Question and answer type agricultural planting system

Abstract

The embodiment of the invention provides a question and answer type agricultural planting system. The system comprises a voice interaction terminal used for receiving voice questioning information of a user and determining a core entity and a user intention according to the voice questioning information, and the core entity comprises identification information of a target planting area, the variety of a target crop in the target planting area and attention information of the user; the data acquisition module is used for acquiring planting data of a target planting area, and the planting data comprises crop data of a target crop and / or environment data of the target planting area; the model deployment module is used for analyzing the user intention and the core entity by using an artificial intelligence model, generating a decision instruction according to an analysis result and a preset knowledge base, issuing the decision instruction to execution equipment to control the execution equipment to execute a corresponding operation task, and generating a question and answer result according to the planting data, and outputting the question and answer result through the voice interaction terminal. The system does not need manual intervention, and can realize integrated and intelligent management of crop planting.

Description

Technical Field

[0001] This invention relates to the field of smart agriculture technology, and more specifically to a question-and-answer agricultural planting system. Background Technology

[0002] With the development of smart agriculture technology, crop cultivation (such as grapes) is gradually upgrading towards precision and intelligence. Currently, in crop cultivation, growers need to manually inspect and monitor crop growth (e.g., abnormalities such as yellowing, fruit cracking, powdery mildew), and combine this with experience to determine environmental control (supplementary lighting, ventilation, shading, etc.) and water and fertilizer irrigation plans, or operate independently through decentralized equipment terminals (e.g., irrigation controllers, environmental monitoring instruments, etc.), resulting in cumbersome procedures and slow response times. Furthermore, high-value cash crops such as grapes have significantly different requirements for environmental data (e.g., air temperature, air humidity, air CO2 concentration, light intensity, etc.) and water and fertilizer ratios at different growth stages (e.g., budding, flowering, fruit expansion, coloring, and ripening). For example, the suitable temperature for fruit expansion is 25-30℃, and the light intensity for coloring is 25,000-30,000 lux. Traditional manual management methods struggle to accurately match these dynamic needs, easily leading to problems such as the spread of pests and diseases and a decline in fruit quality.

[0003] To address these issues, existing technologies include agricultural knowledge question-and-answer systems and independent execution equipment systems. However, these systems operate independently and lack a collaborative closed loop. After farmers consult on crop cultivation issues through the question-and-answer system, they still need to manually control the execution equipment, such as arranging inspection robots or controlling environmental control equipment. This approach results in low management efficiency, poor coordination, and a high risk of errors. Summary of the Invention

[0004] The present invention was proposed in view of the above-mentioned problems. The present invention provides a question-and-answer agricultural planting system.

[0005] According to one aspect of the present invention, a question-and-answer agricultural planting system is provided, comprising: a voice interaction terminal for receiving voice questions from users and determining core entities and user intentions based on the voice questions, wherein the core entities include identification information of a target planting area, the variety of a target crop in the target planting area, and user attention information; a data acquisition module for collecting planting data of the target planting area, wherein the planting data includes crop data of the target crop and / or environmental data of the target planting area; and a model deployment module for deploying an artificial intelligence model, wherein the model deployment module is communicatively connected to the voice interaction terminal and the data acquisition module, and is used to use the artificial intelligence model to analyze the user intentions and core entities, and generate decision instructions based on the analysis results and a preset knowledge base, and send the decision instructions to the execution equipment to control the execution equipment to perform corresponding work tasks, and generate question-and-answer results based on the planting data, and output the question-and-answer results via the voice interaction terminal; wherein the execution equipment includes one or more of the following: inspection robots, mobile execution terminals, irrigation equipment, fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, pesticide application equipment, and plant protection robots.

[0006] For example, the decision-making instructions include inspection instructions corresponding to the inspection robot and / or mobile execution terminal; the data acquisition module includes the inspection robot and / or mobile execution terminal; crop data is collected by the inspection robot and / or mobile execution terminal; and environmental data is collected by one or more of the following devices: inspection robot, mobile execution terminal, weather station, and a first environmental sensor set up inside the target planting area. When the model deployment module generates question-and-answer results based on the planting data, it specifically generates the question-and-answer results based on the crop data and environmental data collected after the inspection robot and / or mobile execution terminal executes the operation task corresponding to the inspection instructions.

