Remote monitoring system for unmanned car wash equipment based on artificial intelligence

Abstract

The present invention discloses a remote monitoring system for unmanned car washing equipment based on artificial intelligence, and relates to the field of remote monitoring technology. The system of the present invention includes: a data collection and processing module, a cleanliness evaluation module, a historical cleaning solution analysis module, and a real-time monitoring and identification module; the data collection and processing module obtains the historical cleaning data and vehicle body image data of the unmanned car washing equipment, and constitutes a historical cleaning data set and the focus areas of different vehicle models; the cleanliness evaluation module evaluates the historical cleanliness based on the focus areas of different vehicle models, and constitutes an associated data group; the historical cleaning solution analysis module compares the associated data group, obtains the historical cleaning solution templates of different vehicle models, and adjusts the corresponding historical cleaning parameters to generate an intelligent control strategy; the real-time monitoring and identification module obtains the real-time vehicle body image data of the target vehicle, identifies the vehicle model and the focus area, and matches the corresponding historical cleaning solution template and the intelligent control strategy.

Description

Technical Field

[0001] The present invention relates to the field of remote monitoring technology, and in particular to a remote monitoring system for unmanned car washing equipment based on artificial intelligence. Background Art

[0002] With the continuous advancement of science and technology, unmanned driving and automation technology have made significant progress in many fields. Unmanned car washing equipment, as one of the important applications, has gradually become popular in the market in recent years. This type of equipment integrates high-precision sensors, robotic arms, intelligent cleaning systems and other technologies to achieve automatic cleaning of vehicles, which greatly facilitates users, improves car washing efficiency, and reduces labor costs. However, in the widespread application of unmanned car washing equipment, a significant problem has gradually emerged: that is, during the car washing process, due to insufficient equipment control accuracy or improper contact between the cleaning brush and the car body surface, it is easy to cause fine scratches on the car body surface, affecting the car washing effect and customer satisfaction.

[0003] Most of the unmanned car wash equipment on the market currently uses preset programs and fixed parameters for cleaning strategies, which are difficult to adjust in real time according to the actual conditions of the surfaces of different models of vehicles. Specifically, if the contact force, angle and movement trajectory of the cleaning brush with the surface of the vehicle body are not accurately controlled, it is easy to cause unnecessary physical wear and tear on the vehicle body during the cleaning process, resulting in scratches. Summary of the invention

[0004] The purpose of the present invention is to provide a remote monitoring system for unmanned car washing equipment based on artificial intelligence to solve the problems raised in the above background technology.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] A remote monitoring system for unmanned car wash equipment based on artificial intelligence, the system includes: data collection and processing module, cleanliness assessment module, historical cleaning program analysis module and real-time monitoring and identification module;

[0007] The data collection and processing module obtains the historical cleaning data of the unmanned car wash equipment and the corresponding vehicle image data, extracts the corresponding vehicle model information based on the vehicle image data, divides the historical cleaning data based on the vehicle model information and forms a historical cleaning data set; analyzes the vehicle image data to obtain the focus areas of different vehicle models;

[0008] The cleanliness evaluation module obtains the corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different car models, and evaluates the historical cleanliness of the unmanned car wash equipment in combination with the vehicle image data; extracts the corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleanliness to form an associated data group;

[0009] The historical cleaning solution analysis module compares and analyzes the associated data groups to obtain the historical cleaning solution templates for the focus areas corresponding to different vehicle models; based on the historical cleaning solution templates for the focus areas corresponding to different vehicle models, the corresponding historical cleaning parameters are adjusted, and the corresponding intelligent control strategy is generated according to the historical cleaning parameter adjustment process;

[0010] The real-time monitoring and recognition module obtains the real-time body image data of the target vehicle, identifies the model of the target vehicle and the corresponding area of ​​interest, and matches the corresponding historical cleaning solution template and intelligent control strategy according to the model of the target vehicle and the corresponding area of ​​interest.

[0011] Further, the data collection and processing module includes a data acquisition and extraction unit, a historical clean data set construction unit, and a region of interest identification unit;

[0012] The data acquisition and extraction unit acquires the historical cleaning data of the unmanned car washing equipment and the corresponding vehicle body image data, and extracts the corresponding vehicle model information according to the vehicle body image data; the historical cleaning data set construction unit divides the historical cleaning data based on the vehicle model information to form a historical cleaning data set; the focus area identification unit analyzes the vehicle body image data to identify the focus areas of different vehicle models.

