An intelligent control system and method for unmanned car washing equipment

Abstract

The present invention discloses an intelligent control system and method for unmanned car washing equipment, and relates to the technical field of intelligent equipment control. The present invention divides the outer surface of a vehicle body to be washed into a plurality of target cleaning monitoring areas by performing stain area frame selection on a stain area distribution image of the outer surface of the vehicle body to be washed, and then analyzes the stain distribution changes presented after the unmanned car washing equipment continuously executes cleaning operation instructions on the vehicle in the target cleaning monitoring area, thereby evaluating and calculating the cleaning state index presented accordingly by the unmanned car washing equipment in the target cleaning area, capturing characteristic cleaning monitoring areas showing cleaning efficiency lower than the average value, and collecting data of the program executed by the control terminal of the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths, so as to realize continuous improvement of the adaptability and automation degree of the unmanned car washing equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent equipment control, and in particular to an intelligent control system and method applied to unmanned car washing equipment. Background Art

[0002] Intelligent control system refers to the automated control of intelligent equipment to achieve full-process automated operation of equipment without the need for human supervision, greatly saving labor costs. For example, the Chinese patent (CN117885698A) discloses "a fully automatic car washing machine that can intelligently identify vehicle models", which specifically discloses: building and optimizing vehicle models, planning cleaning trajectories based on the vehicle models, dynamically adjusting the cleaning trajectories, identifying key cleaning areas such as windows, wheels, and roofs, using Bezier curves to describe smooth paths for key cleaning areas, and planning cleaning trajectories for brushes and nozzles.

[0003] However, in actual operation, unmanned car wash equipment can never reach the upper limit of human operation. Although unmanned car wash equipment can identify the model and size of the vehicle through sensors and image recognition technology, this automatic recognition may not be able to fully adapt to all complex situations, such as complex stains on the vehicle surface or surface treatment of special materials. In addition, although the brushing process of unmanned car wash equipment uses a soft brush for 360-degree rotation without dead angles, this brushing method may not be as flexible as manual car washing to deal with various stains and details; secondly, technical limitations are also an important factor affecting the cleaning effect. Although unmanned car wash equipment can automatically adjust the spraying amount and mode during the spraying and rinsing process of car wash liquid, this adjustment may not be able to fully adapt to the actual situation of all vehicles. Manual car washing can perform more detailed operations according to specific circumstances, such as using different detergents or tools to deal with different stains. Summary of the invention

[0004] The object of the present invention is to provide an intelligent control system and method for unmanned car washing equipment to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: an intelligent control method applied to unmanned car washing equipment, the method comprising:

[0006] Step S1: Whenever a vehicle to be washed enters the active car washing area, the unmanned car washing equipment is triggered to obtain the body model and size information of the vehicle to be washed, as well as the stain area distribution image on the outer surface of the vehicle to be washed based on laser scanning technology;

[0007] Step S2: creating a unit frame selection area by using a frame selection tool, frame selection the stain area on the stain area distribution image by using the unit frame selection area, obtaining an optimal frame selection scheme for the stain area distribution image, and extracting a target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the optimal frame selection scheme;

[0008] Step S3: for each historical cleaning record of the unmanned car washing device, sort out the stain distribution changes presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing device executing the corresponding cleaning operation instructions; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard;

[0009] Step S4: sorting out the cleaning consumption of each target cleaning monitoring area by the unmanned car washing equipment in each historical cleaning record, and calculating the corresponding cleaning status index for each target cleaning monitoring area;

[0010] Step S5: Sort out the distribution of numerical differences in the cleaning status index between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas with cleaning efficiency lower than the average, and identify and extract data of the program executed by the control terminal that can assist in optimizing the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths.

