A plant cultivation intelligent management method and system

Through the construction of robot group inspection and visual model, online remote management of large plantations is realized, the problem of low human inspection efficiency is solved, and management efficiency and convenience are improved.

CN118607898BActive Publication Date: 2024-11-22SHIJIAZHUANG HANXU AGRI TECH CO LTD
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202411087444.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-11-22
Estimated Expiration
2044-08-09

AI Technical Summary

Technical Problem

The management efficiency and convenience of large-scale plantations are limited by human inspections, resulting in high costs and low efficiency.

Method used

A robot group is used to conduct inspections, and a visual model is built based on the inspection screen and the park map to realize online remote management, and determine the knowledge base and array knowledge base through perspective to assist user management.

Benefits of technology

It reduces labor costs, improves management efficiency and convenience, and reduces the on-site inspection needs of managers.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118607898B_ABST
    Figure CN118607898B_ABST
Patent Text Reader

Abstract

The present invention provides a plant cultivation intelligent management method and system, wherein the method comprises: controlling a robot group to inspect the cultivated plants in a plantation; building a visualization model based on the inspection screen and position distribution of the robot group and the park map of the plantation; and assisting users to remotely manage the plantation online based on the visualization model. The plant cultivation intelligent management method and system of the present invention controls a robot group to inspect the cultivated plants in a plantation; building a visualization model based on the inspection screen and position distribution of the robot group and the park map of the plantation; and assisting users to remotely manage the plantation online based on the visualization model, without the need for management personnel to go to the plantation for inspection and manage on site, thereby reducing labor costs, improving the management efficiency of cultivated plants, and improving the convenience of management personnel to manage.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of plant cultivation management, and in particular to a plant cultivation intelligent management method and system. Background Art

[0002] At present, park-style plant cultivation is becoming increasingly popular. In order to ensure the healthy and stable growth of cultivated plants in plantations, it is necessary to manage the cultivated plants.

[0003] However, when managing cultivated plants, managers are often required to conduct inspections in the plantation. Whenever problems are found during the inspection, managers will manage them on site. However, the plantation area is often large, which results in high labor costs, reduces the efficiency of plant management, and reduces the convenience of management for managers.

[0004] Therefore, a solution is urgently needed. Summary of the invention

[0005] One of the purposes of the present invention is to provide an intelligent management method for plant cultivation, which controls a robot group to inspect the cultivated plants in a plantation, and builds a visualization model based on the inspection screen and location distribution of the robot group and the park map of the plantation to assist users in online remote management of the plantation based on the visualization model. There is no need for management personnel to go to the plantation for inspection and manage on-site, which reduces labor costs, improves the management efficiency of cultivated plants, and improves the convenience of management for management personnel.

[0006] An embodiment of the present invention provides a plant cultivation intelligent management method, comprising:

[0007] Control the robot group to inspect the cultivated plants in the plantation;

[0008] Build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation;

[0009] Assist users to remotely manage plantations online based on visualization models.

[0010] Preferably, the assisting user performs online remote management of the plantation based on the visualization model, comprising:

[0011] When the user views the visualization model, the knowledge base is determined based on the perspective, and the key viewing perspective is determined based on the incremental information generated by the visualization model in the most recent first time period;

[0012] If the user meets the guidance timing condition within the latest second time period, guiding the user to view the visualization model from the key viewing perspective; the second time period is shorter than the first time period;

[0013] and / or,

[0014] When the user inputs a trigger command for an abnormal plant event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the abnormal plant event;

[0015] and / or,

[0016] When the user inputs a management policy, the management policy is executed on the plantation.

[0017] Preferably, the determining of the knowledge base based on the perspective, according to the incremental information generated by the visualization model in the most recent first time period, to determine the key viewing perspective, includes:

[0018] Performing feature description on the incremental information to obtain a first description feature set;

[0019] Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base;

[0020] Determine knowledge based on perspective and determine multiple basic perspective requirements;

[0021] Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements;

[0022] Obtaining evaluation values ​​of respective candidate perspectives;

[0023] The candidate viewing angle with the maximum evaluation value is used as the key viewing angle;

[0024] The step of obtaining the evaluation value of each candidate viewing angle includes:

[0025] Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, based on the first quantization table, determining the evaluation value of the traversed candidate view according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view;

[0026] Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model.

