Data processing method and system for AI model data

Abstract

The invention discloses a data processing method and system for AI model data, and belongs to the technical field of artificial intelligence data processing, and the method comprises the steps: S10, obtaining environment data of a target region, the environment data of the target region comprises dirt coordinate data and fixed object coordinate data of a preset room, analyzing the distance between each two adjacent fixed objects according to the fixed object coordinate data, marking each fixed object based on the fixed object coordinate data, and generating a database based on the collected fixed object coordinate data; according to the invention, the environment data of the target area is acquired, the dirt coordinate data and the fixed object coordinate data are acquired from the environment data of the target area, the distance between the dirt and the fixed object can be analyzed and calculated, and the cleaning route for the dirt is generated according to the calculation result. Meanwhile, the obstacle possibly formed by the moving object in the sweeping route is analyzed, an early warning is given out according to the obstacle, the sweeping route is corrected, and therefore the obstacle is accurately avoided to complete sweeping.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence data processing technology, and in particular to a data processing method and system for AI model data. Background Art

[0002] AI technology is globally recognized as a cutting-edge field and a technological frontier. With the continuous development and popularization of AI technology, it has spread throughout people's lives. In the past, smart homes were often questioned as pseudo-intelligent due to limited technical applications. In recent years, AI technology has been increasingly used in the fields of home appliances and the Internet of Vehicles. Through deep integration with AI technology, the intelligence level and efficiency of equipment have been significantly improved. Among them, home appliance companies have introduced AI technology and built AI models for training and deep learning by collecting user usage data to improve the intelligence level of home appliances, achieving natural human-computer dialogue in home scenarios and deep and precise control of home appliances and scenarios. At the same time, with the help of AI technology, home appliances can better understand user instructions and needs to improve services.

[0003] According to statistics released by IDC, the global smart home market size has climbed to US$151.4 billion in 2022, an increase of 16.2% over the previous year. It is predicted that by 2025, the overall penetration rate of AI in the smart home field will be close to 50%. It is predicted that by 2025, the global smart home market will reach US$236.7 billion, with a compound annual growth rate of approximately 11.8%.

[0004] As the world's leading smart home market, China is expected to expand its market size to US$87.8 billion by 2025, with a compound annual growth rate of nearly 15%. This significant market expansion is due to breakthroughs in AI model technology in natural language processing, image recognition, and deep semantic understanding, which enable smart devices to more accurately capture user needs and provide personalized services. Empowered by AI models, home appliances will be able to go beyond traditional functions and expand their value to a wider range of fields.

[0005] According to domestic media reports, the internationally renowned American company Apple will launch AI-based smart home devices as early as March next year. The device will have FaceTime, Siri, and Apple Intelligence functions, and will be connected to cameras, doorbells, and run a series of Apple apps. The company hopes to compete with Amazon and Google in the smart home field.

[0006] It can be seen that household appliances provide users with a convenient way to interact with AI models, thereby optimizing user experience and promoting the widespread application of technology. At present, many home appliance companies on the market have also begun to test smart home robots with richer functions. Among them, as the pace of life in modern society accelerates, sweepers, as a new cleaning tool, have entered thousands of households, and the application of AI technology has improved the performance of sweepers. However, the home scenarios in which sweepers are used are different, resulting in the fact that sweepers supported by modern AI technology still have deficiencies in data processing in collecting image data information and adaptively adjusting cleaning strategies according to real-time conditions. At the same time, in terms of image data information, they are mostly limited to processing image data information of fixed targets, and there are deficiencies in processing mobile image data information. In this case, it is difficult to accurately avoid obstacles in the process of cleaning the dirt on the ground, resulting in obstacles to cleaning the dirt. Summary of the invention

[0007] In view of the above-mentioned problems existing in the existing field of artificial intelligence data processing technology, the present invention is proposed.