[0011] Preferably: step S2 comprises:

[0012] Step S2-1: Constraints that need to be satisfied when selecting a stain area on the stain area distribution image using a unit selection area include: in a unit image selected from the stain area distribution image by each unit selection area, the total area of ​​the distributed stain area is greater than or equal to an area threshold; based on the unit image selected by each unit selection area, vehicle body part structure information can be identified, and the number of vehicle body part structures that can be identified based on one unit image is 1; wherein the vehicle body part structure includes: windows, rearview mirrors, vehicle body exteriors, bumpers, fenders, ABCD pillars, side door frame components, engine hoods, trunks, roof covers, doors, wheel covers, door handles, air intake grilles, etc.;

[0013] Step S2-2: Set the selection scheme that can select all the stain areas distributed in the stain area distribution image in a manner that meets the constraint conditions and requires the least number of unit selection areas to be created as the best scheme; mark the position of the vehicle body surface area corresponding to each unit selection area in the best scheme and set it as the target cleaning monitoring area.

[0014] Preferably: step S3 comprises:

[0015] Step S3-1: In each historical cleaning record of the unmanned car washing device, the target cleaning monitoring area locked for the corresponding vehicle to be washed entering the active car washing area is extracted, and the stain degree value obtained by the unmanned car washing device for evaluating each target cleaning monitoring area before the corresponding vehicle to be washed is fully automatically cleaned is set as the first target value of each target cleaning monitoring area;

[0016] Step S3-2: In each historical cleaning record of the unmanned car washing device, for each corresponding target cleaning monitoring area, extract the sequence of the stain degree value change presented by the unmanned car washing device according to the order in which the corresponding cleaning operation instructions are executed, and evaluate the corresponding target cleaning monitoring area;

[0017] Step S3-3: The first soil level value that meets the cleaning standard in the soil level value change sequence is set as the second target value of each corresponding target cleaning monitoring area; for each corresponding target cleaning monitoring area, the cleaning operation instruction sequence executed by the unmanned car washing device is extracted in the process that the corresponding soil level value changes from the corresponding first target value to the corresponding second target value;

[0018] The unmanned car wash equipment described in this application, that is, the fully automatic car wash equipment, is usually equipped with touch sensors and pressure sensors, which can detect the type and density of stains on the car body. Among them, the touch sensor can sense the particles on the surface of the car body, such as dust and dirt, and the pressure sensor can detect the hardness and size of the stains, such as resin, oil stains and bird droppings. Usually, by analyzing the data of the sensor, the car wash equipment can preliminarily judge the degree of stains on the car body. Secondly, it is also equipped with a camera and a computer vision system, which can perform high-precision image recognition and analysis on the car body. Among them, the camera can capture the details of the stains on the car body and transmit the image to the computer vision system for processing. The computer vision system can use image processing algorithms to identify different areas and colors in the image, and further analyze and obtain the stain degree value.

[0019] Preferably: step S4 comprises:

[0020] Step S4-1: Assume that in a certain historical cleaning record, in the process of capturing the corresponding stain level value of a certain target cleaning monitoring area A changing from the corresponding first target value a1 to the corresponding second target value a2, the cleaning operation instruction sequence executed by the unmanned car washing device is P;

[0021] Step S4-2: obtaining a total number Y of cleaning operation instructions included in the cleaning operation instruction sequence P, and accumulating a total number of types of cleaning operation instructions included in the Y cleaning operation instructions as N;

[0022] Step S4-3: Collect the total time T taken by the unmanned car wash equipment to complete the cleaning operation instruction sequence P, set the unit reference time Tr; calculate the cleaning state index β of a target cleaning monitoring area A = |a1-a2| / [N×Y×(T / Tr)].

