Vehicle-mounted laser radar cleaning method and system

Abstract

The invention relates to the technical field of vehicle intelligent control, and provides a vehicle-mounted laser radar cleaning method and system. The vehicle-mounted laser radar cleaning method comprises the steps that historical point cloud data of a laser radar detection mirror surface are acquired and preprocessed, and the type of a shielding object is marked; training a down-sampling model based on the preprocessed historical point cloud data to obtain ordered and regular point cloud data; training a deep learning model based on the ordered and regular point cloud data and the corresponding marker type; predicting the type of a shielding object corresponding to the point cloud data of the laser radar detection mirror surface by adopting a trained down-sampling model and a deep learning model; the type comprises a first type and a second type; if the type is the first type, the laser radar detection mirror surface is controlled to be heated, and drip washing and scraping washing are controlled to be carried out in sequence after heating; and if the type is the second type, the laser radar detection mirror surface is firstly controlled to be washed, and then is controlled to be scraped and washed.

Description

Technical Field

[0001] The present invention relates to the field of vehicle intelligent control technology, and in particular to a vehicle-mounted laser radar cleaning method and system. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the continuous development of LiDAR performance and environmental perception algorithms, its application fields have become more mature, and LiDAR has been widely used in industry, agriculture, manufacturing, etc. Especially in the field of autonomous driving, LiDAR has become an indispensable sensor for cars to achieve unmanned driving due to its unique advantages.

[0004] LiDAR mainly uses the built-in laser to emit laser beams through the surface protection cover to the surroundings and receive the reflected signals to generate point cloud data to obtain information. In an open-air working environment, the surface protection cover is inevitably blocked by dust, snow, frost, etc., which blocks the laser beam trajectory, resulting in data loss, affecting system operation, and may cause relatively large safety hazards.

[0005] In the prior art, the automatic cleaning function of the laser radar cannot identify ice / frost or dirt, cannot provide corresponding cleaning levels for different situations, and has low intelligence. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a vehicle-mounted laser radar cleaning method and system. The present invention can clean the laser radar mirror without affecting the operation of the laser radar, and has the advantages of strong real-time performance, high degree of automation, energy saving, high degree of cleaning, and strong system reliability.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A first aspect of the present invention provides a method for cleaning a vehicle-mounted laser radar.

[0009] A vehicle-mounted laser radar cleaning method, comprising:

[0010] Obtain historical point cloud data of the LiDAR detection mirror, perform preprocessing, and mark the type of occlusion;

[0011] Based on the preprocessed historical point cloud data, the downsampling model is trained to obtain orderly and regular point cloud data; based on the orderly and regular point cloud data and the corresponding marker types, the deep learning model is trained;

[0012] Using the trained downsampling model and deep learning model, predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type;

[0013] If it is the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then controlled to be scraped.

[0014] Furthermore, after scraping and cleaning, the point cloud data of the laser radar detection mirror is obtained and classified using the trained deep learning model. If no liquid is detected, the laser radar detection mirror is controlled to be heated until it is completely evaporated. If the point cloud data of the laser radar detection mirror detects that there is no obstruction, the operation is stopped.

[0015] Furthermore, the first type includes: the obstruction is frost, ice and / or snow; the second type includes: the obstruction is bird excrement or small insect corpses.

[0016] Furthermore, the deep learning model adopts a PointNet model or a GNN model.

[0017] Furthermore, the downsampling model adopts a PointSD model.

[0018] Furthermore, the preprocessing includes: removing noise and standardizing the point cloud data of the laser radar detection mirror.

[0019] A second aspect of the present invention provides a method for cleaning a vehicle-mounted laser radar.

[0020] A vehicle-mounted laser radar cleaning system, comprising:

[0021] A data acquisition and preprocessing module is configured to: acquire historical point cloud data of the laser radar detection mirror, perform preprocessing, and mark the type of the occlusion;

[0022] A model training module is configured to: train a downsampling model based on the preprocessed historical point cloud data to obtain orderly and regular point cloud data; and train a deep learning model based on the orderly and regular point cloud data and the corresponding marker type;

[0023] A prediction module is configured to: use a trained downsampling model and a deep learning model to predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type;

[0024] The control module is configured as follows: if it is of the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is of the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then scraped.

[0025] A third aspect of the present invention provides a computer-readable storage medium.

[0026] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the vehicle-mounted laser radar cleaning method as described in the first aspect above.

[0027] A fourth aspect of the present invention provides a computer device.

[0028] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the vehicle-mounted laser radar cleaning method as described in the first aspect above are implemented.

[0029] A fifth aspect of the present invention provides a computer program product or a computer program.

