201. Acquire point cloud data scanned by the lidar; the point cloud data includes reflected light intensity.

S202. Identify and determine each obstacle on the point cloud data.

S203: Determine the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders.

S204. If the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, it is determined that there is a blocking object on the window of the lidar.

S205. If the reflected light intensity of the shielding object is greater than the first preset light intensity and less than or equal to the second preset light intensity, the shielding object is a partially transparent shielding object, and if the reflected light intensity of the shielding object is greater than the second preset light intensity , the shielding object is an opaque shielding object; wherein, the second preset light intensity is greater than the first preset light intensity.

Wherein, the reflected light intensity of the blocking object is the reflected light intensity of the suspicious blocking object determined as the blocking object in step S204. The light transmittance of the partially transparent shade is greater than the light transmittance of the opaque shade, and the reflectivity of the partially transparent shade is lower than that of the opaque shade.

Exemplarily, a partially transparent shield refers to a shield with a transmittance greater than 30% and less than 60%. The partially transparent cover can be, for example, rain, snow, or the like. An opaque shade refers to a shade with a transmittance of less than or equal to 30%. For example, the opaque shield can be dust, sundries and the like.

Exemplarily, the second preset light intensity is 3uw. The second preset light intensity may be determined according to the reflection intensity of dirt that may exist in the environment where the lidar is actually used, and the like.

In this embodiment of the present invention, after the blocking objects are determined in step S204, the type of blocking objects can also be judged according to the reflected light intensity of each blocking object. Specifically, the reflected light intensity is greater than the first preset light intensity and less than or equal to The shielding object with the second preset light intensity is a partially transparent shielding object, and the shielding object with the reflected light intensity greater than the second preset light intensity is an opaque shielding object. In the embodiment of the present invention, since the type of the obstruction can be judged, the corresponding type of cleaning can be carried out according to the corresponding type of obstruction, for example, dust (opaque obstruction) is cleaned accordingly, and water droplets ( Partially transparent coverings) are dewatered and cleaned accordingly.

Optionally, in step S205, if the reflected light intensity of the shielding object is greater than the first preset light intensity and less than or equal to the second preset light intensity, the shielding object is a partially transparent shielding object, including: if the reflected light intensity of the shielding object is If it is greater than the first preset light intensity and less than or equal to the second preset light intensity, and each occluder is evenly distributed in the point cloud data, the occluder is a partially transparent occluder. Since some transparent obstructions such as rain and snow are often evenly distributed in all positions of the lidar window, on the basis that the reflected light intensity of the obstructions is greater than the first preset light intensity and less than or equal to the second preset light intensity, combined with The uniform distribution of some transparent occluders can further increase the accuracy of judgment.

Optionally, after step S205, if the ratio of the data points of the occluders in the point cloud data is greater than the preset ratio, a cleaning instruction is generated to control the cleaning mechanism to clean the window. Among them, the cleaning mechanism refers to the cleaning mechanism of the lidar, and the cleaning mechanism is used to clean and remove the dirt on the lidar window. Therefore, this step can also be based on the above-mentioned implementable manner, not only can detect whether there is dirt on the lidar window, but also can automatically detect whether the ratio of the data points of the occluder in the point cloud data is greater than the preset ratio. Start the cleaning mechanism to clean the lidar window to remove the obstructions on the window, without relying on manpower for cleaning, saving manpower, and ensuring the timeliness of cleaning the lidar windows.

Embodiment 3

FIG. 3 is a flowchart of a method for detecting dirt on a lidar window in Embodiment 3 of the present invention. This embodiment is based on the above-mentioned embodiment and further optimized, and specifically provides how to detect the dirt on the lidar window according to the An introduction to the situation that the temperature automatically heats and cleans the shelter. As shown in Figure 3, the operation process includes the following steps:

S301. Acquire point cloud data scanned by the lidar; the point cloud data includes reflected light intensity.

S302. Identify and determine each obstacle based on the point cloud data.

S303 , determining an obstacle located within a preset distance range of a window position of the lidar among the obstacles as a suspicious occluder.

S304 , if the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, determine that there is a blocking object on the window of the lidar.

S305. Acquire the temperature of the window of the lidar.