[0023] Preferably: step S5 comprises:

[0024] Step S5-1: obtaining the body model and size information of the corresponding vehicle to be cleaned in each historical cleaning record, and setting the body model and size information as the first attribute data of each target cleaning monitoring area extracted from each historical cleaning record; extracting the corresponding body part structure that can be identified based on each target cleaning monitoring area, and setting the body part structure as the second attribute data of each target cleaning monitoring area;

[0025] Step S5-2: traverse the first attribute data and the second attribute data of each target cleaning monitoring area extracted from all historical cleaning records; respectively collect all target cleaning monitoring areas whose corresponding first attribute data satisfy a similarity greater than or equal to a similarity threshold and whose corresponding second attribute data are the same, to obtain a plurality of target cleaning monitoring area sets;

[0026] Step S5-3: Calculate the average cleaning state index of all target cleaning monitoring areas included in each target cleaning monitoring area set; capture the target cleaning monitoring area whose corresponding cleaning state index is greater than the corresponding average cleaning state index in each target cleaning monitoring area set, and set it as a characteristic cleaning monitoring area;

[0027] Step S5-4: for each characteristic cleaning monitoring area, before fully automatic cleaning of the corresponding vehicle to be cleaned, the unit image corresponding to each characteristic cleaning monitoring area is captured from the corresponding stain area distribution image of the vehicle to be cleaned based on the unit frame selection area from the corresponding stain area distribution image of the vehicle to be cleaned, and the unit image and the position characteristic distribution information between the corresponding stain area and the corresponding vehicle body part structure obtained from the unit image recognition are stored, and the position characteristic distribution information includes the position orientation distribution between each stain area and the corresponding vehicle body part structure;

[0028] Step S5-5: regularly transmit all stored unit images and position feature distribution information to the management terminal, and assist the management terminal in optimizing the execution program of the control terminal of the unmanned car washing equipment; the execution program refers to the program executed by the control terminal when generating corresponding cleaning operation instructions and cleaning operation paths according to the body model, size information and stain distribution of the vehicle to be washed entering the car washing area.

[0029] In order to better implement the above method, an intelligent control system is also proposed, the system includes:

[0030] The vehicle data acquisition module is used to collect the body model and size information of the vehicle to be cleaned obtained by the unmanned car washing equipment based on laser scanning technology, as well as the stain area distribution image on the outer surface of the vehicle to be cleaned;

[0031] A target cleaning monitoring area extraction module is used to create a unit frame selection area through a frame selection tool, frame the stain area on the stain area distribution image with the unit frame selection area, obtain the best frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the best frame selection scheme;

[0032] The stain distribution change monitoring and management module is used to sort out the stain distribution changes presented by each target cleaning monitoring area of ​​the corresponding vehicle to be cleaned during the process of the unmanned car washing equipment executing the corresponding cleaning operation instructions for each historical cleaning record of the unmanned car washing equipment; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard;

[0033] The cleaning status index calculation module is used to sort out the cleaning consumption of the unmanned car washing equipment for each corresponding target cleaning monitoring area in each historical cleaning record, and calculate the corresponding cleaning status index for each corresponding target cleaning monitoring area;

[0034] The optimized reference data collection and management module is used to sort out the distribution of numerical differences in the cleaning status index between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas where the cleaning efficiency is lower than the average, and identify and extract data of the programs executed by the control terminal when generating cleaning operation instructions and cleaning operation paths that can assist in optimizing the unmanned car washing equipment.

[0035] Preferably: the target cleaning monitoring area extraction module includes: a stain area frame selection management unit and a target cleaning monitoring area extraction unit;

[0036] A stain area selection management unit is used to create a unit selection area through a selection tool, and to select the stain area on the stain area distribution image with the unit selection area;

[0037] The target cleaning monitoring area extraction unit is used to obtain the best frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the best frame selection scheme.

[0038] Preferably: the stain distribution change monitoring and management module includes a stain distribution change situation combing unit and a cleaning operation instruction sequence extraction unit;

[0039] The stain distribution change combing unit is used to comb each historical cleaning record of the unmanned car washing equipment and comb the stain distribution change presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing equipment executing the corresponding cleaning operation instruction;

[0040] The cleaning operation instruction sequence extraction unit is used to identify and extract the cleaning operation instruction sequence that each target cleaning monitoring area has gone through in the process of meeting the cleaning standard.

