Municipal environmental sanitation sweeper operation control system and sweeping method

Abstract

The invention relates to the technical field of control and adjustment, in particular to a municipal environmental sanitation sweeper operation control system and a sweeping method.The municipal environmental sanitation sweeper operation control system comprises a sweeping device and an image obtaining module, the pollution condition identification module is used for identifying pollution types based on the environment images; the coordinate extraction module is used for acquiring coordinate information of a pollution type region in the environment image; the cleaning module is used for automatically switching a matched cleaning device for cleaning according to the coordinate information and the corresponding pollution type; and the parameter adjusting module adjusts the working parameters of each cleaning device in real time according to the residual pollution amount of the corresponding coordinate information after cleaning. Therefore, the related cleaning devices are automatically switched to clean the corresponding areas according to the received data, so that the labor burden is reduced, and the cleaning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of control and regulation, and particularly to an operation control system and a cleaning method for a municipal sanitation sweeper. Background Art

[0002] As an important tool for urban cleaning, the municipal sanitation sweeper undertakes the task of keeping the streets clean and tidy. It is equipped with powerful dust suction devices, sweeping brushes, water spraying systems and other equipment. During operation, it can effectively remove garbage, dust and stains on the road surface, reduce dust pollution at the same time, and ensure the sanitation and beauty of the urban environment. Its flexible operability and high cleaning ability not only improve the efficiency of sanitation work, but also greatly improve the quality of life of citizens. It is an indispensable part of modern urban management.

[0003] Existing municipal sanitation automatic sweepers usually adopt a combination of multiple cleaning methods to improve cleaning efficiency and effect, mainly including technologies such as dust suction, sweeping brush cleaning and water spraying for dust reduction. By manually identifying corresponding scenarios and then turning on corresponding function modules to achieve corresponding functions, however, during the actual working process, humans may be negligent or forgetful. Summary of the Invention

[0004] The purpose of the present invention is to provide an operation control system and a cleaning method for a municipal sanitation sweeper, aiming to automatically switch relevant cleaning devices according to the received data to clean corresponding areas, thereby reducing the manual burden and improving the cleaning efficiency.

[0005] To achieve the above purpose, in the first aspect, the present invention provides an operation control system for a municipal sanitation sweeper, including a cleaning device. The cleaning device includes a powerful dust suction module, a sweeping brush module and a water spraying module. The powerful dust suction module is used for adsorbing dust; the sweeping brush module is used for cleaning the ground; the water spraying module is used for spraying water mist into the air to reduce dust. It also includes an image acquisition module, a pollution situation identification module, a coordinate extraction module, a cleaning module and a parameter adjustment module; the image acquisition module is used for acquiring the environmental image in the moving direction of the vehicle body;

[0006] The pollution situation identification module is used for identifying the pollution type based on the environmental image, and the pollution type includes airborne dust, ground garbage and ground ash layer;

[0007] The coordinate extraction module is used for acquiring the coordinate information of the pollution type area in the environmental image;

[0008] The cleaning module is used for automatically switching the matching cleaning device according to the coordinate information and the corresponding pollution type for cleaning;

[0009] The parameter adjustment module adjusts the working parameters of each cleaning device in real time according to the residual pollution amount of the corresponding coordinate information after cleaning.

[0010] Among them, the operation control system of the municipal sanitation cleaning vehicle further includes a voice prompt module, which is used to remind the staff when switching the cleaning device.

[0011] Among them, the image acquisition module includes a parameter setting unit, a collection unit and an image processing unit;

[0012] The parameter setting unit is used to adjust the working parameters of the camera;

[0013] The collection unit is used to start the camera to continuously collect environmental images;

[0014] The image processing unit is used to preprocess the collected environmental images.

[0015] Among them, the image processing unit includes a denoising subunit, a contrast enhancement subunit and a grayscale subunit;

[0016] The denoising subunit is used to smooth the image using a Gaussian filter to reduce random noise;

[0017] The contrast enhancement subunit is used to stretch the brightness range of the image to enhance the contrast;

[0018] The grayscale subunit is used to convert the color image into a grayscale image.