[0027] Preferably, the guiding timing conditions include:

[0028] The frequency of the user's perspective change when viewing the visualization model is greater than or equal to the frequency threshold;

[0029] and / or,

[0030] The matching degree between the visual content position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model and the standard position relationship set is greater than or equal to the matching degree threshold;

[0031] and / or,

[0032] The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

[0033] Preferably, the control of the robot group's formation based on the formation knowledge base and according to abnormal plant events includes:

[0034] Describing the abnormal plant events by features to obtain a second description feature set;

[0035] Determining the formation knowledge corresponding to the second description feature set from the formation knowledge base;

[0036] Based on the array knowledge, determine the search rules and task formulation rules;

[0037] Based on the search rules, search for the basis for the formation task in the visual model;

[0038] Formulate the formation tasks based on the task formulation rules and the formation task formulation basis;

[0039] Control the robot group to perform formation tasks.

[0040] Preferably, the plant cultivation intelligent management method further comprises:

[0041] When the robot group performs the formation task, if at least two robots in the robot group meet the joint trigger condition, a display frame generation template is generated based on the display frame corresponding to the met joint trigger condition, and a display frame is generated according to the inspection pictures generated by the corresponding at least two robots in the third historical time period;

[0042] The display frame is set in an idle area of ​​the visible area of ​​the visualization model in the real-time perspective of the user currently viewing the visualization model;

[0043] The combined triggering conditions include:

[0044] The current inspection objects of each robot are the same;

[0045] and / or,

[0046] There is a standard task association relationship between the subtasks in the formation task performed by each robot;

[0047] and / or,

[0048] The current straight-line distance between each pair of robots is less than or equal to the distance threshold and the current inspection angle vector angle between each pair of robots falls within the standard angle range.

[0049] An embodiment of the present invention provides a plant cultivation intelligent management system, comprising:

[0050] A control module is used to control the robot group to inspect the cultivated plants in the plantation;

[0051] Build a module to build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation;

[0052] Auxiliary module, used to assist users in online remote management of plantations based on visualization models.

[0053] Preferably, the auxiliary module assists the user to perform online remote management of the plantation based on the visualization model, including:

[0054] When the user views the visualization model, the knowledge base is determined based on the perspective, and the key viewing perspective is determined based on the incremental information generated by the visualization model in the most recent first time period;

[0055] If the user meets the guidance timing condition within the latest second time period, guiding the user to view the visualization model from the key viewing perspective; the second time period is shorter than the first time period;

[0056] and / or,

[0057] When the user inputs a trigger command for an abnormal plant event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the abnormal plant event;

[0058] and / or,

[0059] When the user inputs a management policy, the management policy is executed on the plantation.

[0060] Preferably, the auxiliary module determines the key viewing perspective based on the perspective-based knowledge base and according to the incremental information generated by the visualization model in the most recent first time period, including:

[0061] Performing feature description on the incremental information to obtain a first description feature set;

[0062] Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base;

[0063] Determine knowledge based on perspective and determine multiple basic perspective requirements;

[0064] Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements;

[0065] Obtaining evaluation values ​​of respective candidate perspectives;

[0066] The candidate viewing angle with the maximum evaluation value is used as the key viewing angle;

[0067] The step of obtaining the evaluation value of each candidate viewing angle includes:

[0068] Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, based on the first quantization table, determining the evaluation value of the traversed candidate view according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view;

[0069] Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model.

[0070] Preferably, the guiding timing conditions include:

[0071] The frequency of the user's perspective change when viewing the visualization model is greater than or equal to the frequency threshold;

[0072] and / or,

[0073] The matching degree between the visual content position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model and the standard position relationship set is greater than or equal to the matching degree threshold;

[0074] and / or,

[0075] The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

[0076] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0077] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] 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:

[0079] Figure 1 A schematic diagram of a plant cultivation intelligent management method according to an embodiment of the present invention;

[0080] Figure 2 Schematic diagram of a plant cultivation intelligent management system in an embodiment of the present invention. DETAILED DESCRIPTION

[0081] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0082] The embodiment of the present invention provides a plant cultivation intelligent management method, such as Figure 1 As shown, including:

[0083] S1, control the robot group to inspect the cultivated plants in the plantation;

[0084] S2. Build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation;

[0085] S3. Assist users to remotely manage plantations online based on visualization models.