[0008] Therefore, one of the objects of the present invention is to provide a data processing method and system for AI model data, which can analyze and calculate the distance between dirt and fixed objects by acquiring environmental data of the target area and collecting dirt coordinate data and fixed object coordinate data in the environmental data of the target area, and generate a cleaning route for the dirt based on the calculation results. At the same time, it analyzes the obstacles that may be caused by moving objects in the cleaning route, and issues an early warning accordingly, and corrects the cleaning route to accurately avoid obstacles and complete cleaning.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] In one aspect, the present invention provides a data processing method for AI model data, comprising the following steps:

[0011] Step S10: Acquire environmental data of the target area, wherein the environmental data of the target area includes dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between adjacent fixed objects according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data;

[0012] Step S20: Divide the interval into ▽1, ▽2, ..., ▽ n , n represents the nth interval, and the movement trajectory data of each moving object is collected in the divided interval, and the movement trajectory data is intercepted in a given time period in the collected movement trajectory data to analyze the movement trend of the moving object;

[0013] Step S30: presetting a reference value based on the last piece of the movement trajectory data in the collected movement trajectory data, and calculating the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value;

[0014] Step S40: Calculate the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object.

[0015] As a preferred embodiment of the present invention, in step S20, the given time period includes intercepting the movement trajectory data every 1 to 1.5 seconds, calculating the distance between the moving object and the nearest fixed object when stationary based on the intercepted movement trajectory data, and dividing the distance into large distance, medium distance and small distance in a step-by-step manner, wherein the large distance is ≥50 cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise, it is not generated.

[0016] As a preferred solution of the present invention, the first position coordinate data of the sweeper is obtained in the generated cleaning route, the first position coordinate data is the position coordinate data of the sweeper before the cleaning route is generated, and the distance from the sweeper to the dirt is calculated based on the position coordinate data, and the distance is divided into 10 path points G in equal proportion. σ , generate a set of path point data When the sweeper moves toward the dirt, each time it passes through the path point G σ , the distance between the moving object and the dirt is calculated once, and the position change of the moving object when the sweeper arrives at the dirt is analyzed.

[0017] As a preferred solution of the present invention, when at a certain path point G σ When the position of the moving object changes, the path point G σ Based on the distance change between the moving object and the dirt, the path point G σ Marked as control point, and 10 path points G σ The last three points in the path are used to determine the distance between the moving object and the dirt. When the sweeper reaches the path point When the distance between the moving object and the dirt is obtained to be reduced, it is determined that there is an obstacle to the cleaning of the dirt, and the sweeping machine issues an obstacle warning and stops moving; otherwise, no determination is made.

[0018] As a preferred solution of the present invention, when an obstacle warning is issued, the position coordinate data of the moving object is collected, the position coordinate data is marked as a reference coordinate, and the distance change between the reference coordinate and the dirt is calculated, wherein the calculation method includes calculating at a frequency of every 0.5 seconds, and the duration is 3 seconds. When the distance between the moving object and the reference coordinate is ≥50cm within the time period, it is determined that the obstacle to cleaning the dirt no longer exists, otherwise, it is not determined.

[0019] As a preferred solution of the present invention, it also includes: when an obstacle warning is issued, 2 to 3 virtual position coordinates are generated with the reference coordinate as the center, and the distance change between the reference coordinate and the virtual position coordinate is collected. Among the 2 to 3 virtual position coordinates, when the distance between the reference coordinate and any one of the virtual position coordinates shows a decreasing trend, the sweeper moves from the reference coordinate to the opposite direction of the movement of the virtual position coordinate, otherwise, it does not move forward.

[0020] As a preferred scheme of the present invention, when it is determined that there is no obstacle to cleaning the dirt, the position coordinate data of the sweeping machine is updated in real time until the sweeping machine finishes cleaning the corresponding dirt, and new first position coordinate data about the sweeping machine is generated based on the position coordinate data of the corresponding dirt, and at the same time, based on the first position coordinate data, the coordinate data of the 3 to 5 dirt closest to it are collected, and the coordinate data are sorted according to the distance from the first position coordinate data, wherein the dirt closest to it is marked as the first cleaning target, and a cleaning route for the first cleaning target is generated.