[0030] The present invention provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the vehicle-mounted laser radar cleaning method as described in the first aspect above.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] The present invention identifies obstructions to the laser radar detection mirror and adopts different cleaning processes for different types, thereby improving the cleanliness of the laser radar detection mirror and improving intelligence.

[0033] The present invention controls the laser radar detection mirror to be heated after cleaning, thereby avoiding liquid remaining after cleaning, ensuring the cleanliness of the laser radar detection mirror, improving the clarity of laser radar collection data, and reducing noise interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0035] Figure 1 is a flow chart of a vehicle-mounted laser radar cleaning method shown in the present invention;

[0036] Figure 2 is a structural diagram of a vehicle-mounted laser radar cleaning system shown in the present invention;

[0037] Figure 3 is a structural diagram of a vehicle-mounted laser radar cleaning device shown in the present invention;

[0038] Figure 4 It is a structural diagram of the cleaning device layer and the fan-shaped diverter shown in the present invention;

[0039] Among them, 1. detection device, 2. controller, 3. heating device, 4. liquid pump switch, 5. wiper device, 6. detection mirror, 21. wiper, 22. wiper groove, 23. fan-shaped diverter, 24. detection mirror, 25. cleaning device layer, 26. wiper stop working area, 31. arc-shaped spray hole, 32. infusion tube, 33. diverter drainage trough. DETAILED DESCRIPTION

[0040] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0042] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0043] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the methods and systems according to various embodiments of the present disclosure. It should be noted that each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of a code may include one or more executable instructions for implementing the logical functions specified in each embodiment. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, or they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the flowchart and / or block diagram, and the combination of boxes in the flowchart and / or block diagram can be implemented using a dedicated hardware-based system that performs a specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0044] Embodiment 1

[0045] like Figure 1 As shown, this embodiment provides a vehicle-mounted laser radar cleaning method. This embodiment uses the method applied to a server as an example for illustration. It can be understood that the method can also be applied to a terminal, and can also be applied to a terminal, a server, and a system, and is implemented through the interaction between the terminal and the server. The server can be an independent physical server, or a server cluster or a distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network servers, cloud communications, middleware services, domain name services, security services CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in this application. In this embodiment, the method includes the following steps:

[0046] Obtain historical point cloud data of the LiDAR detection mirror, perform preprocessing, and mark the type of occlusion;

[0047] Based on the preprocessed historical point cloud data, the downsampling model is trained to obtain orderly and regular point cloud data; based on the orderly and regular point cloud data and the corresponding marker types, the deep learning model is trained;

[0048] Using the trained downsampling model and deep learning model, predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type;

[0049] If it is the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then controlled to be scraped.

[0050] The present embodiment is described in detail below:

[0051] Step 1: Obtain historical point cloud data of the LiDAR detection mirror and clean, filter and smooth the data to remove noise and redundant data, and standardize and smooth the point cloud data for subsequent processing.

[0052] Step 2: Mark the pre-processed historical point cloud data of the lidar detection mirror surface, and mark the type of the occlusion, including a first type of frost, ice and / or snow, and a second type of bird excrement or small insect carcasses;

[0053] Step 3: Use the preprocessed historical point cloud dataset of the lidar detection mirror and the corresponding label type to train the downsampling model to obtain an orderly and regular point cloud dataset; based on the orderly and regular point cloud dataset and the corresponding label type, construct the training set and test set.

[0054] The downsampling model may adopt the PointSD model, and the PointSD model adopts the first loss function to optimize the learning parameters of the PointSD model during the training process;

[0055] Among them, the first loss function is expressed by the following formula:

[0056] L1=αL DKD +βL CE

[0057] Among them, α and β represent the hyperparameters of DKD loss and cross entropy loss respectively, and L DKD represents the DKD loss, L CE represents the cross entropy loss.

[0058]

[0059] Among them, γ and λ are introduced hyperparameters, which are used as the weights of the loss values ​​of the target category and the non-target category respectively; p t and p′ t Respectively represent the probability of two samples being predicted as the target class; and Respectively represent the probability of two samples being predicted as non-target classes; and They represent the probability distribution results recalculated proportionally after the target class is not included in the two samples; KL(·) represents the KL divergence.

[0060]

[0061] Among them, P(x) and Q(x) represent two probability distributions on the random variable x.

[0062]

[0063] Among them, P and P′ represent the probability distribution results of the two samples being predicted as the target class, n is the number of samples in the batch, and y i Represents the smoothed probability distribution of the label corresponding to each sample.

[0064] The smoothing operation is expressed by the following formula:

[0065]

[0066] Among them, ε represents the smoothing parameter. Taking a smaller constant can avoid overfitting to a certain extent and alleviate the impact of incorrect labels.