[0019] Among them, the pollution situation recognition module includes a sample acquisition unit, a data set generation unit, a training unit, and a testing unit;

[0020] The sample acquisition unit is used to acquire image samples of the cleaning area, and the samples contain labels marking pollution categories;

[0021] The data set generation unit is used to divide the image into small pieces and extract key feature points to form a data set, and divide the data set into a training set and a validation set;

[0022] The training unit is used to train the convolutional neural network model using the training set to obtain a classification model;

[0023] The testing unit uses the validation set to test the classification model.

[0024] Among them, the coordinate extraction module includes an edge detection unit, a calibration unit, a conversion unit and a display unit;

[0025] The edge detection unit is used to detect the edges of the pollution type area using the Canny algorithm;

[0026] The calibration unit is used to calculate the bounding box for the edge of each pollution type area;

[0027] The conversion unit is used to convert the pixel coordinates of the bounding box into actual physical coordinates;

[0028] The display unit is used to draw the bounding box and the center point on the environmental image and display them.

[0029] Wherein, the cleaning module includes a position acquisition unit, a cleaning path generation unit and a cleaning unit;

[0030] The position acquisition unit is used to acquire the physical coordinates of the identified pollution area;

[0031] The cleaning path generation unit is used to calculate the cleaning path according to the physical coordinates;

[0032] The cleaning unit is used to clean the pollution area based on the cleaning path and the corresponding cleaning device.

[0033] In a second aspect, the present invention also provides a method for cleaning a municipal sanitation sweeper, including: acquiring an environmental image in the moving direction of the vehicle body;

[0034] Identifying the pollution type based on the environmental image, and the pollution type includes airborne dust, ground garbage and ground dust;

[0035] Acquiring the coordinate information of the pollution type area in the environmental image;

[0036] Automatically switching to a matching cleaning device for cleaning based on the coordinate information and the pollution type;

[0037] Real-time adjusting the working parameters of each cleaning device according to the pollution amount corresponding to the cleaned coordinate information.

[0038] For an operation control system and a cleaning method of a municipal sanitation sweeper according to the present invention, the cleaning device is the core execution component of the whole system. The strong dust suction module is used to adsorb airborne dust and other suspended particulate matters, effectively reducing air pollution. The brush module is used to physically clean the ground, removing various garbage, dust and other sundries on the ground. The water spraying module can spray fine water droplets into the air or on the ground to form a water mist to achieve the effect of dust reduction, and at the same time can also help clean the ground.

[0039] The image acquisition module is installed on a camera or other visual sensor device of the vehicle, which captures the environmental conditions in the traveling direction of the sweeper in real time and transmits this image data to the subsequent processing unit. The pollution situation recognition module uses advanced image processing technologies and algorithms (such as deep learning) to analyze the video stream provided by the image acquisition module, automatically distinguishing different types of pollutants, such as differentiating between airborne dust and specific types of garbage on the ground. Once the pollution area is determined, the coordinate extraction module is responsible for accurately locating the position information of these areas, that is, the exact coordinates of the pollution source. This step is crucial for subsequent precise cleaning measures. Based on the information obtained from the previous steps (the type of pollutant and its location), the cleaning module will intelligently select the most suitable cleaning method for the current situation, whether it is to start the vacuum cleaner to deal with light dust, use a sweeper brush to remove larger objects, or even combine water spraying operations to enhance the effect. In addition, it can flexibly adjust the operating modes of each subsystem according to actual needs. After the initial cleaning is completed, the parameter adjustment module will continue to monitor the status of the relevant area. If residual pollution is found, it will correspondingly adjust the working intensity of each cleaning device or other parameter settings until a satisfactory cleanliness level is achieved. This can greatly improve work efficiency, reduce the manual labor burden, and more importantly, it can make the most optimized selection according to different environmental conditions to ensure that the urban environment is cleaner and tidier. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a structural diagram of an operation control system for a municipal sanitation sweeper of the present invention.

[0042] Figure 2 It is a structural diagram of the image acquisition module of the present invention.

[0043] Figure 3 It is a structural diagram of the image processing unit of the present invention.

[0044] Figure 4 It is a structural diagram of the pollution situation recognition module of the present invention.