[0021] As a preferred solution of the present invention, after the cleaning route is generated, the coordinates of fixed objects and / or moving objects within 50 cm of the first cleaning target are collected. When the coordinates of fixed objects and / or moving objects are not collected within the distance, the cleaning task is performed on the dirt according to the generated cleaning route, otherwise it is not performed.

[0022] In another aspect, the present invention provides a system for a data processing method for AI model data, comprising:

[0023] An environmental data acquisition module is used to acquire environmental data of a target area, wherein the environmental data of the target area includes dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between each adjacent fixed object according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data;

[0024] A data partitioning module, used to divide the interval into ▽1, ▽2, ..., ▽ n , n represents the nth interval, and the movement trajectory data of each moving object is collected in the divided interval, and the movement trajectory data is intercepted in a given time period in the collected movement trajectory data to analyze the movement trend of the moving object;

[0025] A movement data calculation module, configured to preset a reference value based on the last movement trajectory data in the collected movement trajectory data, and calculate the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value;

[0026] A distance data calculation module, used to calculate the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object;

[0027] The route generation module is used to calculate the distance between the moving object and the nearest fixed object when it is stationary based on the intercepted moving trajectory data, and divide the distance into large distance, medium distance and small distance in a step-by-step manner, wherein the large distance is ≥50cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise it is not generated.

[0028] Beneficial effects:

[0029] The present invention acquires environmental data of the target area and collects coordinate data of dirt and coordinate data of fixed objects in the environmental data of the target area, thereby analyzing and calculating the distance between dirt and fixed objects, and generating a cleaning route for the dirt according to the calculation result. At the same time, when the sweeper moves to the dirt according to the generated cleaning route, it can detect and collect moving objects close to the cleaning route in this process, and calculate the moving trajectory of the moving object to analyze the possible obstacles formed by the moving object in the cleaning route, and issue an early warning accordingly to let people in the regional environment know the obstacles formed, and after the early warning is issued, the moving trend of the moving object can be obtained by setting the virtual position coordinates, and then the cleaning route can be corrected in time according to its moving trend to accurately avoid obstacles and complete cleaning. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0031] Figure 1 A schematic diagram of a modular structure of a data processing system for AI model data according to an embodiment of the present invention;

[0032] Figure 2 A schematic diagram of a method flow of an embodiment of the present invention;

[0033] Figure 3 It is a schematic diagram of the process structure of an embodiment of the present invention;

[0034] Numbers in the figure: 110 - environmental data acquisition module; 120 - data division module; 130 - movement data calculation module; 140 - distance data calculation module; 150 - route generation module. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0036] Due to the different home usage scenarios, sweeping robots supported by modern AI technology still have deficiencies in data processing in collecting image data information and adaptively adjusting cleaning strategies according to real-time situations. At the same time, in terms of image data information, they are mostly limited to processing image data information of fixed targets, while there are deficiencies in the processing of moving image data information. In this case, it is difficult to accurately avoid obstacles during the cleaning of dirt on the ground, which leads to obstacles in cleaning the dirt.

[0037] Based on this, the present invention proposes a data processing method and system for AI model data, which can analyze and calculate the distance between the dirt and the fixed object by acquiring the environmental data of the target area and collecting the coordinate data of the dirt and the coordinate data of the fixed object in the environmental data of the target area, and generate a cleaning route for the dirt based on the calculation results. At the same time, it analyzes the obstacles that may be caused by moving objects in the cleaning route, and issues an early warning accordingly, and corrects the cleaning route to accurately avoid obstacles and complete cleaning.

[0038] The present invention is further described in detail below through embodiments and in conjunction with the accompanying drawings.