[0086] The robot group is composed of multiple robots. The robots can perform mobile inspections on the cultivated plants in the plantation. During the inspection, an inspection screen is generated. The inspection screen includes on-site images of the cultivated plants and marks abnormal plant events identified by the robot using machine vision technology, such as disease and insect pest problems. The position distribution of the robot group is the position distribution of each robot in the robot group in the plantation. The park map of the plantation is a preset three-dimensional map of the plantation. A visualization model is built based on the inspection screen and position distribution of the robot group and the park map of the plantation. When building the model, the position distribution is marked on the park map, and the inspection screen is set at the location of the robot in the park map. The user can be the manager of the plantation. The user can use the visualization model to overview the inspection status of the robot group, etc., and the system will assist the user in conducting online remote management of the plantation based on the visualization model.

[0087] This application controls a robot group to inspect the cultivated plants in the plantation, and builds a visualization model based on the inspection screen and location distribution of the robot group and the park map of the plantation to assist users in online remote management of the plantation based on the visualization model. There is no need for management personnel to go to the plantation for inspection and manage on-site, which reduces labor costs, improves the management efficiency of cultivated plants, and improves the convenience of management for management personnel.

[0088] In one embodiment, the assisting user performs online remote management of the plantation based on the visualization model, including:

[0089] When a user views a visualization model, the key viewing perspective is determined based on the perspective determination knowledge base and the incremental information generated by the visualization model in the most recent first time period; the most recent first time period may be the most recent 300 seconds; the incremental information includes: abnormal plant events newly inspected by the robot group, new position changes of the robot group, changes in the inspected areas of the robot group, etc.; when a user views the visualization model from a key viewing perspective, the user can view key information that the user needs to pay attention to, such as: on-site conditions that require the user to make management decisions immediately, etc.; there is a large amount of perspective determination knowledge in the perspective determination knowledge base, and the perspective determination knowledge can assist in determining the key viewing perspective; the number of inspection screens will be large, the map area of ​​the regional map will also be large, and the volume of the constructed visualization model will be large. When a user views the visualization model, it may not be possible to view the key content conveniently and quickly, which affects management efficiency. Therefore, the introduction of the key viewing perspective can help users conveniently and quickly view the key content, improve management efficiency, and is particularly applicable;

[0090] If the user meets the guidance timing condition within the second most recent time period, the user is guided to view the visualization model from the key viewing perspective; the second time period is less than the first time period; the second most recent time period can be the most recent 100 seconds; if the guidance timing condition is met, it means that the user can be guided to enter the key viewing perspective, and the user is guided to view the visualization model from the key viewing perspective; when the user views the visualization model, he may have his own management plan, and if he is directly guided to enter the key viewing perspective, it may affect him and reduce the user experience. Therefore, the guidance timing condition is introduced to guide the user to enter the key viewing perspective at the right time to improve humanization;

[0091] and / or,

[0092] When the user inputs a trigger instruction for a plant abnormal event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the plant abnormal event. When a plant abnormal event occurs in the plantation, the robot group needs to track the plant abnormal event and temporarily suspend the inspection. At this time, the robot group needs to be temporarily deployed and controlled. After the deployment and control, the newly generated inspection screen of the robot group will be updated on the visualization model. The user can understand the situation of the plant abnormal event in detail and in detail through the visualization model. There is no need for the user to perform temporary deployment and control of the robot group by himself, which greatly improves the applicability of the system.

[0093] and / or,

[0094] When the user inputs a management strategy, the management strategy is executed on the plantation. When the user inputs a management strategy, it means that the user has completed the decision on how to manage the plantation and the management strategy is executed on the plantation.

[0095] In one embodiment, determining the knowledge base based on the perspective, according to the incremental information generated by the visualization model in the most recent first time period, to determine the key viewing perspective includes:

[0096] Performing feature description on the incremental information to obtain a first description feature set;

[0097] Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base;

[0098] Determine knowledge based on perspective and determine multiple basic perspective requirements;

[0099] Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements;

[0100] Obtaining evaluation values ​​of respective candidate perspectives;

[0101] The candidate viewing angle with the maximum evaluation value is used as the key viewing angle;

[0102] The step of obtaining the evaluation value of each candidate viewing angle includes:

[0103] Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, based on the first quantization table, determining the evaluation value of the traversed candidate view according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view;

[0104] Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model.