[0041] Compared with the prior art, the beneficial effects of the present invention are as follows: by selecting the stain area distribution image of the stain area on the outer surface of the vehicle body to be cleaned, the outer surface of the vehicle body to be cleaned is divided into several target cleaning monitoring areas, and then by analyzing the stain distribution changes in the target cleaning monitoring areas as the unmanned car washing equipment continuously executes cleaning operation instructions on the vehicle, the cleaning state index of the unmanned car washing equipment in the target cleaning area is evaluated and calculated, the characteristic cleaning monitoring areas with cleaning efficiency lower than the average are captured, and data that can assist in optimizing the program executed by the control terminal of the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths are collected, so as to realize the continuous improvement of the adaptability and automation level of the unmanned car washing equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 A schematic flow chart of an intelligent control method for unmanned car washing equipment according to the present invention;

[0043] Figure 2 The present invention is a schematic structural diagram of an intelligent control system applied to unmanned car washing equipment. DETAILED DESCRIPTION

[0044] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in the field without making any creative work shall fall within the scope of protection of the present invention.

[0045] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution: an intelligent control method applied to unmanned car washing equipment, the method comprising:

[0046] Step S1: Whenever a vehicle to be washed enters the active car washing area, the unmanned car washing equipment is triggered to obtain the body model and size information of the vehicle to be washed, as well as the stain area distribution image on the outer surface of the vehicle to be washed based on laser scanning technology;

[0047] Step S2: creating a unit frame selection area by using a frame selection tool, frame selection the stain area on the stain area distribution image by using the unit frame selection area, obtaining an optimal frame selection scheme for the stain area distribution image, and extracting a target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the optimal frame selection scheme;

[0048] Step S3: for each historical cleaning record of the unmanned car washing device, sort out the stain distribution changes presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing device executing the corresponding cleaning operation instructions; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard;

[0049] Step S4: sorting out the cleaning consumption of each target cleaning monitoring area by the unmanned car washing equipment in each historical cleaning record, and calculating the corresponding cleaning status index for each target cleaning monitoring area;

[0050] Step S5: Sort out the distribution of numerical differences in the cleaning status index between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas with cleaning efficiency lower than the average, and identify and extract data of the program executed by the control terminal that can assist in optimizing the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths.

[0051] Wherein, step S2 comprises:

[0052] Step S2-1: when selecting a stain region on the stain region distribution image using a unit selection area, the constraint conditions to be satisfied include: within the unit image selected by each unit selection area from the stain region distribution image, the total area of ​​the distributed stain region is greater than or equal to the area threshold; based on the unit image selected by each unit selection area, the vehicle body structure information can be identified, and the number of vehicle body structure that can be identified based on one unit image is 1;

[0053] For example, from one unit image, it can be identified that the stain distribution image corresponding to the car window position is present, and from another unit image, it can be identified that the stain distribution image corresponding to the car door handle position is present; but it is not possible to identify both the car window and the car door handle from one unit image. The purpose of doing so is mainly to split the car body surface as finely as possible, so as to better distinguish the influence of the stain distribution situation presented at different car body position structures on the unmanned car washing equipment during the vehicle cleaning process;

[0054] Step S2-2: a selection scheme that can select all the stain areas distributed in the stain area distribution image in a manner that meets the constraint conditions and requires the least number of unit selection areas to be created is set as the best scheme; the vehicle body surface area corresponding to each unit selection area in the best scheme is marked and set as the target cleaning monitoring area;

[0055] For example, there are three frame selection schemes to ensure that all stain areas in the stain area distribution image on the outer surface of a vehicle to be cleaned are completely covered, and that the total area of ​​the corresponding covered stain areas in each unit image obtained after all the stain areas are completely covered is greater than or equal to the area threshold, and at the same time, the structural information of the body part where the corresponding stain area is distributed can be identified from each unit image. In the first frame selection scheme, 75 unit frame selection areas need to be created, in the second frame selection scheme, 70 unit frame selection areas need to be created, and in the third frame selection scheme, 68 unit frame selection areas need to be created. In summary, the third frame selection scheme is the best scheme.