[0045] Figure 5 It is a structural diagram of the coordinate extraction module of the present invention.

[0046] Figure 6 It is a structural diagram of the cleaning module of the present invention.

[0047] Figure 7 It is the structural diagram of the parameter adjustment module of the present invention.

[0048] Figure 8 It is the flowchart of a cleaning method for a municipal sanitation sweeper of the present invention.

[0049] Cleaning device 101, image acquisition module 102, pollution situation identification module 103, coordinate extraction module 104, cleaning module 105, parameter adjustment module 106, voice prompt module 107, parameter setting unit 108, acquisition unit 109, image processing unit 110, denoising subunit 111, contrast enhancement subunit 112, grayscale subunit 113, sample acquisition unit 114, dataset generation unit 115, training unit 116, testing unit 117, edge detection unit 118, calibration unit 119, conversion unit 120, display unit 121, position acquisition unit 122, cleaning path generation unit 123, cleaning unit 124, dust amount judgment unit 125, power matching unit 126, speed adjustment unit 127. Specific embodiments

[0050] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

[0051] Please refer to Figures 1 to 7 , the present invention provides a municipal sanitation sweeper operation control system, including a cleaning device 101. The cleaning device 101 includes a strong dust suction module, a sweeping brush module, and a water spraying module. The strong dust suction module is used for adsorbing dust; the sweeping brush module is used for cleaning the ground; the water spraying module is used for spraying water mist into the air to reduce dust. It also includes an image acquisition module 102, a pollution situation identification module 103, a coordinate extraction module 104, a cleaning module 105, and a parameter adjustment module 106. The image acquisition module 102 is used for acquiring the environmental image in the moving direction of the vehicle body; the pollution situation identification module 103 is used for identifying the pollution type based on the environmental image, and the pollution type includes airborne dust, ground garbage, and ground ash layer; the coordinate extraction module 104 is used for acquiring the coordinate information of the pollution type area in the environmental image; the cleaning module 105 is used for automatically switching and matching the corresponding cleaning device 101 according to the coordinate information and the pollution type for cleaning; the parameter adjustment module 106 is used for adjusting the working parameters of each cleaning device 101 in real time according to the residual pollution amount of the corresponding coordinate information after cleaning.

[0052] In this embodiment, the cleaning device 101 is the core execution component of the entire system. The powerful vacuuming module is used to adsorb airborne dust and other suspended particulate matters, effectively reducing air pollution. The bristle brush module is used for physically cleaning the ground to remove various types of garbage, dust, and other sundries on the ground. The water spraying module can spray fine water droplets into the air or onto the ground to form a water mist, so as to achieve the effect of dust suppression and also help clean the ground.

[0053] The image acquisition module 102 is a camera or other vision sensor device installed on the vehicle, which captures the environmental conditions in the traveling direction of the cleaning vehicle in real time and transmits this image data to the subsequent processing unit. The pollution situation recognition module 103 uses advanced image processing technologies and algorithms (such as deep learning) to analyze the video stream provided by the image acquisition module 102, automatically distinguishing different types of pollutants, such as differentiating between airborne dust and specific types of garbage on the ground. Once the pollution areas are determined, the coordinate extraction module 104 is responsible for accurately locating the position information of these areas, that is, the exact coordinates of the pollution sources. This step is crucial for subsequent precise application of cleaning measures. Based on the information obtained from the previous steps (the type and location of pollutants), the cleaning module 105 will intelligently select the most suitable cleaning method for the current situation, whether to start the vacuum cleaner to deal with light dust, or use the bristle brush to remove larger objects, or even combine water spraying operations to enhance the effect. In addition, it can also flexibly adjust the operating modes of each subsystem according to actual needs. After the initial cleaning is completed, the parameter adjustment module 106 will continue to monitor the status of the relevant areas. If residual pollution is found, it will correspondingly adjust the working intensity of each cleaning device 101 or other parameter settings until a satisfactory cleanliness level is achieved.

[0054] In summary, such a comprehensive cleaning vehicle operation control system can not only greatly improve work efficiency and reduce the manual labor burden, but more importantly, it can make the most optimized choices according to different environmental conditions to ensure that the urban environment is cleaner and tidier. With the progress and development of technology, such a system will also possess more advanced features in the future and contribute to building a beautiful homeland.