[0105] There are multiple first description features in the first description feature set, including: the event type of the plant abnormal event newly inspected by the robot group, the location of the event, etc.; the first description feature set represents the situation that requires key viewing perspective determination, and the perspective determination knowledge base has the perspective determination knowledge corresponding to the situation; the perspective determination knowledge can determine the basic perspective requirements, and when the created candidate perspective meets all the basic perspective requirements, it means that the user can view the key information that he needs to pay attention to under the candidate perspective, for example: the first description feature in the first feature description feature set is that the event type of the plant abnormal event newly inspected by the robot group is pests and diseases, and the location of the event is location A, then the basic perspective requirement determined by the corresponding perspective determination knowledge is that all cultivated plants with the pests and diseases at and near location A can be viewed under the candidate perspective, as well as the growth time and responsible personnel of the cultivated plants; there are multiple candidate perspectives, and the evaluation value of each candidate perspective is obtained, and the evaluation value represents the suitability of the candidate perspective. The degree of degree, the candidate perspective with the maximum evaluation value is taken as the key viewing perspective; the standard conditional association relationship between the two perspective constraints can be that the candidate perspective is required to be visible pests and diseases, and the corresponding depth perspective requirement can be that under the candidate perspective, plants of different pests and diseases are arranged left and right or up and down, so as to facilitate user partition viewing; the depth perspective requirement is preset with a requirement weight, and the larger the requirement weight, the greater the suitability of the candidate perspective when the candidate perspective meets the depth perspective requirement; the first quantitative table has evaluation values ​​corresponding to the sum of different requirement weights, and the larger the sum of the requirement weights, the larger the corresponding evaluation value; when the traversed candidate perspective meets the depth perspective requirement, the evaluation value is directly determined based on the requirement weights and the sum; otherwise, it means that the traversed candidate perspective does not meet any depth perspective requirement; the larger the overlap, the more convenient it is for the user to switch from the current real-time perspective to the traversed candidate perspective, and the greater the corresponding suitability, and the second quantitative table has evaluation values ​​corresponding to different overlaps.

[0106] The embodiment of the present invention introduces basic perspective conditions and depth perspective conditions when determining the key viewing perspective based on the perspective determination knowledge base and according to the incremental information generated by the visualization model in the most recent first time period, so as to facilitate the rapid determination of the selected perspective and the evaluation value of the selected perspective. Secondly, the overlap is introduced to comprehensively determine the evaluation value, and the key viewing perspective is finally determined using the evaluation value, which improves the accuracy and efficiency of determining the key viewing perspective.

[0107] In one embodiment, the boot timing condition includes:

[0108] The frequency of perspective changes when the user views the visualization model is greater than or equal to the frequency threshold; the perspective change frequency refers to the frequency of the user switching the perspective to view the visualization model, and the frequency threshold can be 4 times / minute. When the perspective change frequency is greater than or equal to the frequency threshold, it means that the user is not continuously paying attention to a certain part of the visualization model, and the user can be guided to enter the key viewing perspective;

[0109] and / or,

[0110] The matching degree between the visual content position relationship set and the standard position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model is greater than or equal to the matching degree threshold; there are multiple visual content position relationships in the visual content position relationship set, and the visual content position relationship is the content relative position relationship between the visual content of the historical perspective and the key viewing perspective in the visualization model; there are multiple standard position relationships in the standard position relationship set, and the standard position relationship reflects that the user is suitable to enter the key viewing perspective, for example: the cultivated plants near the location A where pests and diseases occur under the historical perspective represent that the user has historically understood the situation near location A, the visual content of the key viewing perspective is the cultivated plants at location A, and the user is suitable to enter the key viewing perspective, then the standard position relationship is that the visual content is adjacent on the visualization model, etc.; the matching degree threshold can be 80%; when the matching degree is greater than or equal to the matching degree threshold, the user can be guided to enter the key viewing perspective;

[0111] and / or,

[0112] The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

[0113] The visual content represents the situation that the user needs to manage. The capability map contains the capability value corresponding to the situation. The larger the capability value, the greater the user's ability to manage the situation. The capability threshold can be 8. When the capability value is greater than or equal to the capability threshold, it means that the user can be guided into the key viewing perspective.