[0067] Step 4: Take the ordered and regular point cloud dataset in the training set as input and the corresponding label type as output to train the deep learning model;

[0068] Among them, the deep learning model can adopt a PointNet network, including a multi-layer perceptron (MLP) and a maximum pooling layer, and optimize the hyperparameters of the PointNet network through training; the deep learning model can also adopt a GNN model. In addition, the deep learning model can also be implemented using other models, which will not be repeated in the present invention.

[0069] Step 5: Obtain the point cloud data of the laser radar detection mirror and perform preprocessing; input the preprocessed point cloud data of the laser radar detection mirror into the trained downsampling model for downsampling processing to obtain regular point cloud data; input the regular point cloud data into the trained deep learning model for classification to obtain the type of occlusion.

[0070] Step 6: If it is the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then controlled to be scraped.

[0071] After scraping and cleaning, the point cloud data of the LiDAR detection mirror is obtained and classified using the trained deep learning model. If no liquid is detected, the LiDAR detection mirror is heated until it is completely evaporated. If the point cloud data of the LiDAR detection mirror shows that there is no obstruction, the system stops working.

[0072] The present invention identifies obstructions to the laser radar detection mirror and adopts different cleaning processes for different types, thereby improving the cleanliness of the laser radar detection mirror and improving intelligence.

[0073] Embodiment 2

[0074] This embodiment provides a vehicle-mounted laser radar cleaning system.

[0075] A vehicle-mounted laser radar cleaning system, comprising:

[0076] A data acquisition and preprocessing module is configured to: acquire historical point cloud data of the laser radar detection mirror, perform preprocessing, and mark the type of the occlusion;

[0077] A model training module is configured to: train a downsampling model based on the preprocessed historical point cloud data to obtain orderly and regular point cloud data; and train a deep learning model based on the orderly and regular point cloud data and the corresponding marker type;

[0078] A prediction module is configured to: use a trained downsampling model and a deep learning model to predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type;

[0079] The control module is configured as follows: if it is of the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is of the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then scraped.

[0080] In some embodiments, after scraping and cleaning, the point cloud data of the laser radar detection mirror is obtained and classified using a trained deep learning model. If no liquid is detected, the laser radar detection mirror is controlled to be heated until it is completely evaporated. If the point cloud data of the laser radar detection mirror detects that there is no obstruction, the operation is stopped.

[0081] In some embodiments, the first type includes: the obstruction is frost, ice and / or snow; the second type includes: the obstruction is bird excrement or small insect carcasses.

[0082] In some embodiments, the deep learning model adopts a PointNet model or a GNN model.

[0083] In some embodiments, the downsampling model adopts a PointSD model.

[0084] In some embodiments, the preprocessing includes: removing noise and standardizing the point cloud data of the laser radar detection mirror.

[0085] In this embodiment, if Figure 2 , Figure 3 , Figure 4 As shown, the laser radar cleaning system has the functions of scraping, washing and heating, and the hardware structure may include: a detection device 1, a controller 2, a heating device 3, a liquid pump switch 4, a wiper device 5, a detection mirror 6, a wiper 21, a wiper groove 22, a fan-shaped diverter 23, a cleaning device layer 25, a wiper stop working area 26, an arc-shaped spray hole 31, an infusion tube 32 and a diverter drainage groove 33.

[0086] The detection device 1 can process the detected point cloud information, identify what kind of obstruction it is, and convey this information to the laser radar cleaning controller 2; the controller 2 processes the signal sent by the detection device 1 and issues instructions to the cleaning device. If it is ice and snow, the heating device 3 is started first and then the wiper device 5 is turned on (the wiper will return to the stop area 26 after the work is completed to avoid blocking the laser radar detection mirror and affecting the laser radar work). If it is other dirt, first turn on the liquid pump switch 4 to spray the laser radar detection mirror 6 with liquid and then activate the wiper device 5 and the heating device 3. After the above steps are completed, the detection device will also self-check once to detect whether the obstruction is completely cleaned. If it is completely cleaned, the cleaning device will no longer work. If it is not completely cleaned, the cleaning system will continue to work.

[0087] The cleaning device layer 25 is low in front and high in the back, which is convenient for the cleaning agent to flow out; the wiper groove 22 on the laser radar body; the leftmost side of the wiper groove 22 is the resting area when the wiper 21 is not working. Specifically, the cleaning device layer 25 contains a liquid infusion tube 32, a diverter drainage groove 33 and an arc-shaped liquid dripping hole 31.

[0088] The laser radar detection mirror is an outwardly convex arc mirror, the drip hole is arc-shaped and parallel to the mirror direction, and the number of fan-shaped diverters 23 depends on the size of the laser radar detection mirror.

[0089] The arc-shaped shower hole has an arc length of 4-6 mm and a height of 2-3 mm. A drainage groove is provided in the fan-shaped diverter 23 to facilitate uniform distribution of the cleaning liquid on the laser radar detection mirror.