[0056] Wherein, step S3 comprises:

[0057] Step S3-1: In each historical cleaning record of the unmanned car washing device, the target cleaning monitoring area locked for the corresponding vehicle to be washed entering the active car washing area is extracted, and the stain degree value obtained by the unmanned car washing device for evaluating each target cleaning monitoring area before the corresponding vehicle to be washed is fully automatically cleaned is set as the first target value of each target cleaning monitoring area;

[0058] Step S3-2: In each historical cleaning record of the unmanned car washing device, for each corresponding target cleaning monitoring area, extract the sequence of the stain degree value change presented by the unmanned car washing device according to the order in which the corresponding cleaning operation instructions are executed, and evaluate the corresponding target cleaning monitoring area;

[0059] Step S3-3: The first stain level value that meets the cleaning standard in the stain level value change sequence is set as the second target value of each corresponding target cleaning monitoring area; for each corresponding target cleaning monitoring area, extract the cleaning operation instruction sequence executed and completed by the unmanned car washing equipment in the process of the corresponding stain level value changing from the corresponding first target value to the corresponding second target value.

[0060] Wherein, step S4 comprises:

[0061] Step S4-1: Assume that in a certain historical cleaning record, in the process of capturing the corresponding stain level value of a certain target cleaning monitoring area A changing from the corresponding first target value a1 to the corresponding second target value a2, the cleaning operation instruction sequence executed by the unmanned car washing device is P;

[0062] Step S4-2: obtaining a total number Y of cleaning operation instructions included in the cleaning operation instruction sequence P, and accumulating a total number of types of cleaning operation instructions included in the Y cleaning operation instructions as N;

[0063] Step S4-3: Collect the total time T taken by the unmanned car wash equipment to complete the cleaning operation instruction sequence P, set the unit reference time Tr; calculate the cleaning state index β of a target cleaning monitoring area A = |a1-a2| / [N×Y×(T / Tr)].

[0064] Wherein, step S5 comprises:

[0065] Step S5-1: obtaining the body model and size information of the corresponding vehicle to be cleaned in each historical cleaning record, and setting the body model and size information as the first attribute data of each target cleaning monitoring area extracted from each historical cleaning record; extracting the corresponding body part structure that can be identified based on each target cleaning monitoring area, and setting the body part structure as the second attribute data of each target cleaning monitoring area;

[0066] Step S5-2: traverse the first attribute data and the second attribute data of each target cleaning monitoring area extracted from all historical cleaning records; respectively collect all target cleaning monitoring areas whose corresponding first attribute data satisfy a similarity greater than or equal to a similarity threshold and whose corresponding second attribute data are the same, to obtain a plurality of target cleaning monitoring area sets;

[0067] Step S5-3: Calculate the average cleaning state index of all target cleaning monitoring areas included in each target cleaning monitoring area set; capture the target cleaning monitoring area whose corresponding cleaning state index is greater than the corresponding average cleaning state index in each target cleaning monitoring area set, and set it as a characteristic cleaning monitoring area;

[0068] Step S5-4: for each characteristic cleaning monitoring area, before fully automatic cleaning of the corresponding vehicle to be cleaned, the unit image corresponding to each characteristic cleaning monitoring area is captured from the corresponding stain area distribution image of the vehicle to be cleaned based on the unit frame selection area from the corresponding stain area distribution image of the vehicle to be cleaned, and the unit image and the position characteristic distribution information between the corresponding stain area and the corresponding vehicle body part structure obtained from the unit image recognition are stored, and the position characteristic distribution information includes the position orientation distribution between each stain area and the corresponding vehicle body part structure;

[0069] Step S5-5: regularly transmit all stored unit images and position feature distribution information to the management terminal, and assist the management terminal in optimizing the execution program of the control terminal of the unmanned car washing equipment; the execution program refers to the program executed by the control terminal when generating corresponding cleaning operation instructions and cleaning operation paths according to the body model, size information and stain distribution of the vehicle to be washed entering the car washing area.