[0114] In one embodiment, the control of the robot group's formation based on the formation knowledge base and according to the abnormal plant event includes:

[0115] Describing the abnormal plant events by features to obtain a second description feature set;

[0116] Determining the formation knowledge corresponding to the second description feature set from the formation knowledge base;

[0117] Based on the array knowledge, determine the search rules and task formulation rules;

[0118] Based on the search rules, search for the basis for the formation task in the visual model;

[0119] Formulate the formation tasks based on the task formulation rules and the formation task formulation basis;

[0120] Control the robot group to perform formation tasks.

[0121] There are multiple second description features in the second description feature set, including: event type and occurrence location of plant abnormal events; the second description feature set reflects the situation where robot formation control is required, and the formation knowledge base contains formation knowledge corresponding to the situation. The formation knowledge can determine the execution rules based on search rules and task formulation rules. The search rules are used to search for execution rules for formulating the basis for formation, and the task formulation rules are used to formulate execution rules for formation tasks based on the basis for formulating formation tasks. For example: the second description feature in the second description feature set is that the event type of the plant abnormal event is dry branches and leaves, and the occurrence location is position B. The corresponding formation task formulation basis is the terrain information of area C where position B is located, the distribution area of ​​each plant in area C, etc. The task formulation rule is to formulate a situation of whether there are other plants drying up in monitoring area C based on the terrain information and the distribution area; finally, the robot group is controlled to perform the formation task.

[0122] In one embodiment, the plant cultivation intelligent management method further includes:

[0123] When the robot group performs the formation task, if at least two robots in the robot group meet the joint trigger condition, a display frame generation template is generated based on the display frame corresponding to the joint trigger condition, and a display frame is generated according to the inspection pictures generated by the corresponding at least two robots in the third historical time period; the third historical time period can be 50 seconds in history;

[0124] The display frame is set in an idle area of ​​the visible area of ​​the visualization model in the real-time perspective of the user currently viewing the visualization model;

[0125] The combined triggering conditions include:

[0126] The current inspection objects of the robots are the same; when the current inspection objects of the robots are the same, a display frame may be generated based on the inspection images generated by at least two robots in the third historical time period, so as to facilitate the users to view together;

[0127] and / or,

[0128] There is a standard task association relationship between the subtasks in the formation task performed by each robot; when the formation task is issued to the robot group, the robot will receive the subtask related to itself; the standard task association relationship means that the task type of the subtask is the same; when there is a standard task association relationship between the subtasks in the formation task performed by each robot, a display frame can be generated based on the inspection screens generated by at least two robots in the third historical time period, so that users can view them together;

[0129] and / or,

[0130] The current straight-line distance between each pair of robots is less than or equal to the distance threshold and the angle between the current inspection angle vectors between the two pairs of robots falls within the standard angle range. The current straight-line distance is the straight-line distance between the two robots at the current moment; the distance threshold can be, for example, 8 meters; the angle between the current inspection angle vectors of each robot is the angle between the current inspection angle vectors of the robots, and the current inspection angle vector is constructed based on the current position of the robot and the direction in which the robot takes the inspection picture;

[0131] The embodiment of the present invention introduces a joint trigger condition. If at least two robots in a robot group meet the joint trigger condition, a display frame generation template is generated based on the display frame corresponding to the met joint trigger condition. A display frame is generated according to the inspection screens generated by the at least two robots in the third historical time period. This facilitates users to quickly and promptly view the screen content that needs to be viewed together, greatly improves the user's management efficiency, and further improves the applicability of the system.

[0132] The embodiment of the present invention provides a plant cultivation intelligent management system, such as Figure 2 As shown, including:

[0133] Control module 1, used to control the robot group to inspect the cultivated plants in the plantation;

[0134] Module 2 is used to build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation;

[0135] Auxiliary module 3 is used to assist users in online remote management of plantations based on visualization models.

[0136] The auxiliary module assists the user to perform online remote management of the plantation based on the visualization model, including:

[0137] When the user views the visualization model, the knowledge base is determined based on the perspective, and the key viewing perspective is determined based on the incremental information generated by the visualization model in the most recent first time period;

[0138] If the user meets the guidance timing condition within the latest second time period, guiding the user to view the visualization model from the key viewing perspective; the second time period is shorter than the first time period;

[0139] and / or,

[0140] When the user inputs a trigger command for an abnormal plant event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the abnormal plant event;

[0141] and / or,

[0142] When the user inputs a management policy, the management policy is executed on the plantation.