[0039] Reference Figures 1 to 3 , is an embodiment of the present invention, which provides a data processing method for AI model data, comprising the following steps:

[0040] Step S10: Acquire environmental data of the target area, the environmental data of the target area including dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between adjacent fixed objects according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data;

[0041] Step S20: Divide the interval into ▽1, ▽2, ..., ▽ n , n represents the nth interval, and the moving trajectory data of each moving object is collected in the divided interval, and the moving trajectory data is intercepted at a given time period in the collected moving trajectory data to analyze the moving trend of the moving object;

[0042] It should be noted that in this embodiment, the given time period includes intercepting the moving trajectory data every 1 to 1.5 seconds, calculating the distance between the moving object and the nearest fixed object when it is stationary based on the intercepted moving trajectory data, and dividing the distance into large distance, medium distance and small distance in a step-by-step manner, wherein the large distance is ≥50 cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise, it is not generated;

[0043] According to the data of sweeping robots sold on the market, the diameter of sweeping robots is generally between 30cm and 40cm. Specifically, the diameter of a small sweeping robot is about 30cm, the diameter of a medium-sized sweeping robot is about 35cm, and the diameter of a large sweeping robot can reach 40cm. Therefore, this application uses the large spacing as a reference basis to meet the actual situation;

[0044] In this embodiment, the first position coordinate data of the sweeping machine is obtained in the generated cleaning route. The first position coordinate data is the position coordinate data of the sweeping machine before the cleaning route is generated, and the distance from the sweeping machine to the dirt is calculated based on the position coordinate data, and the distance is divided into 10 path points G in equal proportion. σ , generate a set of path point data When the sweeper moves toward the dirt, each time it passes a path point G σ , then calculate the distance between the moving object and the dirt, and analyze the position change of the moving object when the sweeper reaches the dirt;

[0045] At the same time, this embodiment further, when a certain path point G σ When the position of the moving object changes, the path point G σ Based on the distance change between the moving object and the dirt, the path point G σ Marked as control point, and with 10 path points G σThe last three points are used as decision points to determine the distance between the moving object and the dirt. When the sweeper reaches the path point G8 σ When the distance between the moving object and the dirt is reduced, it is determined that there is an obstacle to the cleaning of the dirt, and the sweeping machine issues an obstacle warning and stops moving. Otherwise, no determination is made;

[0046] On the basis of the above, the present embodiment also includes generating 2 to 3 virtual position coordinates with the reference coordinate as the center when an obstacle warning is issued, collecting the distance change between the reference coordinate and the virtual position coordinate, and when the distance between the reference coordinate and any one of the virtual position coordinates shows a decreasing trend among the 2 to 3 virtual position coordinates, the sweeper moves from the reference coordinate to the opposite direction of the movement of the virtual position coordinate, otherwise, it does not move;

[0047] It should be emphasized in this embodiment that, when it is determined that there is no obstacle to cleaning the dirt, the position coordinate data of the sweeping machine is updated in real time until the sweeping machine has finished cleaning the corresponding dirt, and new first position coordinate data about the sweeping machine is generated based on the position coordinate data of the corresponding dirt, and at the same time, the coordinate data of the 3 to 5 dirt closest to it is collected based on the first position coordinate data, and the coordinate data are sorted based on the distance from the first position coordinate data, wherein the dirt closest to it is marked as the first cleaning target, and a cleaning route for the first cleaning target is generated;

[0048] After the cleaning route is generated, the coordinates of fixed objects and / or moving objects within 50 cm of the first cleaning target are collected. If the coordinates of fixed objects and / or moving objects are not collected within the distance, the cleaning task is performed on the dirt according to the generated cleaning route, otherwise, it is not performed;

[0049] Step S30: presetting a reference value based on the last movement trajectory data in the collected movement trajectory data, and calculating the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value;

[0050] Step S40: Calculate the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object.