[0055] The municipal sanitation cleaning vehicle operation control system further includes a voice prompt module 107, which is used to remind the staff when switching the cleaning device 101.

[0056] The voice prompt module 107 is to enhance the interactivity between the operator and the system, improve work efficiency while ensuring the safety and accuracy of operations. Specifically, the functions and roles of the voice prompt module 107 are as follows:

[0057] When the system automatically switches different cleaning devices 101 according to the analysis result of the environmental image (for example, switching from the sweeping brush mode to the vacuuming mode or the water spraying mode), the voice prompt module 107 will immediately issue a clear voice prompt to inform the operator of the current operation change. This helps the operator timely understand the change of the vehicle state, so as to make corresponding preparations or adjustments.

[0058] In addition to the basic work mode switching notification, in some cases, such as encountering obstacles or special situations such as emergency stop, the voice prompt module 107 will also immediately issue a warning message to help the driver quickly take measures to avoid potential risks.

[0059] The image acquisition module 102 includes a parameter setting unit 108, a collection unit 109 and an image processing unit 110; the parameter setting unit 108 is used to adjust the working parameters of the camera; the collection unit 109 is used to start the camera to continuously collect environmental images; the image processing unit 110 is used to preprocess the collected environmental images.

[0060] The parameter setting unit 108 allows the user to adjust the working parameters of the camera according to actual needs, such as exposure time, aperture size, ISO sensitivity, etc. By reasonably setting these parameters, it is ensured that clear and stable image quality can be obtained under different lighting conditions, thereby improving the recognition accuracy of the entire system. Once the parameters are set, the collection unit 109 will start the camera to continuously capture the environmental scene in front of or around the vehicle. This step is very crucial because the continuous image stream provides a real-time data source for the system, enabling the cleaning vehicle to respond immediately to environmental changes. After the original image is captured, the image processing unit 110 will perform a series of preprocessing operations on it to optimize the image quality and prepare for the subsequent analysis.

[0061] The image processing unit 110 includes a denoising unit 111, a contrast enhancement unit 112 and a grayscale unit 113; the denoising unit 111 is used to smooth the image using a Gaussian filter to reduce random noise; the contrast enhancement unit 112 is used to stretch the brightness range of the image to enhance the contrast; the grayscale unit 113 is used to convert the color image into a grayscale image.

[0062] The denoising unit 111 uses techniques such as Gaussian filters to remove random noise in the image. Gaussian filtering is a commonly used smoothing method that can effectively reduce image noise caused by insufficient light or other factors, while keeping edges and other important features unaffected. The contrast enhancer unit 112 increases the contrast of the image by stretching the brightness range of the image. This process helps to highlight the detailed parts of the image. Especially under low-light conditions, enhancing the contrast makes the object contours more obvious, facilitating subsequent feature extraction and pattern recognition. The grayscale unit 113 converts the color image into a grayscale image. Although color images contain more color information, in many cases, such as edge detection, texture analysis and other tasks, grayscale images are sufficient and greatly reduce the computational complexity. In addition, grayscale conversion also helps to reduce the data volume in subsequent processing steps and speed up the processing speed.

[0063] In summary, through the coordinated work of its internal units, the image acquisition module 102 can not only efficiently capture high-quality environmental images, but also improve the image quality through a series of preprocessing means, providing a solid foundation for subsequent advanced functions such as pollution situation recognition.

[0064] The pollution situation recognition module 103 includes a sample acquisition unit 114, a data set generation unit 115, a training unit 116, and a testing unit 117; the sample acquisition unit 114 is used to acquire image samples of the cleaning area, and the samples contain labels marking pollution categories; the data set generation unit 115 is used to segment the image into small pieces and extract key feature points to form a data set, and divide the data set into a training set and a validation set; the training unit 116 is used to train a convolutional neural network model using the training set to obtain a classification model; the testing unit 117 uses the validation set to test the classification model.