[0143] The auxiliary module determines a knowledge base based on the perspective, and determines a key viewing perspective according to the incremental information generated by the visualization model in the most recent first time period, including:

[0144] Performing feature description on the incremental information to obtain a first description feature set;

[0145] Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base;

[0146] Determine knowledge based on perspective and determine multiple basic perspective requirements;

[0147] Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements;

[0148] Obtaining evaluation values ​​of respective candidate perspectives;

[0149] The candidate viewing angle with the maximum evaluation value is used as the key viewing angle;

[0150] The step of obtaining the evaluation value of each candidate viewing angle includes:

[0151] Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, based on the first quantization table, determining the evaluation value of the traversed candidate view according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view;

[0152] Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model.

[0153] The boot timing conditions include:

[0154] The frequency of the user's perspective change when viewing the visualization model is greater than or equal to the frequency threshold;

[0155] and / or,

[0156] The matching degree between the visual content position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model and the standard position relationship set is greater than or equal to the matching degree threshold;

[0157] and / or,

[0158] The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

[0159] The auxiliary module controls the robot group based on the array knowledge base and according to abnormal plant events, including:

[0160] Describing the abnormal plant events by features to obtain a second description feature set;

[0161] Determining the formation knowledge corresponding to the second description feature set from the formation knowledge base;

[0162] Based on the array knowledge, determine the search rules and task formulation rules;

[0163] Based on the search rules, search for the basis for the formation task in the visual model;

[0164] Formulate the formation tasks based on the task formulation rules and the formation task formulation basis;

[0165] Control the robot group to perform formation tasks.

[0166] The plant cultivation intelligent management system also includes:

[0167] Setup module for:

[0168] When the robot group performs the formation task, if at least two robots in the robot group meet the joint trigger condition, a display frame generation template is generated based on the display frame corresponding to the met joint trigger condition, and a display frame is generated according to the inspection pictures generated by the corresponding at least two robots in the third historical time period;

[0169] The display frame is set in an idle area of ​​the visible area of ​​the visualization model in the real-time perspective of the user currently viewing the visualization model;

[0170] The combined triggering conditions include:

[0171] The current inspection objects of each robot are the same;

[0172] and / or,

[0173] There is a standard task association relationship between the subtasks in the formation task performed by each robot;

[0174] and / or,

[0175] The current straight-line distance between each pair of robots is less than or equal to the distance threshold and the current inspection angle vector angle between each pair of robots falls within the standard angle range.

[0176] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A plant cultivation intelligent management method, characterized in that: include: Control the robot group to inspect the cultivated plants in the plantation; Build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation; Assist users to remotely manage plantations online based on visualization models; The auxiliary user performs online remote management of the plantation based on the visualization model, including: When the user views the visualization model, the knowledge base is determined based on the perspective, and the key viewing perspective is determined according to the incremental information generated by the visualization model in the most recent first time period; wherein the incremental information includes: plant abnormal events newly inspected by the robot group, new position changes of the robot group, and changes in the inspected area of ​​the robot group; If the user meets the guidance timing condition within the latest second time period, guiding the user to view the visualization model from the key viewing perspective; the second time period is shorter than the first time period; and / or, When the user inputs a trigger command for an abnormal plant event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the abnormal plant event; and / or, When the user inputs a management strategy, the management strategy is executed on the plantation; The step of determining the knowledge base based on the perspective, and determining the key viewing perspective according to the incremental information generated by the visualization model in the most recent first time period, includes: Performing feature description on the incremental information to obtain a first description feature set; Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base; Determine knowledge based on perspective and determine multiple basic perspective requirements; Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements; Obtaining evaluation values ​​of respective candidate perspectives; The candidate viewing angle with the maximum evaluation value is used as the key viewing angle; The step of obtaining the evaluation value of each candidate viewing angle includes: Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, determining the evaluation value of the traversed candidate view based on the first quantization table according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view; Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model; The boot timing conditions include: The frequency of the user's perspective change when viewing the visualization model is greater than or equal to the frequency threshold; and / or, The matching degree between the visual content position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model and the standard position relationship set is greater than or equal to the matching degree threshold; and / or, The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

2. The plant cultivation intelligent management method according to claim 1, characterized in that: The method of controlling the robot group based on the array knowledge base and according to abnormal plant events includes: Describing the abnormal plant events by features to obtain a second description feature set; Determining the formation knowledge corresponding to the second description feature set from the formation knowledge base; Based on the array knowledge, determine the search rules and task formulation rules; Based on the search rules, search for the basis for the formation task in the visual model; Formulate the formation tasks based on the task formulation rules and the formation task formulation basis; Control the robot group to perform formation tasks.