[0090] The wiper device is controlled by a motor and fits the laser radar detection mirror during cleaning.

[0091] When the laser radar cleaning system executes the work instruction, the detection device starts to run, emits a laser beam, receives the reflected point cloud information for processing, and determines the type of obstruction: ① If it is water droplets, ice and snow, the controller sends a heating instruction to the heating device, and then brushes until the detection device detects that the laser radar mirror is unobstructed, and the laser radar cleaning system stops working. ② If it is detected as bird excrement or a small insect corpse, the controller sends a command to the water pump switch to rinse the laser radar detection mirror, and then starts the wiper device to scrape the laser radar detection mirror, and then the detection device detects it. When it detects that only a small amount of liquid is left, the controller sends a heating instruction to the heating device to evaporate the liquid on the laser radar detection mirror until it evaporates completely. The detection device then detects that there is no obstruction on the mirror and the system stops working.

[0092] The hardware structure of the above-mentioned laser radar cleaning system is only one implementation method of the present invention. The present invention can also be implemented by using an existing laser radar cleaning system, which will not be described in detail here.

[0093] The present invention can effectively avoid the difficulty of cleaning the vehicle-mounted laser radar and the difficulty of cleaning the laser radar in ice and snow weather. The cleaning device has strong practicality, simple structure, high cleaning degree, small size, strong system reliability and simple maintenance.

[0094] The present invention can improve the redundancy of the overall system, reduce the false alarm rate, and enhance the overall driving experience by adding a heartbeat detection mechanism and integrating it with the current camera detection results.

[0095] Embodiment 3

[0096] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the vehicle-mounted laser radar cleaning method described in the above-mentioned embodiment 1 are implemented.

[0097] Embodiment 4

[0098] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the vehicle-mounted laser radar cleaning method described in the above-mentioned embodiment 1 are implemented.

[0099] Embodiment 5

[0100] This embodiment provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steps in the vehicle-mounted laser radar cleaning method described in the first embodiment.

[0101] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0102] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0105] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A vehicle-mounted laser radar cleaning method, characterized in that: include: Obtain historical point cloud data of the LiDAR detection mirror, perform preprocessing, and mark the type of occlusion; Based on the preprocessed historical point cloud data, the downsampling model is trained to obtain orderly and regular point cloud data; Train deep learning models based on ordered and regular point cloud data and corresponding marker types; Using the trained downsampling model and deep learning model, predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type; If it is the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then controlled to be scraped.

2. The vehicle-mounted laser radar cleaning method according to claim 1, characterized in that: After scraping and cleaning, the point cloud data of the LiDAR detection mirror is obtained and classified using the trained deep learning model. If no liquid is detected, the LiDAR detection mirror is heated until it is completely evaporated. If the point cloud data of the LiDAR detection mirror shows that there is no obstruction, the system stops working.

3. The vehicle-mounted laser radar cleaning method according to claim 1, characterized in that: The first type includes: the obstruction is frost, ice and / or snow; the second type includes: the obstruction is bird excrement or small insect carcasses.

4. The vehicle-mounted laser radar cleaning method according to claim 1, characterized in that: The deep learning model adopts a PointNet model or a GNN model.

5. The vehicle-mounted laser radar cleaning method according to claim 1, characterized in that: The downsampling model adopts the PointSD model.

6. The vehicle-mounted laser radar cleaning method according to claim 1, characterized in that: The preprocessing includes: removing noise and standardizing the point cloud data of the laser radar detection mirror.

7. A vehicle-mounted laser radar cleaning system, characterized in that: include: A data acquisition and preprocessing module is configured to: acquire historical point cloud data of the laser radar detection mirror, perform preprocessing, and mark the type of the occlusion; A model training module is configured to: train a downsampling model based on the preprocessed historical point cloud data to obtain orderly and regular point cloud data; and train a deep learning model based on the orderly and regular point cloud data and the corresponding marker type; A prediction module is configured to: use a trained downsampling model and a deep learning model to predict the type of the occluder corresponding to the point cloud data of the laser radar detection mirror; the type includes a first type and a second type; The control module is configured as follows: if it is of the first type, the laser radar detection mirror surface is controlled to be heated, and after heating, it is controlled to be rinsed and scraped in sequence; if it is of the second type, the laser radar detection mirror surface is controlled to be rinsed first, and then scraped.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in the vehicle-mounted laser radar cleaning method as described in any one of claims 1 to 6 are implemented.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the vehicle-mounted laser radar cleaning method as described in any one of claims 1-6 are implemented.

10. A computer program product, characterized in that The computer program product includes a computer program, and when the computer program is executed by a processor, the steps in the vehicle-mounted laser radar cleaning method as described in any one of claims 1 to 6 are implemented.

Images