[0013] Furthermore, the specific contents of the data acquisition and extraction unit include:

[0014] Obtain historical cleaning data of the unmanned car wash equipment and corresponding vehicle body image data, obtain vehicle body contour three-view images according to the vehicle body image data, and number the vehicle body contour three-view images according to the vehicle body to which they belong, and one vehicle body number corresponds to a group of vehicle body contour three-view images; based on the corresponding relationship between the vehicle body number and the vehicle body contour three-view images, obtain the size information of the rectangular frame tangent to the vehicle body contour three-view images corresponding to each vehicle body number, and calculate the volume of the corresponding three-dimensional space according to the size information of the rectangular frame tangent to the vehicle body contour three-view images;

[0015] The specific contents of the historical clean data set construction unit include:

[0016] According to the volume of the three-dimensional space corresponding to several preset vehicle models, all vehicle body numbers are divided into corresponding preset vehicle models; and based on the preset vehicle models and the corresponding vehicle body numbers, the historical clean data corresponding to the vehicle body numbers are divided accordingly, thereby forming a historical clean data set; and the historical clean data set includes historical clean data subsets corresponding to several vehicle models, the number of historical clean data subsets is equal to the number of preset vehicle models, and each element in the historical clean data subset corresponds to the historical clean data of one vehicle body number;

[0017] The specific contents of the focus area identification unit include:

[0018] For each subset of historical cleaning data, the corresponding vehicle body image data is obtained, and the corresponding vehicle body surface model is established according to the vehicle body image data, and the vehicle body surface model is divided into several sub-areas, and the corresponding scratch coordinates are marked on the vehicle body surface model according to the vehicle body image data; for each sub-area of ​​the vehicle body surface model, the corresponding scratch area S1 and the corresponding vehicle body surface area S are counted, and the corresponding scratch distribution density ρ is calculated, and ρ=S1 / S; the scratch distribution density of all sub-areas of the vehicle body surface model is summarized, and the vehicle body surface model and the three-dimensional space tangent to the three-view drawing of the vehicle body contour in the data acquisition and extraction unit are placed in the same world coordinate system, and the sub-area on the vehicle body surface model is mapped to the three-dimensional space tangent to the three-view drawing of the vehicle body contour, so that the three-dimensional space tangent to the three-view drawing of the vehicle body contour is also divided into several mapping sub-areas;

[0019] The scratch distribution density ρ of all mapping sub-areas in the historical clean data subset is counted, and the scratch distribution density ρ is greater than or equal to the preset scratch distribution density threshold ρ0, and the proportion of the scratch distribution density ρ of the same mapping sub-area greater than or equal to the preset scratch distribution density threshold ρ0 is greater than or equal to n. Then, the mapping sub-area that meets the above conditions is taken as the focus mapping sub-area, and the corresponding vehicle surface model sub-area is found according to the focus mapping sub-area, and it is taken as the focus area; the focus areas of all historical clean data subsets are summarized to obtain the focus areas of different models.

[0020] Further, the cleanliness evaluation module includes a cleanliness index calculation unit and a historical cleaning data association unit;

[0021] The cleaning degree index calculation unit obtains the corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different models, and calculates the historical cleaning degree index of the unmanned car washing equipment in combination with the vehicle body image data, thereby obtaining the cleaning status of the focus areas of different models; the historical cleaning data association unit extracts the corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleaning degree to form an associated data group; the historical cleaning parameters include stain parameters, scratch parameters and cleaning operation parameters, among which the cleaning operation parameters specifically refer to the contact force, angle and movement trajectory of the cleaning brush of the unmanned car washing equipment with the surface of the vehicle body.

[0022] Furthermore, the specific contents of the cleanliness index calculation unit include:

[0023] For each focus area of ​​the focus area identification unit, the corresponding historical cleaning data is obtained from the historical cleaning data set, and the corresponding cleanliness index CPI is defined according to the corresponding vehicle body image data before and after cleaning in the historical cleaning data, and CPI=(W+α·R) / A, wherein W represents the total area of ​​stains in the focus area, R represents the total area of ​​scratches in the focus area, A represents the surface area of ​​the focus area, and α represents the weight coefficient; the cleanliness index CPI of each focus area is normalized to obtain the corresponding standard cleanliness index CPI', and the cleanliness status of the focus areas of different vehicle models is evaluated according to the set cleanliness index threshold interval Q, and Q=[q1,q2], when CPI'<q1, the cleanliness status of the corresponding focus area is defined as low cleanliness, indicating the presence of stains; when q1≤CPI'≤q2, the cleanliness status of the corresponding focus area is defined as moderate cleanliness, indicating the presence of neither stains nor scratches; when CPI'>q2, the cleanliness status of the corresponding focus area is defined as excessive cleanliness, indicating the presence of scratches but no stains.

[0024] Furthermore, the historical cleaning scheme analysis module includes a data comparison unit, a parameter adjustment unit, and an intelligent control strategy generation unit;

[0025] The data comparison unit analyzes and compares the associated data groups to obtain the historical cleaning program templates for the areas of concern corresponding to different models; the parameter adjustment unit adjusts the corresponding historical cleaning parameters based on the historical cleaning program templates for the areas of concern corresponding to different models; the intelligent control strategy generation unit generates the corresponding intelligent control strategy according to the historical cleaning parameter adjustment process.