Exemplarily, the temperature of the window of the lidar can be acquired by a temperature sensor installed near the window.

S306. When the temperature of the window of the lidar is less than the first preset temperature value, heat the window of the lidar.

Optionally, in this step, when the temperature of the window of the lidar is lower than the first preset temperature, a heating instruction is activated to heat the window of the lidar. When the temperature of the window of the lidar is greater than or equal to the first preset temperature, the heating command is not activated. There may be many methods for heating the window, which are not limited in this embodiment of the present invention. Specifically, for example, the window of the lidar can be heated by means of heat conduction or heat convection.

Exemplarily, the first preset temperature is 4 degrees Celsius. When the temperature is less than 4 degrees Celsius, water vapor is easy to condense on the lidar window to form water droplets. Therefore, when the temperature of the window of the lidar is less than 4 degrees Celsius, the window of the lidar is heated, so that water vapor does not easily form water vapor on the window. On the other hand, heating the lidar window can also heat and evaporate the water droplets that have formed, thereby removing the water droplets formed on the window. If the window is covered with snow, heating the lidar window can also melt the snow to form water droplets that eventually evaporate, removing the snow from the window.

S307. When the temperature of the window of the lidar is greater than the second preset temperature value, stop heating the window of the lidar; the second preset temperature value is greater than the first preset temperature value.

Optionally, in this step, when the temperature of the window of the lidar is greater than the second preset temperature, an instruction to turn off the heating is started to stop heating the window of the lidar. When the temperature of the lidar window is greater than or equal to the first preset temperature and less than or equal to the second preset temperature, the instruction to turn off the heating is not started, and the lidar window continues to be heated.

Exemplarily, the second preset temperature is 5 degrees Celsius. When the temperature is greater than 5 degrees Celsius, stop heating the lidar window.

Optionally, before step S305 (obtaining the temperature of the lidar window), the method for detecting dirt on the lidar window may further include: if the reflected light intensity of the obstruction is greater than the first preset light intensity and less than or equal to the second If the light intensity is preset, the blocking object is a partially transparent blocking object, and if the reflected light intensity of the blocking object is greater than the second preset light intensity, the blocking object is an opaque blocking object; wherein, the second preset light intensity is greater than the first Preset light intensity. If the blocking object is a partially transparent blocking object, step S305 is performed. It should be noted that, if the shielding object is an opaque shielding object, step S305 may also be performed, but the effect of using heating to clean the shielding object is not as good as when the shielding object is a partially transparent shielding object.

In the embodiment of the present invention, when the temperature of the lidar window is lower than the first preset temperature, the window is automatically heated to increase the temperature of the window, and the obstructions are automatically heated and cleaned by heating. This is removed by heating. The method of covering the windows is especially suitable for partially transparent coverings such as rain and snow.

Embodiment 4

FIG. 4 is a flowchart of a method for detecting dirt on a lidar window in Embodiment 4 of the present invention. This embodiment is based on the above-mentioned embodiment, and further optimization is carried out. Specifically, how to determine the size of the obstruction according to the size of the obstruction is given. As well as briefings on cleaning specific locations of the windows. As shown in Figure 4, the operation process includes the following steps:

S401. Acquire point cloud data scanned by the lidar; the point cloud data includes reflected light intensity.

S402. Identify and determine each obstacle based on the point cloud data.

S403: Determine the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders.

S404. If the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, determine that there is a blocking object on the window of the lidar.

S405. Determine the size of the occluder and the position of the occluder according to the data points of the occluder.

Exemplarily, the size of the occluder may be determined by the number of data points corresponding to the occluder. The greater the number of data points corresponding to the occluder, the larger the occluder; the smaller the number of data points corresponding to the occluder, the larger the occluder. smaller. The size of the occluder refers to the area of the occluder. The position of the occluder in the viewport may be determined by the outer edge of the shape formed by the corresponding data point of the occluder in the point cloud, and/or may be determined by the geometric center of the shape formed by the corresponding data point of the occluder in the point cloud.

S406 , when the area of the obstruction is greater than the preset value, generate a cleaning instruction to control the cleaning mechanism to clean the window.

In this step, when the area of the obstruction is larger than the preset value, a cleaning instruction may be generated to control the cleaning mechanism to clean the window.