[0065] The sample acquisition unit 114 collects a large number of image samples from the cleaning area and ensures that these samples cover various different pollution conditions. Each image sample should be accompanied by marking information clearly indicating the types of pollution present in the image (such as airborne dust, ground garbage, ground ash layer, etc.). These data with labels form the basis for model training.

[0066] The data set generation unit 115 performs preprocessing on the collected images, including operations such as cropping and scaling, to ensure that all input data has a consistent format. In order to improve the learning efficiency of the model and better capture the local features in the image, the original image is divided into smaller blocks or fragments, and then key feature points or feature vectors are extracted from each small block. These features are edges, color distribution, texture, etc., which are essential for distinguishing different types of pollutants. Finally, the data constructed according to the above steps is organized into a structured data set and further divided into a training set and a validation set. Typically, most of the data is used for training, while a small part is reserved for validation to evaluate the performance of the model on unseen data.

[0067] The training unit 116 trains a convolutional neural network (CNN) model using the data in the training set. CNN is a deep learning architecture that is particularly suitable for processing image data, and it can effectively learn the spatial hierarchical structure in the image. By repeatedly iteratively optimizing the model parameters, the model can accurately classify the image into the corresponding pollution category.

[0068] The testing unit 117 is used to test the trained classification model using an independent validation set after the training is completed. This step is very important because it can help developers understand the actual performance of the model and whether it can be generalized to new unknown data. The test results will provide a series of indicators, such as accuracy, recall, F1 score, etc., to measure the effectiveness and reliability of the model. If the test results show that the model performs poorly, it is necessary to return to the previous steps to adjust the parameter settings or improve the model structure until satisfactory performance is achieved.

[0069] Through such a complete process, the pollution situation identification module 103 can establish a powerful classification system, which can not only quickly and accurately identify various pollutants in the cleaning area, but also provide strong support for subsequent automated cleaning decisions, greatly improving the intelligence level of the cleaning work.

[0070] The coordinate extraction module 104 includes an edge detection unit 118, a calibration unit 119, a conversion unit 120 and a display unit 121; the edge detection unit 118 is used to detect the edge of the pollution type area using the Canny algorithm; the calibration unit 119 is used to calculate the bounding box for the edge of each pollution type area; the conversion unit 120 is used to convert the pixel coordinates of the bounding box into actual physical coordinates; the display unit 121 is used to draw the bounding box and the center point on the environment image and display them.

[0071] The edge detection unit 118 uses the Canny algorithm to identify the edges of the pollution type area in the image. The Canny algorithm is a classic multi-stage edge detection algorithm, which first uses a Gaussian filter to smooth the image to reduce noise, then calculates the gradient strength and direction, then applies non-maximum suppression to refine the edge, and finally determines the final edge through a double threshold method.

[0072] The Canny algorithm is widely used in computer vision due to its good signal-to-noise ratio and single response characteristics, especially in situations where clear boundaries are required.

[0073] The calibration unit 119 further processes the edges of each contamination type area and calculates the minimum rectangular bounding box containing these edges. This step helps to simplify the subsequent coordinate conversion process and also provides a more intuitive target area for the cleaning device 101. The bounding box provides important information about the size, shape and relative position of the contaminated area in the image.

[0074] The conversion unit 120 converts the pixel coordinates of the bounding box into physical coordinates in the real world where the vehicle is located. This process usually involves camera parameters (such as focal length, sensor size, etc.) and the relative position relationship between the vehicle and the ground. Through methods such as geometric transformation or perspective transformation, the system can establish a mapping relationship from the image plane to the real world, so that the position information of the bounding box directly corresponds to the actual cleaning area, which facilitates the cleaning device 101 to accurately navigate to the target location.

[0075] The display unit 121 draws a bounding box and its center point on the original environment image and presents it to the operator through the vehicle display or other interface. Such a visual display not only helps the operator to intuitively understand the current working status, but also serves as a debugging tool to help technicians optimize system performance.

[0076] The drawn content also includes auxiliary information such as pollution type annotations and predicted cleaning paths to enhance user experience and improve work efficiency.

[0077] In summary, the coordinate extraction module 104 achieves seamless connection from image analysis to actual operation by combining advanced image processing technology and space coordinate conversion technology.