[0070] In order to better implement the above method, an intelligent control system is also proposed, the system includes:

[0071] The vehicle data acquisition module is used to collect the body model and size information of the vehicle to be cleaned obtained by the unmanned car washing equipment based on laser scanning technology, as well as the stain area distribution image on the outer surface of the vehicle to be cleaned;

[0072] A target cleaning monitoring area extraction module is used to create a unit frame selection area through a frame selection tool, frame the stain area on the stain area distribution image with the unit frame selection area, obtain the best frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the best frame selection scheme;

[0073] The stain distribution change monitoring and management module is used to sort out the stain distribution changes presented by each target cleaning monitoring area of ​​the corresponding vehicle to be cleaned during the process of the unmanned car washing equipment executing the corresponding cleaning operation instructions for each historical cleaning record of the unmanned car washing equipment; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard;

[0074] The cleaning status index calculation module is used to sort out the cleaning consumption of the unmanned car washing equipment for each corresponding target cleaning monitoring area in each historical cleaning record, and calculate the corresponding cleaning status index for each corresponding target cleaning monitoring area;

[0075] The optimized reference data collection and management module is used to sort out the distribution of numerical differences in the cleaning status index between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas where the cleaning efficiency is lower than the average, and extract data that can assist in optimizing the programs executed by the control terminal of the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths.

[0076] Among them, the target cleaning monitoring area extraction module includes: a stain area selection management unit and a target cleaning monitoring area extraction unit;

[0077] A stain area selection management unit is used to create a unit selection area through a selection tool, and to select the stain area on the stain area distribution image with the unit selection area;

[0078] The target cleaning monitoring area extraction unit is used to obtain the best frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the best frame selection scheme.

[0079] Among them, the stain distribution change monitoring and management module includes a stain distribution change situation combing unit and a cleaning operation instruction sequence extraction unit;

[0080] The stain distribution change combing unit is used to comb each historical cleaning record of the unmanned car washing equipment and comb the stain distribution change presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing equipment executing the corresponding cleaning operation instruction;

[0081] The cleaning operation instruction sequence extraction unit is used to identify and extract the cleaning operation instruction sequence that each target cleaning monitoring area has gone through in the process of meeting the cleaning standard.

[0082] 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. An intelligent control method for unmanned car washing equipment, characterized in that: The method comprises: Step S1: Whenever a vehicle to be washed enters the active car washing area, the unmanned car washing equipment is triggered to obtain the body model and size information of the vehicle to be washed, as well as the stain area distribution image on the outer surface of the vehicle to be washed based on laser scanning technology; Step S2: creating a unit frame selection area by using a frame selection tool, frame selection the stain area on the stain area distribution image with the unit frame selection area, obtaining an optimal frame selection scheme for the stain area distribution image, and extracting a target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the optimal frame selection scheme; Step S3: for each historical cleaning record of the unmanned car washing device, sort out the stain distribution changes presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing device executing the corresponding cleaning operation instructions; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard; Step S4: sorting out the cleaning consumption of each target cleaning monitoring area by the unmanned car washing equipment in each historical cleaning record, and calculating the corresponding cleaning status index for each target cleaning monitoring area; Step S5: Sort out the distribution of numerical differences in the cleaning status index between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas with cleaning efficiency lower than the average, and identify and extract data of the program executed by the control terminal that can assist in optimizing the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths.

2. The intelligent control method for unmanned car washing equipment according to claim 1 is characterized in that: The step S2 comprises: Step S2-1: Constraints that need to be satisfied when the stain area is framed on the stain area distribution image by the unit frame selection area include: in the unit image framed by each unit frame selection area from the stain area distribution image, the total area of ​​the distributed stain area is greater than or equal to the area threshold; based on the unit image framed by each unit frame selection area, the vehicle body part structure information can be identified, and the number of vehicle body part structures that can be identified based on one unit image is 1; wherein the vehicle body part structure includes: windows, rearview mirrors, vehicle body exteriors, bumpers, fenders, ABCD pillars, side door frame components, engine hoods, trunks, roof covers, doors, wheel covers, door handles, air intake grilles, etc.; Step S2-2: Set the selection scheme that can select all the stain areas distributed in the stain area distribution image in a manner that meets the constraint conditions and requires the least number of unit selection areas to be created as the best scheme; mark the position of the vehicle body surface area corresponding to each unit selection area in the best scheme and set it as the target cleaning monitoring area.