[0007] For example, the voice interaction terminal is specifically used to: perform speech recognition on voice-based questions; convert the speech recognition results into structured text, and extract core entities and user intent from the structured text.

[0008] For example, the information of concern includes abnormal phenomena. A preset knowledge base is used to store the variety parameters of preset crops, abnormal association rules, and water and fertilizer ratio schemes. The variety parameters include at least one crop variety of preset crops, the growth stage of each crop variety, and the preset suitable environment data of each crop variety at different growth stages. The abnormal association rules are used to indicate the association between the abnormal phenomena of preset crops and the causes of the abnormalities. The water and fertilizer ratio scheme includes the amount of water and the nitrogen, phosphorus and potassium ratio required by preset crops at different growth stages.

[0009] For example, the model deployment module generates question-and-answer results based on planting data in the following way: it uses an image recognition algorithm to identify anomalies in crop data within the planting data to obtain anomaly identification results; and it generates question-and-answer results based on the anomaly identification results in response to the user's question.

[0010] For example, the model deployment module includes a cloud server and edge nodes that are communicatively connected to each other. The cloud server is used to deploy artificial intelligence models and maintain a preset knowledge base. The edge nodes are communicatively connected to a voice interaction terminal and a data acquisition module, respectively, and are used to call the artificial intelligence model of the cloud server to parse user intent and core entities, and generate a first decision instruction based on the parsing results and the preset knowledge base called from the cloud server. The cloud server is equipped with an artificial intelligence model, which is used to generate a second decision instruction based on the parsing results, generate question-and-answer results based on planting data, and output the question-and-answer results via the voice interaction terminal. The difficulty level of the task corresponding to the first decision instruction is lower than the difficulty level of the task corresponding to the second decision instruction.

[0011] For example, the question-and-answer agricultural planting system also includes a decision execution module, which is communicatively connected to the model deployment module. The model deployment module sends decision instructions to the execution device through the decision execution module. The decision execution module is used to generate control signals based on the decision instructions and send the control signals to the execution device to control the execution device to perform the corresponding operation task.

[0012] For example, the decision execution module is also used to monitor the operating status of the execution equipment and the execution progress of the task in real time. When it is determined that the execution equipment has malfunctioned based on the operating status, or when it is determined that the task execution is abnormal based on the execution progress, an alarm message is issued and / or the system automatically switches to a backup execution equipment.

[0013] For example, the decision instruction includes a job task; the decision execution module generates control signals based on the decision instruction and sends the control signals to the execution device in the following manner: adding the job task to the task queue, sorting the job tasks in the task queue according to a preset priority rule, generating control signals for the job tasks in the task queue in sequence and sending them to the execution device, wherein, during sorting, if the priority of the first job task currently received is higher than that of the second job task most recently to be executed, the second job task is paused and the first job task is inserted in the task queue before the second job task.

[0014] For example, the task queue is a first-in-first-out queue, and the decision execution module is also used for:

[0015] Merge similar work tasks on the same or adjacent plots; and / or, freeze the work task after its execution is interrupted and record its breakpoint information, and continue the work task based on the breakpoint information when the preset recovery conditions are met; and / or, mark the execution progress of each work task in real time, including pending execution, in execution, completed, and abnormally terminated.

[0016] The question-and-answer agricultural planting system according to embodiments of the present invention can use an AI model as the central hub to establish a complete closed loop of "consultation – data collection – execution equipment control – feedback (i.e., outputting question-and-answer results)". The AI ​​model can combine a preset knowledge base, user intent, core entities, and planting data to generate decision instructions and question-and-answer results, achieving integrated and intelligent management of crop planting. This technical solution integrates voice interaction, knowledge-based question-and-answer, and execution equipment control functions through an AI model for the first time, completing the entire "consultation – data collection – execution equipment control – feedback" chain without human intervention. This effectively solves the problem of collaborative gaps in existing systems, reducing the anomaly response time in crop planting from hours to minutes. Furthermore, the agricultural planting system according to embodiments of the present invention supports voice access, eliminating the need for touchscreen operation, adapting to field operation scenarios for farmers, and reducing the difficulty of use for farmers such as the elderly. Attached Figure Description

[0017] The above and other objects, features, and advantages of the present invention will become more apparent from the more detailed description of the embodiments of the invention in conjunction with the accompanying drawings. The drawings are provided to further illustrate the embodiments of the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same parts or steps.