[0026] Furthermore, the specific contents of the data comparison unit include:

[0027] The cleaning operation parameters in the associated data group corresponding to each historical cleaning data subset are analyzed, and the associated data group corresponding to the historical cleaning data subset is classified according to the cleaning state of the concerned area obtained by the cleaning degree index calculation unit, and associated data groups of three cleaning states, namely, low cleaning degree, moderate cleaning degree and excessive cleaning degree, are obtained respectively; for the associated data groups of the same cleaning state, the corresponding cleaning operation parameters are compared, and the completely identical cleaning operation parameters are marked as cleaning operation parameter set A, cleaning operation parameter set B and cleaning operation parameter set C, respectively, and the cleaning operation parameter set A corresponds to the cleaning state of low cleaning degree, the cleaning operation parameter set B corresponds to the cleaning state of moderate cleaning degree, and the cleaning operation parameter set C corresponds to the cleaning state of excessive cleaning degree; and the cleaning operation parameters corresponding to the cleaning operation parameter set B are used as the historical cleaning scheme template of the concerned area;

[0028] The specific contents of the parameter adjustment unit include:

[0029] Compare the cleaning operation parameter set A and the cleaning operation parameter set C with the corresponding cleaning operation parameters in the cleaning operation parameter set B in turn, identify the differences between the cleaning operation parameter set A and the cleaning operation parameter set C and the cleaning operation parameters in the cleaning operation parameter set B, and adjust the cleaning operation parameter set A and the cleaning operation parameter set C according to the above differences and the cleaning operation parameters in the cleaning operation parameter set B;

[0030] The identification process of the distinguishing point is as follows: for each cleaning parameter in the cleaning operation parameter set A and the cleaning operation parameter set C, the difference D with the cleaning parameter in the cleaning operation parameter set B is calculated, and DA = A_i-B_i, DC = C_i-B_i, where A_i, B_i and C_i represent the i-th cleaning parameter in the cleaning operation parameter set A, B and C respectively; compare the value of DA and DC with the threshold D, for the cleaning operation parameter set A and the cleaning operation parameter set C, when DA>D, DC>D, the corresponding cleaning parameter is used as the distinguishing point;

[0031] The specific adjustment process is as follows:

[0032] According to the identification process of the difference points, the differences in the cleaning operation parameter set A and the cleaning operation parameter set C are summarized respectively to form the difference point sets A and C; for each difference point in the difference point sets A and C, it is mapped with the corresponding cleaning operation parameter in the cleaning operation parameter set B, and the difference between the mapped cleaning operation parameter and the difference point is calculated as the corresponding adjustment amplitude K, and the cleaning operation parameters in the corresponding difference point sets A and C are updated according to the adjustment amplitude, and A_i'=A_i+βK_A_i, where β represents the adjustment factor, and 0<β<1, K_A_i represents the adjustment amplitude corresponding to the i-th cleaning operation parameter in the cleaning operation parameter set A mapped by the difference point set A.

[0033] Furthermore, the real-time monitoring and identification module includes a real-time data analysis unit and a data matching and decision-making unit;

[0034] The real-time data analysis unit obtains the real-time body image data of the target vehicle, identifies the model of the target vehicle and the corresponding focus area; the target vehicle is a vehicle currently using the unmanned car washing equipment;

[0035] The data matching and decision-making unit matches the corresponding historical cleaning solution template and intelligent control strategy according to the model of the target vehicle and the corresponding area of ​​interest.

[0036] Furthermore, the specific contents of the real-time data analysis unit include:

[0037] Acquire real-time vehicle body image data of the target vehicle, obtain corresponding three-view images of the vehicle body contour according to the real-time vehicle body image data, and calculate the volume of the corresponding three-dimensional space based on the size information of the rectangular frame tangent to the three-view images of the vehicle body contour; identify the model of the target vehicle and the corresponding area of ​​interest according to the calculated volume of the three-dimensional space corresponding to the real-time vehicle body image data;

[0038] The specific contents of the data matching and decision-making unit include:

[0039] The model of the target vehicle and the corresponding area of ​​interest are obtained according to the real-time data analysis unit, and several historical cleaning solution templates are matched according to the model of the target vehicle and the corresponding area of ​​interest, and similarity calculation is performed based on the real-time body image data and the body image data corresponding to the preset model in the historical cleaning data set, and the historical cleaning solution template with the largest similarity is selected as the cleaning solution for the target vehicle; the target vehicle is cleaned according to the matched historical cleaning solution template, and the corresponding cleaning parameters are obtained in real time; the difference D0 between the cleaning parameters and the cleaning parameters corresponding to the historical cleaning solution template is compared, the corresponding difference points are identified, and the similarity calculation is performed based on the currently identified difference points and the difference point sets A and C, the maximum value of the similarity calculation is selected to match the corresponding difference point set A, and the corresponding intelligent control strategy is matched according to the same adjustment method of the parameter adjustment unit of the difference point set A.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: through the real-time monitoring and identification module, the system can adjust the cleaning strategy in real time according to the actual surface conditions of different vehicle models, avoiding unsatisfactory cleaning effects or damage to the vehicle body caused by fixed procedures; using the cleaning degree evaluation module and the historical cleaning solution analysis module, the system can accurately control the contact force, angle and movement trajectory of the cleaning brush, significantly reducing the risk of scratches caused by improper operation; based on the historical cleaning solution templates generated for vehicle models and areas of concern, the cleaning strategy is more personalized, and more effective cleaning solutions can be formulated for the actual conditions of different vehicles; through the analysis and comparison of historical cleaning data, the system can continuously optimize the cleaning parameters, and based on the generation of intelligent control strategies, realize self-improvement of the cleaning process and enhance the effect of long-term use; real-time matching of the best cleaning solution template improves cleaning efficiency, while ensuring cleaning quality, reducing the need for repeated cleaning, and thus saving time and resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0042] Figure 1 It is a schematic diagram of a remote monitoring system module of an unmanned car washing device based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] See also Figure 1 , the present invention provides a technical solution:

[0045] A remote monitoring system for unmanned car wash equipment based on artificial intelligence, the system includes: data collection and processing module, cleanliness assessment module, historical cleaning program analysis module and real-time monitoring and identification module;

[0046] The data collection and processing module obtains the historical cleaning data of the unmanned car wash equipment and the corresponding vehicle image data, extracts the corresponding vehicle model information based on the vehicle image data, divides the historical cleaning data based on the vehicle model information and forms a historical cleaning data set; analyzes the vehicle image data to obtain the focus areas of different vehicle models;

[0047] The cleanliness evaluation module obtains the corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different car models, and evaluates the historical cleanliness of the unmanned car wash equipment in combination with the vehicle image data; extracts the corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleanliness to form an associated data group;

[0048] The historical cleaning solution analysis module compares and analyzes the associated data groups to obtain the historical cleaning solution templates for the focus areas corresponding to different vehicle models; based on the historical cleaning solution templates for the focus areas corresponding to different vehicle models, the corresponding historical cleaning parameters are adjusted, and the corresponding intelligent control strategy is generated according to the historical cleaning parameter adjustment process;

[0049] The real-time monitoring and recognition module obtains the real-time body image data of the target vehicle, identifies the model of the target vehicle and the corresponding area of ​​interest, and matches the corresponding historical cleaning solution template and intelligent control strategy according to the model of the target vehicle and the corresponding area of ​​interest.

[0050] The data collection and processing module includes a data acquisition and extraction unit, a historical clean data set construction unit, and a region of interest identification unit;

[0051] The data acquisition and extraction unit acquires the historical cleaning data of the unmanned car washing equipment and the corresponding vehicle body image data, and extracts the corresponding vehicle model information according to the vehicle body image data; the historical cleaning data set construction unit divides the historical cleaning data based on the vehicle model information to form a historical cleaning data set; the focus area identification unit analyzes the vehicle body image data to identify the focus areas of different vehicle models.

[0052] The specific contents of the data acquisition and extraction unit include:

[0053] Obtain historical cleaning data of the unmanned car wash equipment and corresponding vehicle body image data, obtain vehicle body contour three-view images according to the vehicle body image data, and number the vehicle body contour three-view images according to the vehicle body to which they belong, and one vehicle body number corresponds to a group of vehicle body contour three-view images; based on the corresponding relationship between the vehicle body number and the vehicle body contour three-view images, obtain the size information of the rectangular frame tangent to the vehicle body contour three-view images corresponding to each vehicle body number, and calculate the volume of the corresponding three-dimensional space according to the size information of the rectangular frame tangent to the vehicle body contour three-view images;

[0054] In this embodiment, the specific process of obtaining the three-view image of the vehicle body contour according to the historical cleaning data of the unmanned car washing equipment and the vehicle body image data and calculating the corresponding three-dimensional space volume is as follows:

[0055] The historical cleaning data and the corresponding vehicle body image data are obtained from the unmanned car washing equipment. Assume that there is a set of vehicle body image data, which includes three images: a front view, a side view, and a top view.

[0056] For the vehicle number "CAR001", extract the three corresponding vehicle images, use the edge detection algorithm to process the images to obtain the outer contour of the vehicle body, assuming that the three views of the vehicle body contour corresponding to the number "CAR001" are: front view (W×H), side view (L×H) and top view (L×W);

[0057] Based on the outline of each view, calculate the size of the tangent rectangle.

[0058] Assumptions:

[0059] The rectangular frame size of the front view is: width (W_1 = 2.0) m, height (H_1 = 1.5) m,

[0060] The dimensions of the rectangular frame of the side view are: length (L_1 = 4.5) m, height (H_2 = 1.5) m,

[0061] The dimensions of the rectangular frame of the top view are: length (L_2 = 4.5) m, width (W_2 = 2.0) m;

[0062] The three-dimensional space can be approximated as a cuboid, whose dimensions are provided by the rectangular frame mentioned above.

[0063] Based on these dimensions, calculate the volume (V):

[0064] V=L_1×W_1×H_1=4.5×2.0×1.5=13.5(m^3).

[0065] The specific contents of the historical clean data set construction unit include:

[0066] According to the volume of the three-dimensional space corresponding to several preset vehicle models, all vehicle body numbers are divided into corresponding preset vehicle models; and based on the preset vehicle models and the corresponding vehicle body numbers, the historical clean data corresponding to the vehicle body numbers are divided accordingly, thereby forming a historical clean data set; and the historical clean data set includes historical clean data subsets corresponding to several vehicle models, the number of historical clean data subsets is equal to the number of preset vehicle models, and each element in the historical clean data subset corresponds to the historical clean data of one vehicle body number;

[0067] The specific contents of the focus area identification unit include:

[0068] For each subset of historical cleaning data, the corresponding vehicle body image data is obtained, and the corresponding vehicle body surface model is established according to the vehicle body image data, and the vehicle body surface model is divided into several sub-areas, and the corresponding scratch coordinates are marked on the vehicle body surface model according to the vehicle body image data; for each sub-area of ​​the vehicle body surface model, the corresponding scratch area S1 and the corresponding vehicle body surface area S are counted, and the corresponding scratch distribution density ρ is calculated, and ρ=S1 / S; the scratch distribution density of all sub-areas of the vehicle body surface model is summarized, and the vehicle body surface model and the three-dimensional space tangent to the three-view drawing of the vehicle body contour in the data acquisition and extraction unit are placed in the same world coordinate system, and the sub-area on the vehicle body surface model is mapped to the three-dimensional space tangent to the three-view drawing of the vehicle body contour, so that the three-dimensional space tangent to the three-view drawing of the vehicle body contour is also divided into several mapping sub-areas;

[0069] The scratch distribution density ρ of all mapping sub-areas in the historical clean data subset is counted, and the scratch distribution density ρ is greater than or equal to the preset scratch distribution density threshold ρ0, and the proportion of the scratch distribution density ρ of the same mapping sub-area greater than or equal to the preset scratch distribution density threshold ρ0 is greater than or equal to n. Then, the mapping sub-area that meets the above conditions is taken as the focus mapping sub-area, and the corresponding vehicle surface model sub-area is found according to the focus mapping sub-area, and it is taken as the focus area; the focus areas of all historical clean data subsets are summarized to obtain the focus areas of different models.

[0070] The cleanliness evaluation module includes a cleanliness index calculation unit and a historical cleaning data association unit;

[0071] The cleaning degree index calculation unit obtains the corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different models, and calculates the historical cleaning degree index of the unmanned car washing equipment in combination with the vehicle body image data, thereby obtaining the cleaning status of the focus areas of different models; the historical cleaning data association unit extracts the corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleaning degree to form an associated data group; the historical cleaning parameters include stain parameters, scratch parameters and cleaning operation parameters, among which the cleaning operation parameters specifically refer to the contact force, angle and movement trajectory of the cleaning brush of the unmanned car washing equipment with the surface of the vehicle body.

[0072] The specific contents of the cleanliness index calculation unit include:

[0073] For each focus area of ​​the focus area identification unit, the corresponding historical cleaning data is obtained from the historical cleaning data set, and the corresponding cleanliness index CPI is defined according to the corresponding vehicle body image data before and after cleaning in the historical cleaning data, and CPI=(W+α·R) / A, wherein W represents the total area of ​​stains in the focus area, R represents the total area of ​​scratches in the focus area, A represents the surface area of ​​the focus area, and α represents the weight coefficient; the cleanliness index CPI of each focus area is normalized to obtain the corresponding standard cleanliness index CPI', and the cleanliness status of the focus areas of different vehicle models is evaluated according to the set cleanliness index threshold interval Q, and Q=[q1,q2], when CPI'<q1, the cleanliness status of the corresponding focus area is defined as low cleanliness, indicating the presence of stains; when q1≤CPI'≤q2, the cleanliness status of the corresponding focus area is defined as moderate cleanliness, indicating the presence of neither stains nor scratches; when CPI'>q2, the cleanliness status of the corresponding focus area is defined as excessive cleanliness, indicating the presence of scratches but no stains.

[0074] The historical cleaning scheme analysis module includes a data comparison unit, a parameter adjustment unit, and an intelligent control strategy generation unit;

[0075] The data comparison unit analyzes and compares the associated data groups to obtain the historical cleaning program templates for the areas of concern corresponding to different models; the parameter adjustment unit adjusts the corresponding historical cleaning parameters based on the historical cleaning program templates for the areas of concern corresponding to different models; the intelligent control strategy generation unit generates the corresponding intelligent control strategy according to the historical cleaning parameter adjustment process.

[0076] The specific contents of the data comparison unit include:

[0077] The cleaning operation parameters in the associated data group corresponding to each historical cleaning data subset are analyzed, and the associated data group corresponding to the historical cleaning data subset is classified according to the cleaning state of the concerned area obtained by the cleaning degree index calculation unit, and associated data groups of three cleaning states, namely, low cleaning degree, moderate cleaning degree and excessive cleaning degree, are obtained respectively; for the associated data groups of the same cleaning state, the corresponding cleaning operation parameters are compared, and the completely identical cleaning operation parameters are marked as cleaning operation parameter set A, cleaning operation parameter set B and cleaning operation parameter set C, respectively, and the cleaning operation parameter set A corresponds to the cleaning state of low cleaning degree, the cleaning operation parameter set B corresponds to the cleaning state of moderate cleaning degree, and the cleaning operation parameter set C corresponds to the cleaning state of excessive cleaning degree; and the cleaning operation parameters corresponding to the cleaning operation parameter set B are used as the historical cleaning scheme template of the concerned area;

[0078] The specific contents of the parameter adjustment unit include:

[0079] Compare the cleaning operation parameter set A and the cleaning operation parameter set C with the corresponding cleaning operation parameters in the cleaning operation parameter set B in turn, identify the differences between the cleaning operation parameter set A and the cleaning operation parameter set C and the cleaning operation parameters in the cleaning operation parameter set B, and adjust the cleaning operation parameter set A and the cleaning operation parameter set C according to the above differences and the cleaning operation parameters in the cleaning operation parameter set B;