3. The intelligent control method for unmanned car washing equipment according to claim 2 is characterized in that: The step S3 comprises: Step S3-1: In each historical cleaning record of the unmanned car washing device, a target cleaning monitoring area locked for a corresponding vehicle to be washed that enters the active car washing area is extracted, and a stain degree value obtained by the unmanned car washing device for evaluating each target cleaning monitoring area before fully-automatically cleaning the corresponding vehicle to be washed is set as a first target value of each target cleaning monitoring area; Step S3-2: In each historical cleaning record of the unmanned car washing device, for each corresponding target cleaning monitoring area, extract the sequence of the stain degree value change presented by the unmanned car washing device according to the order in which the corresponding cleaning operation instructions are executed, and evaluate the corresponding target cleaning monitoring area; Step S3-3: The first stain level value that meets the cleaning standard in the stain level value change sequence is set as the second target value of each corresponding target cleaning monitoring area; for each corresponding target cleaning monitoring area, extract the cleaning operation instruction sequence executed and completed by the unmanned car washing equipment in the process of the corresponding stain level value changing from the corresponding first target value to the corresponding second target value. The unmanned car wash equipment described in this application, that is, the fully automatic car wash equipment, is usually equipped with touch sensors and pressure sensors, which can detect the type and density of stains on the car body. Among them, the touch sensor can sense the particles on the surface of the car body, such as dust and dirt, and the pressure sensor can detect the hardness and size of the stains, such as resin, oil stains and bird droppings. Usually, by analyzing the data of the sensor, the car wash equipment can preliminarily judge the degree of stains on the car body. Secondly, it is also equipped with a camera and a computer vision system, which can perform high-precision image recognition and analysis on the car body. Among them, the camera can capture the details of the stains on the car body and transmit the image to the computer vision system for processing. The computer vision system can use image processing algorithms to identify different areas and colors in the image, and further analyze and obtain the stain degree value.

4. The intelligent control method for unmanned car washing equipment according to claim 3 is characterized in that: The step S4 comprises: Step S4-1: Assume that in a certain historical cleaning record, in the process of capturing the corresponding stain level value of a certain target cleaning monitoring area A changing from the corresponding first target value a1 to the corresponding second target value a2, the cleaning operation instruction sequence executed by the unmanned car washing device is P; Step S4-2: obtaining a total number Y of cleaning operation instructions included in the cleaning operation instruction sequence P, and accumulating a total number of types of cleaning operation instructions included in the Y cleaning operation instructions as N; Step S4-3: Collect the total time T taken by the unmanned car wash equipment to complete the cleaning operation instruction sequence P, set the unit reference time Tr; calculate the cleaning state index β of a target cleaning monitoring area A = |a1-a2| / [N×Y×(T / Tr)].