3. The plant cultivation intelligent management method according to claim 1, characterized in that: Also includes: When the robot group performs the formation task, if at least two robots in the robot group meet the joint trigger condition, a display frame generation template is generated based on the display frame corresponding to the met joint trigger condition, and a display frame is generated according to the inspection pictures generated by the corresponding at least two robots in the third historical time period; The display frame is set in an idle area of ​​the visible area of ​​the visualization model in the real-time perspective of the user currently viewing the visualization model; The combined triggering conditions include: The current inspection objects of each robot are the same; and / or, There is a standard task association relationship between the subtasks in the formation task performed by each robot; and / or, The current straight-line distance between each pair of robots is less than or equal to the distance threshold and the current inspection angle vector angle between each pair of robots falls within the standard angle range.

4. A plant cultivation intelligent management system, characterized in that: include: A control module is used to control the robot group to inspect the cultivated plants in the plantation; Build a module to build a visualization model based on the inspection images and location distribution of the robot group and the park map of the plantation; Auxiliary module, used to assist users in online remote management of plantations based on visualization models; The auxiliary module assists the user to perform online remote management of the plantation based on the visualization model, including: When the user views the visualization model, the knowledge base is determined based on the perspective, and the key viewing perspective is determined according to the incremental information generated by the visualization model in the most recent first time period; wherein the incremental information includes: plant abnormal events newly inspected by the robot group, new position changes of the robot group, and changes in the inspected area of ​​the robot group; If the user meets the guidance timing condition within the latest second time period, guiding the user to view the visualization model from the key viewing perspective; the second time period is shorter than the first time period; and / or, When the user inputs a trigger command for an abnormal plant event in the visualization model, the robot group is deployed and controlled based on the deployment knowledge base and the abnormal plant event; and / or, When the user inputs a management strategy, the management strategy is executed on the plantation; The auxiliary module determines a knowledge base based on the perspective, and determines a key viewing perspective according to the incremental information generated by the visualization model in the most recent first time period, including: Performing feature description on the incremental information to obtain a first description feature set; Determining the perspective determination knowledge corresponding to the first descriptive feature set from the perspective determination knowledge base; Determine knowledge based on perspective and determine multiple basic perspective requirements; Create multiple candidate perspectives for viewing the visualization model; each candidate perspective meets all basic perspective requirements; Obtaining evaluation values ​​of respective candidate perspectives; The candidate viewing angle with the maximum evaluation value is used as the key viewing angle; The step of obtaining the evaluation value of each candidate viewing angle includes: Traversing each candidate view in turn, and each time when the traversed candidate view meets the depth view requirement corresponding to the standard condition association relationship between the two view constraint conditions, determining the evaluation value of the traversed candidate view based on the first quantization table according to the sum of the requirement weights of the depth view requirement met by the traversed candidate view; Otherwise, based on the second quantization table, the evaluation value of the traversed candidate perspective is determined according to the overlap between the traversed candidate perspective and the real-time perspective of the user currently viewing the visualization model; The boot timing conditions include: The frequency of the user's perspective change when viewing the visualization model is greater than or equal to the frequency threshold; and / or, The matching degree between the visual content position relationship set between the historical perspective and the key viewing perspective of the user viewing the visualization model and the standard position relationship set is greater than or equal to the matching degree threshold; and / or, The ability value corresponding to the visual content of the key viewing perspective in the visualization model in the user's ability map is greater than or equal to the ability threshold.

Citation Information

Patent Citations

  • Intelligent management system for agricultural cultivation

    CN116645232A

  • Intelligent inspection method and equipment for thermal power plant

    CN116824731A

  • Inspection method, device and system, electronic equipment and storage medium

    CN117558074A

  • Hydraulic power plant intelligent inspection method and system based on inspection robot

    CN117742334A

  • Intelligent park intelligent monitoring method and system based on artificial intelligence

    CN118097923A