[0078] The cleaning module 105 includes a position acquisition unit 122, a cleaning path generation unit 123 and a cleaning unit 124; the position acquisition unit 122 is used to obtain the physical coordinates of the identified contaminated area; the cleaning path generation unit 123 is used to calculate the cleaning path according to the physical coordinates; the cleaning unit 124 is used to clean the contaminated area based on the cleaning path and the corresponding cleaning device 101.

[0079] The location acquisition unit 122 receives the actual physical coordinates of the identified contaminated area from the coordinate extraction module 104. These coordinates are determined by the previous image processing step and have been converted from pixel coordinates to real-world coordinates of the vehicle. By cooperating with the vehicle positioning system (such as GPS or SLAM technology based on laser radar / camera), the location acquisition unit 122 accurately locates the specific location of each contaminated point on the map.

[0080] The cleaning path generation unit 123 uses the received physical coordinates to plan the most effective cleaning route. This usually involves the application of a path planning algorithm, such as an A* search algorithm, a Dijkstra algorithm, or a more complex optimization algorithm. Path planning not only needs to consider how to cover all contaminated areas, but also needs to comprehensively consider factors such as cleaning efficiency, energy consumption, and safety. For example, the path should try to avoid repeatedly cleaning the same area, and the turning radius and operating limitations of the cleaning vehicle should also be taken into account. The generated cleaning path includes a series of points, each of which corresponds to a specific cleaning action, such as starting a specific type of cleaning device 101, adjusting the direction of travel, etc.

[0081] The cleaning unit 124 controls the cleaning vehicle to move along a predetermined route according to the path information provided by the cleaning path generation unit 123, and activates the corresponding cleaning device 101 to clean the polluted area. Based on different types of pollution (such as air dust, ground garbage, and ground ash layer), the cleaning unit 124 will select the cleaning method that best suits the current situation. For example, a powerful dust suction module is used for light dust; a sweeping brush module is used for solid garbage; and a water spray module is used for situations where dust reduction is required. During the cleaning process, the cleaning unit 124 will also monitor the cleaning effect in real time, and adjust the cleaning parameters such as suction strength, water spray volume, etc. according to actual conditions to ensure the best cleaning effect.

[0082] The parameter adjustment module 106 includes a dust amount judgment unit 125, a power matching unit 126 and a speed adjustment unit 127; the dust amount judgment unit 125 is used to judge the dust amount based on the texture of the polluted area in the environmental image; the power matching unit 126 is used to match the power of the water spray module and the strong dust suction module based on the dust amount; the speed adjustment unit 127 is used to match the power of the corresponding sweeping brush module based on the garbage area.

[0083] The dust amount judgment unit 125 estimates the dust amount by analyzing the texture of the polluted area in the environmental image. For example, places with dense dust usually appear as granular structures or specific grayscale distribution patterns in the image. Using computer vision techniques, such as texture analysis algorithms (including but not limited to gray-level co-occurrence matrix, local binary pattern, etc.), the dust density of different areas is quantitatively evaluated. This image feature-based method can provide relatively accurate dust amount information, providing a basis for subsequent power matching.

[0084] The power matching unit 126 determines the power levels that the water spraying module and the powerful dust suction module should use according to the data provided by the dust amount judgment unit 125. For areas with more dust, it is necessary to increase the water volume of the water spraying module or increase the suction intensity of the powerful dust suction module to ensure effective dust removal. On the contrary, in the case of less dust, the power of these modules is appropriately reduced, thus saving energy and reducing unnecessary resource consumption. By precisely controlling the power output of each cleaning device 101, not only the cleaning efficiency is improved, but also the economy and environmental friendliness of the entire system are ensured.

[0085] The rotation speed adjustment unit 127 adjusts the rotation speed of the sweeping brush module according to the size of the garbage area. Larger garbage accumulation areas require higher rotation speeds to completely remove the garbage, while smaller or scattered garbage uses lower rotation speeds. In addition to considering the garbage area, the rotation speed adjustment also needs to comprehensively consider factors such as the ground type (such as hard pavement and soft lawn) and the type of garbage (light garbage and heavy garbage). Through the intelligent control of the sweeping brush rotation speed, not only the cleaning quality is ensured, but also the excessive wear of the cleaning tools is avoided, and the service life of the equipment is extended.