[0080] The identification process of the distinguishing point is as follows: for each cleaning parameter in the cleaning operation parameter set A and the cleaning operation parameter set C, the difference D with the cleaning parameter in the cleaning operation parameter set B is calculated, and DA = A_i-B_i, DC = C_i-B_i, where A_i, B_i and C_i represent the i-th cleaning parameter in the cleaning operation parameter set A, B and C respectively; compare the value of DA and DC with the threshold D, for the cleaning operation parameter set A and the cleaning operation parameter set C, when DA>D, DC>D, the corresponding cleaning parameter is used as the distinguishing point;

[0081] The specific adjustment process is as follows:

[0082] According to the identification process of the difference points, the differences in the cleaning operation parameter set A and the cleaning operation parameter set C are summarized respectively to form the difference point sets A and C; for each difference point in the difference point sets A and C, it is mapped with the corresponding cleaning operation parameter in the cleaning operation parameter set B, and the difference between the mapped cleaning operation parameter and the difference point is calculated as the corresponding adjustment amplitude K, and the cleaning operation parameters in the corresponding difference point sets A and C are updated according to the adjustment amplitude, and A_i'=A_i+βK_A_i, where β represents the adjustment factor, and 0<β<1, K_A_i represents the adjustment amplitude corresponding to the i-th cleaning operation parameter in the cleaning operation parameter set A mapped by the difference point set A.

[0083] The real-time monitoring and identification module includes a real-time data analysis unit and a data matching and decision-making unit;

[0084] The real-time data analysis unit obtains the real-time body image data of the target vehicle, identifies the model of the target vehicle and the corresponding focus area; the target vehicle is a vehicle currently using the unmanned car washing equipment;

[0085] The data matching and decision-making unit matches the corresponding historical cleaning solution template and intelligent control strategy according to the model of the target vehicle and the corresponding area of ​​interest.

[0086] The specific contents of the real-time data analysis unit include:

[0087] Acquire real-time vehicle body image data of the target vehicle, obtain corresponding three-view images of the vehicle body contour according to the real-time vehicle body image data, and calculate the volume of the corresponding three-dimensional space based on the size information of the rectangular frame tangent to the three-view images of the vehicle body contour; identify the model of the target vehicle and the corresponding area of ​​interest according to the calculated volume of the three-dimensional space corresponding to the real-time vehicle body image data;

[0088] The specific contents of the data matching and decision-making unit include:

[0089] The model of the target vehicle and the corresponding area of ​​interest are obtained according to the real-time data analysis unit, and several historical cleaning solution templates are matched according to the model of the target vehicle and the corresponding area of ​​interest, and similarity calculation is performed based on the real-time body image data and the body image data corresponding to the preset model in the historical cleaning data set, and the historical cleaning solution template with the largest similarity is selected as the cleaning solution for the target vehicle; the target vehicle is cleaned according to the matched historical cleaning solution template, and the corresponding cleaning parameters are obtained in real time; the difference D0 between the cleaning parameters and the cleaning parameters corresponding to the historical cleaning solution template is compared, the corresponding difference points are identified, and the similarity calculation is performed based on the currently identified difference points and the difference point sets A and C, the maximum value of the similarity calculation is selected to match the corresponding difference point set A, and the corresponding intelligent control strategy is matched according to the same adjustment method of the parameter adjustment unit of the difference point set A.

[0090] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0091] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. 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. The remote monitoring system of unmanned car washing equipment based on artificial intelligence is characterized by: The system includes: a data collection and processing module, a cleaning degree assessment module, a historical cleaning solution analysis module, and a real-time monitoring and identification module; The data collection and processing module obtains the historical cleaning data of the unmanned car washing equipment and the corresponding vehicle body image data, extracts the corresponding vehicle model information according to the vehicle body image data, divides the historical cleaning data based on the vehicle model information and forms a historical cleaning data set; analyzes the vehicle body image data to obtain the focus areas of different vehicle models; The cleanliness evaluation module obtains corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different vehicle models, and evaluates the historical cleanliness of the unmanned car washing equipment in combination with the vehicle body image data; extracts corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleanliness to form an associated data group; The historical cleaning solution analysis module compares and analyzes the associated data group to obtain the historical cleaning solution templates of the concerned areas corresponding to different vehicle models; based on the historical cleaning solution templates of the concerned areas corresponding to different vehicle models, the corresponding historical cleaning parameters are adjusted, and the corresponding intelligent control strategy is generated according to the historical cleaning parameter adjustment process; The real-time monitoring and identification module obtains real-time body image data of the target vehicle, identifies the model of the target vehicle and the corresponding focus area, and matches the corresponding historical cleaning solution template and intelligent control strategy according to the model of the target vehicle and the corresponding focus area.

2. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 1 is characterized in that: The data collection and processing module includes a data acquisition and extraction unit, a historical clean data set construction unit, and a region of interest identification unit; The data acquisition and extraction unit acquires the historical cleaning data of the unmanned car washing equipment and the corresponding vehicle body image data, and extracts the corresponding vehicle model information according to the vehicle body image data; the historical cleaning data set construction unit divides the historical cleaning data based on the vehicle model information to form a historical cleaning data set; the focus area identification unit analyzes the vehicle body image data to identify the focus areas of different vehicle models.

3. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 2 is characterized in that: The data acquisition and extraction unit specifically includes: Obtain historical cleaning data of the unmanned car wash equipment and corresponding vehicle body image data, obtain vehicle body contour three-view images according to the vehicle body image data, and number the vehicle body contour three-view images according to the vehicle body to which they belong, and one vehicle body number corresponds to a group of vehicle body contour three-view images; based on the corresponding relationship between the vehicle body number and the vehicle body contour three-view images, obtain the size information of the rectangular frame tangent to the vehicle body contour three-view images corresponding to each vehicle body number, and calculate the volume of the corresponding three-dimensional space according to the size information of the rectangular frame tangent to the vehicle body contour three-view images; The specific contents of the historical clean data set construction unit include: According to the volume of the three-dimensional space corresponding to several preset vehicle models, all vehicle body numbers are divided into corresponding preset vehicle models; and based on the preset vehicle models and the corresponding vehicle body numbers, the historical clean data corresponding to the vehicle body numbers are divided accordingly, thereby forming a historical clean data set; and the historical clean data set includes historical clean data subsets corresponding to several vehicle models, the number of historical clean data subsets is equal to the number of preset vehicle models, and each element in the historical clean data subset corresponds to the historical clean data of one vehicle body number; The specific contents of the focus area identification unit include: For each subset of historical cleaning data, the corresponding vehicle body image data is obtained, and the corresponding vehicle body surface model is established according to the vehicle body image data, and the vehicle body surface model is divided into several sub-areas, and the corresponding scratch coordinates are marked on the vehicle body surface model according to the vehicle body image data; for each sub-area of ​​the vehicle body surface model, the corresponding scratch area S1 and the corresponding vehicle body surface area S are counted, and the corresponding scratch distribution density ρ is calculated, and ρ=S1 / S; the scratch distribution density of all sub-areas of the vehicle body surface model is summarized, and the vehicle body surface model and the three-dimensional space tangent to the three-view drawing of the vehicle body contour in the data acquisition and extraction unit are placed in the same world coordinate system, and the sub-area on the vehicle body surface model is mapped to the three-dimensional space tangent to the three-view drawing of the vehicle body contour, so that the three-dimensional space tangent to the three-view drawing of the vehicle body contour is also divided into several mapping sub-areas; The scratch distribution density ρ of all mapping sub-areas in the historical clean data subset is counted, and the scratch distribution density ρ is greater than or equal to the preset scratch distribution density threshold ρ0, and the proportion of the scratch distribution density ρ of the same mapping sub-area greater than or equal to the preset scratch distribution density threshold ρ0 is greater than or equal to n. Then, the mapping sub-area that meets the above conditions is taken as the focus mapping sub-area, and the corresponding vehicle surface model sub-area is found according to the focus mapping sub-area, and it is taken as the focus area; the focus areas of all historical clean data subsets are summarized to obtain the focus areas of different models.

4. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 1 is characterized in that: The cleanliness evaluation module includes a cleanliness index calculation unit and a historical cleaning data association unit; The cleaning degree index calculation unit obtains corresponding historical cleaning data from the historical cleaning data set based on the focus areas of different vehicle models, and calculates the historical cleaning degree index of the unmanned car washing equipment in combination with the vehicle body image data, thereby obtaining the cleaning status of the focus areas of different vehicle models; the historical cleaning data association unit extracts corresponding historical cleaning parameters from the historical cleaning data, and associates the historical cleaning parameters with the historical cleaning degree to form an associated data group; the historical cleaning parameters include stain parameters, scratch parameters and cleaning operation parameters, wherein the cleaning operation parameters specifically refer to the contact force, angle and movement trajectory of the cleaning brush of the unmanned car washing equipment with the surface of the vehicle body.

5. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 4 is characterized in that: The specific contents of the cleaning index calculation unit include: For each focus area of ​​the focus area identification unit, the corresponding historical cleaning data is obtained from the historical cleaning data set, and the corresponding cleanliness index CPI is defined according to the corresponding vehicle body image data before and after cleaning in the historical cleaning data, and CPI=(W+α·R) / A, wherein W represents the total area of ​​stains in the focus area, R represents the total area of ​​scratches in the focus area, A represents the surface area of ​​the focus area, and α represents the weight coefficient; the cleanliness index CPI of each focus area is normalized to obtain the corresponding standard cleanliness index CPI', and the cleanliness status of the focus areas of different vehicle models is evaluated according to the set cleanliness index threshold interval Q, and Q=[q1,q2], when CPI'<q1, the cleanliness status of the corresponding focus area is defined as low cleanliness, indicating the presence of stains; when q1≤CPI'≤q2, the cleanliness status of the corresponding focus area is defined as moderate cleanliness, indicating the presence of neither stains nor scratches; when CPI'>q2, the cleanliness status of the corresponding focus area is defined as excessive cleanliness, indicating the presence of scratches but no stains.

6. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 1 is characterized by: The historical cleaning scheme analysis module includes a data comparison unit, a parameter adjustment unit and an intelligent control strategy generation unit; The data comparison unit analyzes and compares the associated data group, thereby obtaining a historical cleaning solution template for the area of ​​interest corresponding to different vehicle models; The parameter adjustment unit adjusts the corresponding historical cleaning parameters based on the historical cleaning solution templates of the areas of interest corresponding to different vehicle models; The intelligent control strategy generation unit generates a corresponding intelligent control strategy according to the historical cleaning parameter adjustment process.

7. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 6 is characterized by: The specific contents of the data comparison unit include: The cleaning operation parameters in the associated data group corresponding to each historical cleaning data subset are analyzed, and the associated data group corresponding to the historical cleaning data subset is classified according to the cleaning state of the concerned area obtained by the cleaning degree index calculation unit, and associated data groups of three cleaning states, namely, low cleaning degree, moderate cleaning degree and excessive cleaning degree, are obtained respectively; for the associated data groups of the same cleaning state, the corresponding cleaning operation parameters are compared, and the completely identical cleaning operation parameters are marked as cleaning operation parameter set A, cleaning operation parameter set B and cleaning operation parameter set C, respectively, and the cleaning operation parameter set A corresponds to the cleaning state of low cleaning degree, the cleaning operation parameter set B corresponds to the cleaning state of moderate cleaning degree, and the cleaning operation parameter set C corresponds to the cleaning state of excessive cleaning degree; and the cleaning operation parameters corresponding to the cleaning operation parameter set B are used as the historical cleaning scheme template of the concerned area; The parameter adjustment unit specifically includes: Compare the cleaning operation parameter set A and the cleaning operation parameter set C with the corresponding cleaning operation parameters in the cleaning operation parameter set B in turn, identify the differences between the cleaning operation parameter set A and the cleaning operation parameter set C and the cleaning operation parameters in the cleaning operation parameter set B, and adjust the cleaning operation parameter set A and the cleaning operation parameter set C according to the above differences and the cleaning operation parameters in the cleaning operation parameter set B; The identification process of the distinguishing point is as follows: for each cleaning parameter in the cleaning operation parameter set A and the cleaning operation parameter set C, the difference D with the cleaning parameter in the cleaning operation parameter set B is calculated, and DA = A_i-B_i, DC = C_i-B_i, where A_i, B_i and C_i represent the i-th cleaning parameter in the cleaning operation parameter set A, B and C respectively; compare the value of DA and DC with the threshold D, for the cleaning operation parameter set A and the cleaning operation parameter set C, when DA>D, DC>D, the corresponding cleaning parameter is used as the distinguishing point; The specific adjustment process is as follows: According to the identification process of the difference points, the differences in the cleaning operation parameter set A and the cleaning operation parameter set C are summarized respectively to form the difference point sets A and C; for each difference point in the difference point sets A and C, it is mapped with the corresponding cleaning operation parameter in the cleaning operation parameter set B, and the difference between the mapped cleaning operation parameter and the difference point is calculated as the corresponding adjustment amplitude K, and the cleaning operation parameters in the corresponding difference point sets A and C are updated according to the adjustment amplitude, and A_i'=A_i+βK_A_i, where β represents the adjustment factor, and 0<β<1, K_A_i represents the adjustment amplitude corresponding to the i-th cleaning operation parameter in the cleaning operation parameter set A mapped by the difference point set A.

8. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 1 is characterized by: The real-time monitoring and identification module includes a real-time data analysis unit and a data matching and decision-making unit; The real-time data analysis unit acquires real-time vehicle body image data of a target vehicle, identifies the model of the target vehicle and the corresponding focus area; the target vehicle is a vehicle currently using an unmanned vehicle washing device; The data matching and decision-making unit matches the corresponding historical cleaning solution template and the intelligent control strategy according to the model of the target vehicle and the corresponding focus area.

9. The remote monitoring system for unmanned car washing equipment based on artificial intelligence according to claim 8 is characterized by: The specific contents of the real-time data analysis unit include: Acquire real-time vehicle body image data of the target vehicle, obtain corresponding three-view images of the vehicle body contour according to the real-time vehicle body image data, and calculate the volume of the corresponding three-dimensional space based on the size information of the rectangular frame tangent to the three-view images of the vehicle body contour; identify the model of the target vehicle and the corresponding area of ​​interest according to the calculated volume of the three-dimensional space corresponding to the real-time vehicle body image data; The specific contents of the data matching and decision-making unit include: The model of the target vehicle and the corresponding area of ​​interest are obtained according to the real-time data analysis unit, and several historical cleaning solution templates are matched according to the model of the target vehicle and the corresponding area of ​​interest, and similarity calculation is performed based on the real-time body image data and the body image data corresponding to the preset model in the historical cleaning data set, and the historical cleaning solution template with the largest similarity is selected as the cleaning solution for the target vehicle; the target vehicle is cleaned according to the matched historical cleaning solution template, and the corresponding cleaning parameters are obtained in real time; the difference D0 between the cleaning parameters and the cleaning parameters corresponding to the historical cleaning solution template is compared, the corresponding difference points are identified, and the similarity calculation is performed based on the currently identified difference points and the difference point sets A and C, the maximum value of the similarity calculation is selected to match the corresponding difference point set A, and the corresponding intelligent control strategy is matched according to the same adjustment method of the parameter adjustment unit of the difference point set A.

Images