In another embodiment, step S406 may also be, when the blocking object is located in the working area of the window, a cleaning instruction may be generated to control the cleaning mechanism to clean the window. Among them, the working area of the window is the area where the laser emitted by the lidar passes, and the area where the lidar receives laser echoes. When the obstruction is located in the working area, it will interfere with the transmission and reception of the laser light, so a cleaning mechanism is required to remove it.

In another embodiment, step S406 may also be, when the area of the blocking object is larger than the preset value and when the blocking object is located in the working area of the window, a cleaning instruction may be generated to control the cleaning mechanism to clean the window. That is, in order to prevent the cleaning mechanism from being frequently activated, the cleaning mechanism is triggered to perform cleaning only when the shield is located in the working area and meets a certain size.

In yet another embodiment, step S406 may also be to acquire a cleaning standard corresponding to the location according to the location, and generate a cleaning instruction when the size of the obstruction exceeds the cleaning standard. In this embodiment, the windows are artificially divided, and different cleaning standards are configured for different locations of the lidar window. For example, the central area of the lidar window has a lower cleaning standard, and the lidar window except the central area has a lower cleaning standard Areas have high cleaning standards. In this case, the cleaning standard refers to the standard that triggers the cleaning mechanism to clean, and the standard can be a size standard or a standard such as reflection intensity. Therefore, in this case, the judgment corresponding to a specific area can be made according to the cleaning standards of different areas, and when the size of the obstruction located in the area exceeds the cleaning standard of the area, a cleaning instruction is generated to clean the area, Or clean the entire lidar window.

Exemplarily, the cleaning of the lidar window is applicable to any type of shields, both for partially transparent shields and for opaque shields. For example, the cleaning of the lidar window can be performed by using a liquid, and the liquid can include volatile organic substances. An optional liquid for cleaning the window can be glass water, and the glass water includes water, alcohol, glycol, corrosion inhibitor and various surfactants. In its embodiment, the cleaning agent can also be a high-pressure gas-liquid mixture, or cauterize the dirt with the aid of a laser.

In this embodiment of the present invention, when the area of the obstruction is larger than the preset value, a cleaning instruction may be generated to control the cleaning mechanism to clean the window; and/or, when the obstruction is located in the working area of the window, a cleaning instruction may be generated to control The cleaning mechanism cleans the window. In addition, according to the cleaning standards of different areas, when the size of the obstructions located in the area exceeds the cleaning standard, a cleaning instruction can be generated.

Embodiment 5

FIG. 5 is a schematic structural diagram of an apparatus for detecting dirt on a lidar window according to Embodiment 5 of the present invention. The apparatus can execute the method for detecting dirt on a lidar window provided by any embodiment of the present invention, and has the corresponding method for executing the method. Functional modules and beneficial effects. As shown in Figure 5, the device specifically includes:

The point cloud data acquisition module 501 is used for acquiring point cloud data obtained by laser radar scanning; the point cloud data includes reflected light intensity.

The obstacle identification module 502 is used for identifying the point cloud data to determine each obstacle.

The suspicious occluder determination module 503 is configured to determine, among the obstacles, obstacles located within a preset distance range of the window position of the lidar as suspicious occluders.

The blocking object determination module 504 is configured to determine that there is a blocking object on the window of the lidar if the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity.

Further, the device further includes: a blocker type judgment module, which is used to determine that the blocker is a partially transparent blocker if the reflected light intensity of the blocker is greater than the first preset light intensity and less than or equal to the second preset light intensity, If the reflected light intensity of the shielding object is greater than the second preset light intensity, the shielding object is an opaque shielding object; wherein the second preset light intensity is greater than the first preset light intensity.

Further, the device further includes: a cleaning control module, configured to generate a cleaning instruction to control the cleaning mechanism to clean the window if the ratio of the data points of the obstruction in the point cloud data is greater than a preset ratio.

Further, the device further includes: a heating control module for acquiring the temperature of the window of the lidar;

heating the window of the lidar when the temperature of the window of the lidar is less than the first preset temperature value;

When the temperature of the window of the lidar is greater than the second preset temperature value, the heating of the window of the lidar is stopped; the second preset temperature value is greater than the first preset temperature value.