[0086] In summary, the parameter adjustment module 106 realizes the dynamic optimization of the key parameters in the cleaning process through a series of intelligent processing steps. This adaptive adjustment mechanism enables the cleaning vehicle to perform optimally in different situations, not only improving the working efficiency, but also enhancing the flexibility and reliability of the system. In this way, the municipal sanitation cleaning vehicle can complete the cleaning task of urban streets more efficiently and energy-savingly.

[0087] Second Embodiment

[0088] Please refer to Figure 8 , the present invention also provides a method for cleaning a municipal sanitation cleaning vehicle, including:

[0089] S201 Obtain the environmental image in the moving direction of the vehicle body;

[0090] The cleaning vehicle is equipped with cameras or other visual sensors, which are installed in front of or around the vehicle and can capture environmental images in the traveling direction of the cleaning vehicle in real time. The acquisition unit 109 in the image acquisition module 102 is responsible for starting the camera and continuously taking environmental images. At the same time, the parameter setting unit 108 adjusts the working parameters of the camera according to environmental factors such as the current lighting conditions to ensure the image quality.

[0091] S202 identifies the pollution type based on the environmental image, and the pollution type includes airborne dust, ground garbage, and ground dust;

[0092] Using advanced image processing techniques and machine learning algorithms (such as convolutional neural networks) to analyze the acquired environmental images, and automatically identify different types of pollutants in the images. Specifically, the pollution situation recognition module 103 first collects a large number of labeled image samples through the sample acquisition unit 114, and then the dataset generation unit 115 constructs training and validation datasets. The training unit 116 uses these datasets to train a deep learning model so that it can accurately distinguish different types of pollutants such as airborne dust, ground garbage, and ground dust. The testing unit 117 is used to evaluate the model performance and ensure its good generalization ability.

[0093] S203 obtains the coordinate information of the pollution type area in the environmental image;

[0094] Once the pollution area is identified, the coordinate extraction module 104 will determine the specific positions of these areas. Specifically, the edge detection unit 118 uses the Canny algorithm to detect the edges of the pollution area; the calibration unit 119 calculates the bounding box of each pollution area; the conversion unit 120 converts the pixel coordinates into actual physical coordinates; the display unit 121 draws the bounding box and the center point on the image for intuitive display to the operator.

[0095] S204 automatically switches to the matching cleaning device 101 for cleaning based on the coordinate information and the pollution type;

[0096] According to the identified pollution type and its location, the cleaning module 105 automatically selects an appropriate cleaning method and guides the cleaning vehicle to the corresponding location to perform the cleaning task. Specifically, the position acquisition unit 122 receives the actual physical coordinates of the pollution area; the cleaning path generation unit 123 plans the optimal cleaning route; the cleaning unit 124 controls the cleaning vehicle to move along the predetermined path and activates the corresponding cleaning device 101 (such as a strong suction module, a sweeping brush module, or a water spraying module) for operation.

[0097] S205 adjusts the working parameters of each cleaning device 101 in real time according to the pollution amount corresponding to the cleaned coordinate information.

[0098] During the cleaning process, the system continuously monitors the cleaning effect and dynamically adjusts the working parameters of the cleaning device 101 according to the remaining pollution level. Specifically, the dust amount judgment unit 125 estimates the dust amount based on the texture of the polluted area in the environmental image; the power matching unit 126 adjusts the power of the water spraying module and the strong suction module according to the dust amount; the rotation speed adjustment unit 127 adjusts the rotation speed of the sweeping brush module according to the garbage area. This ensures that the cleaning process is both effective and energy-saving.

[0099] Through the above steps, this cleaning method not only improves the cleaning efficiency but also enhances the intelligent level of cleaning, enabling the cleaning vehicle to autonomously complete high-quality cleaning work in various complex environments. The application of this method helps to improve the urban environmental hygiene, reduce the labor cost, and also promotes the development of environmental protection technologies.