5. The intelligent control method for unmanned car washing equipment according to claim 4 is characterized in that: The step S5 comprises: Step S5-1: obtaining the body model and size information of the corresponding vehicle to be cleaned in each historical cleaning record, and setting the body model and size information as the first attribute data of each target cleaning monitoring area extracted from each historical cleaning record; extracting the corresponding body part structure that can be identified based on each target cleaning monitoring area, and setting the body part structure as the second attribute data of each target cleaning monitoring area; Step S5-2: traverse the first attribute data and the second attribute data of each target cleaning monitoring area extracted from all historical cleaning records; respectively collect all target cleaning monitoring areas whose corresponding first attribute data satisfy a similarity greater than or equal to a similarity threshold and whose corresponding second attribute data are the same, to obtain a plurality of target cleaning monitoring area sets; Step S5-3: Calculate the average cleaning state index of all target cleaning monitoring areas included in each target cleaning monitoring area set; capture the target cleaning monitoring area whose corresponding cleaning state index is greater than the corresponding average cleaning state index in each target cleaning monitoring area set, and set it as a characteristic cleaning monitoring area; Step S5-4: for each characteristic cleaning monitoring area, before fully automatic cleaning of the corresponding vehicle to be cleaned, a unit image corresponding to each characteristic cleaning monitoring area is captured from the corresponding stain area distribution image of the vehicle to be cleaned based on the unit frame selection area from the corresponding stain area distribution image of the vehicle to be cleaned, and the unit image and the position characteristic distribution information between the corresponding stain area and the corresponding vehicle body part structure obtained by identifying the unit image are stored, wherein the position characteristic distribution information includes the position orientation distribution between each stain area and the corresponding vehicle body part structure; Step S5-5: regularly transmit all stored unit images and position feature distribution information to the management terminal, and assist the management terminal in optimizing the execution program of the control terminal of the unmanned car washing equipment; the execution program refers to the program executed by the control terminal when generating corresponding cleaning operation instructions and cleaning operation paths according to the body model, size information and stain distribution of the vehicle to be washed entering the car washing area.

6. An intelligent control system for executing an intelligent control method for unmanned car washing equipment as claimed in any one of claims 1 to 5, characterized in that: The system comprises: The vehicle data acquisition module is used to collect the body model and size information of the vehicle to be cleaned obtained by the unmanned car washing equipment based on laser scanning technology, as well as the stain area distribution image on the outer surface of the vehicle body to be cleaned; The target cleaning monitoring area extraction module is used to create a unit frame selection area through a frame selection tool, use the unit frame selection area to frame the stain area distribution image, obtain the best frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding vehicle body outer surface to be cleaned based on the best frame selection scheme; The stain distribution change monitoring and management module is used to sort out the stain distribution change situation presented by each target cleaning monitoring area of ​​the corresponding vehicle to be cleaned during the process of the unmanned car washing equipment executing the corresponding cleaning operation instruction for each historical cleaning record of the unmanned car washing equipment; identify and extract the cleaning operation instruction sequence experienced by each target cleaning monitoring area in the process of meeting the cleaning standard; The cleaning state index calculation module is used to sort out the cleaning consumption of the unmanned car washing equipment on each corresponding target cleaning monitoring area in each historical cleaning record, and calculate the corresponding cleaning state index for each corresponding target cleaning monitoring area; The optimized reference data collection and management module is used to sort out the distribution of numerical differences in the cleaning status index presented between different target cleaning monitoring areas with the same attribute data, capture characteristic cleaning monitoring areas with cleaning efficiency lower than the average value, and extract data that can assist in optimizing the program executed by the control terminal of the unmanned car washing equipment when generating cleaning operation instructions and cleaning operation paths.

7. An intelligent control system according to claim 6, characterized in that: The target cleaning monitoring area extraction module includes: a stain area frame selection management unit and a target cleaning monitoring area extraction unit; The stain area selection management unit is used to create a unit selection area through a selection tool, and use the unit selection area to select the stain area on the stain area distribution image; The target cleaning monitoring area extraction unit is used to obtain an optimal frame selection scheme for the stain area distribution image, and extract the target cleaning monitoring area for the corresponding outer surface of the vehicle body to be cleaned based on the optimal frame selection scheme.

8. An intelligent control system according to claim 6, characterized in that: The stain distribution change monitoring and management module includes a stain distribution change situation combing unit and a cleaning operation instruction sequence extraction unit; The stain distribution change combing unit is used to comb each historical cleaning record of the unmanned car washing device and comb the stain distribution change presented in each target cleaning monitoring area of ​​the corresponding vehicle to be washed during the process of the unmanned car washing device executing the corresponding cleaning operation instruction; The cleaning operation instruction sequence extraction unit is used to identify and extract the cleaning operation instruction sequence that each target cleaning monitoring area has gone through in the process of meeting the cleaning standard.

Images