Further, the suspicious occluder determination module 503 is also used to determine obstacles that are located within the preset distance range of the lidar window position and have similar position distributions in at least two frames of adjacent point cloud data as Suspicious obstruction.

Further, the suspicious occluder determination module 503 is also used to determine that each obstacle is located within the preset distance range of the window position of the lidar and the distance deviation in at least two consecutive frames of point cloud data is less than the first distance, then Obstacles Identify suspicious obstructions.

Further, the cleaning control module is also used to determine the size of the occluder and the position on the window according to the data points of the occluder;

When the area of the obstruction is larger than the preset value and/or the obstruction is located in the working area of the window, a cleaning instruction is generated to control the cleaning mechanism to clean the window; or

The cleaning standard corresponding to the location is obtained according to the location, and when the size of the obstruction exceeds the cleaning standard, a cleaning instruction is generated.

Embodiment 6

6A is a schematic structural diagram of a laser radar cleaning system according to Embodiment 6 of the present invention, and FIG. 6B is a structural schematic diagram of a cleaning control device in a laser radar cleaning system according to Embodiment 6 of the present invention. The cleaning system shown in FIG. 6A includes a laser radar cleaning mechanism 61 and a cleaning control device 60 . The lidar cleaning mechanism 61 is used for cleaning the window of the lidar. The cleaning control device 60 is connected to the laser radar cleaning mechanism 61 . The cleaning control device 60 includes a processor 601 and a memory 602 . A computer program is stored in the memory 602, so that when the processor 601 executes the computer program, the detection method of the lidar window contamination in the above-mentioned embodiment is implemented. Figure 6B shows a block diagram of an exemplary processing device 60 suitable for use in implementing embodiments of the present invention. The processing device 60 shown in FIG. 6B is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present invention. As shown in Figure 6B, the processing device 60 takes the form of a general purpose computing device. Components of the processing device 60 may include, but are not limited to, one or more processors 601 , a system memory 602 , and a bus 603 connecting different system components (including the system memory 602 and the processor 601 ).

Bus 603 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of a variety of bus structures. By way of example, these architectures include, but are not limited to, Industry Standard Architecture (ISA) bus, Micro Channel Architecture (MAC) bus, Enhanced ISA bus, Video Electronics Standards Association (VESA) local bus, and Peripheral Component Interconnect ( PCI) bus.

Processing device 60 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by processing device 60, including both volatile and non-volatile media, removable and non-removable media.

System memory 602 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 604 and/or cache memory 605 . Processing device 60 may further include other removable/non-removable, volatile/non-volatile computer system storage media. For example only, storage system 606 may be used to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 6B, commonly referred to as a "hard drive"). Although not shown in Figure 6B, a disk drive may be provided for reading and writing to removable non-volatile magnetic disks (eg "floppy disks"), as well as removable non-volatile optical disks (eg CD-ROM, DVD-ROM) or other optical media) to read and write optical drives. In these cases, each drive may be connected to bus 603 through one or more data media interfaces. System memory 602 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present invention.

A program/utility 608 having a set (at least one) of program modules 607, which may be stored, for example, in system memory 602, such program modules 607 including, but not limited to, an operating system, one or more application programs, other program modules, and programs Data, each or some combination of these examples may include an implementation of a network environment. Program modules 607 generally perform the functions and/or methods in the described embodiments of the present invention.

The processing device 60 may also communicate with one or more external devices 609 (eg, keyboard, pointing device, display 610, etc.), may also communicate with one or more devices that enable a user to interact with the device, and/or communicate with the device that enables the user to interact with the device. Processing device 60 can communicate with any device (eg, network card, modem, etc.) that communicates with one or more other computing devices. Such communication may take place through input/output (I/O) interface 611 . Also, processing device 60 may communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network such as the Internet) through a network adapter 612 . As shown in FIG. 6B , network adapter 612 communicates with other modules of processing device 60 via bus 603 . It should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with processing device 60, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives and data backup storage systems.

The processor 601 executes various functional applications and data processing by running programs stored in the system memory 602 , for example, implementing the lidar-based map construction method provided by the embodiments of the present invention for each lidar.