[0051] Based on the above, the present application can analyze and calculate the distance between the dirt and the fixed object by acquiring the environmental data of the target area and collecting the dirt coordinate data and the fixed object coordinate data in the environmental data of the target area, and generate a cleaning route for the dirt based on the calculation results. At the same time, it analyzes the obstacles that may be caused by moving objects in the cleaning route, and issues a warning accordingly, and corrects the cleaning route to accurately avoid obstacles and complete cleaning.

[0052] In combination with the above-mentioned data processing method for AI model data, this embodiment also proposes a system applied to the method, as follows:

[0053] The environmental data acquisition module 110 is used to acquire environmental data of a target area, the environmental data of the target area including dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between each adjacent fixed object according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data;

[0054] The data partitioning module 120 is used to partition the interval into ▽1, ▽2, ..., ▽ n , n represents the nth interval, and the moving trajectory data of each moving object is collected in the divided interval, and the moving trajectory data is intercepted at a given time period in the collected moving trajectory data to analyze the moving trend of the moving object;

[0055] The movement data calculation module 130 is used to preset a reference value based on the last movement trajectory data in the collected movement trajectory data, and calculate the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value;

[0056] The distance data calculation module 140 is used to calculate the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object;

[0057] The route generation module 150 is used to calculate the distance between the moving object and the nearest fixed object when it is stationary based on the intercepted moving trajectory data, and divide the distance into large distance, medium distance and small distance in a step-by-step manner, where the large distance is ≥50cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise it is not generated.

[0058] In summary, the present invention acquires environmental data of the target area, and collects coordinate data of dirt and coordinate data of fixed objects in the environmental data of the target area, so as to analyze and calculate the distance between dirt and fixed objects, and generate a cleaning route for the dirt according to the calculation results. At the same time, when the sweeper moves to the dirt according to the generated cleaning route, it can detect and collect moving objects close to the cleaning route in this process, and calculate the moving trajectory of the moving object to analyze the possible obstacles formed by the moving object in the cleaning route, and issue an early warning accordingly to let people in the regional environment know the obstacles formed, and after the early warning is issued, the moving trend of the moving object can be obtained by setting the virtual position coordinates, and then the cleaning route can be corrected in time according to its moving trend to accurately avoid obstacles and complete cleaning.

[0059] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A data processing method for AI model data, characterized in that: The following steps are involved: Step S10: Acquire environmental data of the target area, wherein the environmental data of the target area includes dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between adjacent fixed objects according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data; Step S20: Divide the interval into n represents the nth interval, and the movement trajectory data of each moving object is collected in the divided intervals, and the movement trajectory data is intercepted at a given time period in the collected movement trajectory data to analyze the movement trend of the moving object; Step S30: presetting a reference value based on the last piece of the movement trajectory data in the collected movement trajectory data, and calculating the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value; Step S40: Calculate the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance to the moving object.

2. A data processing method for AI model data according to claim 1, characterized in that: In the step S20, the given time period includes intercepting the movement trajectory data every 1 to 1.5 seconds, calculating the distance between the moving object and the nearest fixed object when it is stationary based on the intercepted movement trajectory data, and dividing the distance into large distance, medium distance and small distance in a step-by-step manner, wherein the large distance is ≥50 cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise, it is not generated.

3. A data processing method for AI model data according to claim 2, characterized in that: The first position coordinate data of the sweeper is obtained in the generated cleaning route, the first position coordinate data is the position coordinate data of the sweeper before the cleaning route is generated, and the distance from the sweeper to the dirt is calculated based on the position coordinate data, and the distance is divided into 10 path points G in equal proportion. σ , generate a set of path point data When the sweeper moves toward the dirt, each time it passes through the path point G σ , the distance between the moving object and the dirt is calculated once, and the position change of the moving object when the sweeper arrives at the dirt is analyzed.