[0100] What is disclosed above is only a preferred embodiment of the present invention. Of course, the scope of the rights of the present invention cannot be limited thereby. Those of ordinary skill in the art understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present invention still fall within the scope covered by the invention.

Claims

1. A municipal sanitation sweeper operation control system, comprising a sweeping device, the sweeping device comprising a powerful dust suction module, a sweeping brush module and a water spraying module, the powerful dust suction module is used to absorb dust; the sweeping brush module is used to sweep the ground; the water spraying module is used to spray water mist into the air to reduce dust, characterized in that: It also includes an image acquisition module, a pollution situation identification module, a coordinate extraction module, a cleaning module and a parameter adjustment module; The image acquisition module is used to acquire an environment image in the moving direction of the vehicle body; The pollution situation identification module is used to identify pollution types based on the environmental image, and the pollution types include air dust, ground garbage and ground ash layer; The coordinate extraction module is used to obtain the coordinate information of the pollution type area in the environmental image; The cleaning module is used to automatically switch the matching cleaning device for cleaning according to the coordinate information and the corresponding pollution type; The parameter adjustment module adjusts the working parameters of each cleaning device in real time according to the residual pollution amount corresponding to the coordinate information after cleaning.

2. A municipal sanitation sweeper operation control system as claimed in claim 1, characterized in that: The municipal sanitation sweeper operation control system also includes a voice prompt module, which is used to remind staff when switching the cleaning device.

3. A municipal sanitation sweeper operation control system as claimed in claim 2, characterized in that: The image acquisition module includes a parameter setting unit, a collection unit and an image processing unit; The parameter setting unit is used to adjust the working parameters of the camera; The acquisition unit is used to start the camera to start continuously acquiring environmental images; The image processing unit is used to pre-process the collected environment image.

4. A municipal sanitation sweeper operation control system as claimed in claim 3, characterized in that: The image processing unit includes a denoising subunit, a contrast enhancement subunit and a grayscale subunit; The denoising subunit is used to use a Gaussian filter to smooth the image and reduce random noise; The contrast enhancement subunit is used to stretch the brightness range of the image to enhance the contrast; The grayscale subunit is used to convert a color image into a grayscale image.

5. A municipal sanitation sweeper operation control system as claimed in claim 4, characterized in that: The pollution situation identification module includes a sample acquisition unit, a data set generation unit, a training unit, and a testing unit; The sample acquisition unit is used to acquire a sample of a cleaned area image, wherein the sample includes a label marking a pollution category; The data set generation unit is used to divide the image into small blocks and extract key feature points to form a data set, and divide the data set into a training set and a verification set; The training unit is used to train the convolutional neural network model using the training set to obtain a classification model; The testing unit uses the validation set to test the classification model.

6. A municipal sanitation sweeper operation control system as claimed in claim 5, characterized in that: The coordinate extraction module includes an edge detection unit, a calibration unit, a conversion unit and a display unit; The edge detection unit is used to detect the edge of the pollution type area using the Canny algorithm; The calibration unit is used to calculate a bounding box for the edge of each pollution type area; The conversion unit is used to convert the pixel coordinates of the bounding box into actual physical coordinates; The display unit is used to draw a boundary box and a center point on the environment image and display them.

7. A municipal sanitation sweeper operation control system as claimed in claim 6, characterized in that: The cleaning module includes a position acquisition unit, a cleaning path generation unit and a cleaning unit; The position acquisition unit is used to acquire the physical coordinates of the identified contaminated area; The cleaning path generation unit is used to calculate the cleaning path according to the physical coordinates; The cleaning unit is used to clean the contaminated area based on a cleaning path and a corresponding cleaning device.

8. A cleaning method for a municipal sanitation sweeper, using a municipal sanitation sweeper operation control system according to any one of claims 1 to 7, It is characterized in that It includes: obtaining an environmental image in the moving direction of the vehicle body; Identify pollution types based on the environmental image, the pollution types including airborne dust, ground garbage, and ground dust; Obtain coordinate information of pollution type areas in environmental images; Automatically switch to a matching cleaning device for cleaning based on coordinate information and pollution type; The working parameters of each cleaning device are adjusted in real time according to the pollution amount of the corresponding coordinate information after cleaning.

Images