Embodiment 7

Embodiment 7 of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method for detecting dirt on a lidar window described in the foregoing embodiments can be implemented.

The computer storage medium in the embodiments of the present invention may adopt any combination of one or more computer-readable mediums. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or any combination of the above. More specific examples (a non-exhaustive list) of computer readable storage media include: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

A computer-readable signal medium may include a propagated data signal in baseband or as part of a carrier wave, with computer-readable program code embodied thereon. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device .

Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations of the present invention may be written in one or more programming languages, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional Procedural programming language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or wide area network (WAN), or may be connected to an external computer (eg, through the Internet using an Internet service provider) connect).

Note that the above are only preferred embodiments of the present invention and applied technical principles. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, combinations and substitutions can be made by those skilled in the art without departing from the protection scope of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and can also include more other equivalent embodiments without departing from the concept of the present invention. The scope is determined by the scope of the appended claims.

Claims (10)

1. a detection method of lidar window dirt, is characterized in that, comprises: acquiring point cloud data obtained by scanning the lidar; the point cloud data includes reflected light intensity; Identifying the point cloud data to determine each obstacle; Determining the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders; If the reflected light intensity of at least one of the suspicious blocking objects is greater than the first preset light intensity, it is determined that there is a blocking object on the window of the lidar. 2. The method of claim 1, further comprising: If the reflected light intensity of the shield is greater than the first preset light intensity and less than or equal to the second preset light intensity, the shield is a partially transparent shield, and if the reflected light intensity of the shield is greater than the With the second preset light intensity, the shielding object is an opaque shielding object; wherein, the second preset light intensity is greater than the first preset light intensity. 3. The method according to claim 1 or 2, characterized in that, further comprising: If the ratio of the data points of the occluder in the point cloud data is greater than a preset ratio, a cleaning instruction is generated to control the cleaning mechanism to clean the window. 4. The method of claim 1, further comprising: obtain the temperature of the lidar's window; heating the window of the lidar when the temperature of the window of the lidar is less than a first preset temperature value; When the temperature of the window of the lidar is greater than a second preset temperature value, the heating of the window of the lidar is stopped; the second preset temperature value is greater than the first preset temperature value. 5 . The method according to claim 1 , wherein the method further comprises: acquiring multi-frame point cloud data collected by the lidar by scanning multiple times; 5 . Determining the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious obstructions includes: determining the obstacles located within the preset distance range of the window position of the lidar among the obstacles Obstacles with similar location distribution within at least two frames of adjacent point cloud data are determined as suspicious occluders. 6 . The method according to claim 1 , wherein the determining of obstacles located within a preset distance range of the window position of the lidar as suspicious occluders in each obstacle comprises: 6 . If each obstacle is located within a preset distance range of the window position of the lidar and the distance deviation in at least two consecutive frames of the point cloud data is less than the first distance, the obstacle is determined as a suspicious occluder. 7. The method of claim 1, further comprising: Determine the size of the occluder and the position on the window according to the data points of the occluder; When the area of the shield is larger than a preset value and/or the shield is located in the working area of the window, a cleaning instruction is generated to control the cleaning mechanism to clean the window; or The cleaning standard corresponding to the position is acquired according to the position, and when the size of the obstruction exceeds the cleaning standard, a cleaning instruction is generated. 8. A device for detecting dirt on a lidar window, comprising: a point cloud data acquisition module for acquiring point cloud data obtained by scanning the lidar; the point cloud data includes reflected light intensity; an obstacle identification module, used for identifying the point cloud data to determine each obstacle; a suspicious occluder determination module, configured to determine the obstacles located within the preset distance range of the window position of the lidar among the obstacles as suspicious occluders; A blocker determination module, configured to determine that there is a blocker on the window of the lidar if the reflected light intensity of at least one of the suspicious blockers is greater than the first preset light intensity. 9. A laser radar cleaning system, comprising: A lidar cleaning mechanism for cleaning the lidar window; and A cleaning control device, connected to the lidar cleaning mechanism, the cleaning control device includes a processor and a memory; a computer program is stored in the memory, so that when the processor executes the computer program, the process as claimed in claim 1 is realized The method of any one of ~7. 10 . A computer storage medium on which a computer program is stored, characterized in that, when the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images