4. A data processing method for AI model data according to claim 3, characterized in that: When at a certain path point G σ When the position of the moving object changes, the path point G σ Based on the distance change between the moving object and the dirt, the path point G σ Marked as control point, and the 10 path points G σ The last three points in the path are used to determine the distance between the moving object and the dirt. When the sweeper reaches the path point When the distance between the moving object and the dirt is obtained to be reduced, it is determined that there is an obstacle to the cleaning of the dirt, and the sweeping machine issues an obstacle warning and stops moving; otherwise, no determination is made.

5. A data processing method for AI model data according to claim 4, characterized in that: When an obstacle warning is issued, the position coordinate data of the moving object is collected, the position coordinate data is marked as a reference coordinate, and the change in distance between the reference coordinate and the dirt is calculated, wherein the calculation method includes calculating at a frequency of every 0.5 seconds, and the duration is 3 seconds. When the distance between the moving object and the reference coordinate is ≥50cm within the time period, it is determined that the obstacle to cleaning the dirt no longer exists, otherwise, it is not determined.

6. A data processing method for AI model data according to claim 5, characterized in that: It also includes generating 2 to 3 virtual position coordinates with the reference coordinate as the center when an obstacle warning is issued, collecting the distance change between the reference coordinate and the virtual position coordinate, and when the distance between the reference coordinate and any one of the virtual position coordinates shows a decreasing trend among the 2 to 3 virtual position coordinates, the sweeper moves from the reference coordinate to the opposite direction of the movement of the virtual position coordinate, otherwise, it does not move forward.

7. A data processing method for AI model data according to claim 5, characterized in that: When it is determined that there is no obstacle to cleaning the dirt, the position coordinate data of the sweeping machine is updated in real time until the sweeping machine completes cleaning the corresponding dirt, and new first position coordinate data about the sweeping machine is generated based on the position coordinate data of the corresponding dirt. At the same time, based on the first position coordinate data, the coordinate data of the 3 to 5 dirt closest to it are collected, and the coordinate data are sorted based on the distance from the first position coordinate data, wherein the dirt closest to it is marked as the first cleaning target, and a cleaning route for the first cleaning target is generated.

8. A data processing method for AI model data according to claim 7, characterized in that: After the cleaning route is generated, the coordinates of fixed objects and / or moving objects within 50 cm of the first cleaning target are collected. When the coordinates of fixed objects and / or moving objects are not collected within the distance, the cleaning task is performed on the dirt according to the generated cleaning route, otherwise it is not performed.

9. A system applied to a data processing method for AI model data as claimed in claim 1, characterized in that: include: An environmental data acquisition module is used to acquire environmental data of a target area, wherein the environmental data of the target area includes dirt coordinate data and fixed object coordinate data of a preset room, analyze the distance between each adjacent fixed object according to the fixed object coordinate data, mark each fixed object based on the fixed object coordinate data, and generate a database based on the collected fixed object coordinate data; A data partitioning module, used to divide the interval into ▽1, ▽2, ..., ▽ n , n represents the nth interval, and the movement trajectory data of each moving object is collected in the divided interval, and the movement trajectory data is intercepted at a given time period in the collected movement trajectory data to analyze the movement trend of the moving object; A movement data calculation module, configured to preset a reference value based on the last movement trajectory data in the collected movement trajectory data, and calculate the distance between the moving object and the center point of the interval in the corresponding interval based on the reference value; A distance data calculation module, used to calculate the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object according to the distance, and calculate the diameter difference with the sweeper according to the maximum distance between the dirt and the adjacent fixed object and the distance from the moving object; The route generation module is used to calculate the distance between the moving object and the nearest fixed object when it is stationary based on the intercepted moving trajectory data, and divide the distance into large distance, medium distance and small distance in a step-by-step manner, wherein the large distance is ≥50cm. When the calculated distance is a large distance, a cleaning route for the dirt is generated, otherwise it is not generated.

Images