[0018] Figure 1 A schematic block diagram of a question-and-answer agricultural planting system according to an embodiment of the present invention is shown. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the present invention more apparent, exemplary embodiments according to the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are merely a part of the embodiments of the present invention, and not all of the embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein. Based on the embodiments of the present invention described herein, all other embodiments obtained by those skilled in the art without inventive effort should fall within the protection scope of the present invention.

[0020] In existing technologies, agricultural knowledge question-and-answer systems and execution equipment operate independently. After farmers inquire about crop anomalies, they still need to manually convert their inquiries into inspection tasks, and then manually operate execution equipment such as environmental control devices based on the inspection results. This lacks an automated collaborative mechanism of "questioning-data collection-answering-controlling," resulting in low efficiency and a high risk of errors. However, the question-and-answer agricultural planting system of this invention can achieve full-link linkage of "voice questioning → automatic data collection (e.g., automatic inspection) → data-supported answering → automatic control of execution equipment."

[0021] Figure 1 A schematic block diagram of a question-and-answer agricultural planting system 100 according to an embodiment of the present invention is shown. Figure 1 As shown, the question-and-answer agricultural planting system 100 includes a voice interaction terminal 110, a data acquisition module 120, and a model deployment module 130.

[0022] The voice interaction terminal 110 is used to receive the user's voice question information and determine the core entity and user intent based on the voice question information. The core entity includes the identification information of the target planting area, the variety of the target crop in the target planting area, and the user's attention information.

[0023] The target planting area can be any area, including but not limited to open fields, greenhouses, plastic tunnels, and orchards. The target crop is the crop within the target planting area. The target crop can be any type of crop, including but not limited to grapes, apples, cucumbers, and cabbage. For example, one or more of the following can be used as a voice interaction terminal: a mobile app (e.g., a mobile phone), an inspection robot, or a dedicated voice terminal, to receive voice questions from the user (i.e., the grower). For example, the voice questions could be, "Why are the grape leaves in greenhouse 1 turning yellow?" or "What is the development progress of the wheat in field 3?". For example, the inspection robot can have a built-in voice module to support voice input. The voice interaction terminal 110 can communicate with the model deployment module 130 via methods such as WiFi or Ethernet. The voice interaction terminal 110 can determine the core entities and user intent based on the voice question information. The core entities include the identification information of the target planting area (e.g., greenhouse number, wheat field number, etc.), the variety of the target crop in the target planting area (e.g., grape variety, wheat variety, etc.), and the user's focus information (e.g., "grape leaves are turning yellow," "development progress," etc.). Those skilled in the art will understand that user intent refers to the fundamental goal, task, or problem a user hopes to achieve when asking a question via voice. It transcends the literal meaning of the user's actions or input, pointing to their deeper motivations and expectations. Understanding user intent is a multimodal, context-aware reasoning process. Any existing method for determining user intent based on voice information can be used to implement the operation of determining user intent based on voice question information in this invention. For user intent, multiple preset intent categories can be pre-set, such as querying the cause of anomalies, inquiring about growth status, etc., and the intent category to which the user intent belongs can be determined based on the voice question information. For example, the voice interaction terminal can have an offline wake-up function, that is, it supports offline wake-up via voice.

[0024] The data acquisition module 120 is used to collect planting data of the target planting area, including crop data of the target crop and / or environmental data of the target planting area.

[0025] Planting data may include crop data and / or environmental data. Crop data is data collected for the target crop. For example, crop data may include one or more of color (RGB) images, depth images, thermal infrared images, hyperspectral images, etc. Each of the RGB, depth, thermal infrared, and hyperspectral images can be a static image or a continuous dynamic image, i.e., a video stream. Environmental data may include: soil data reflecting the soil conditions within the target planting area, and / or meteorological data reflecting the air conditions within and / or outside the target planting area. For example, environmental data may include one or more of soil temperature, soil moisture, soil electrical conductivity, soil nutrient content, air temperature, air humidity, light intensity, weather conditions (e.g., sunny, cloudy, etc.), sunshine duration, precipitation, and atmospheric carbon dioxide concentration, etc. Soil electrical conductivity (EC) may reflect the total amount of salt and soluble ions in the soil. Soil nutrient content may be the content of nitrogen (N), phosphorus (P), and potassium (K).

[0026] For example, the data acquisition module includes one or more of the following: a first environmental sensor installed inside the target planting area, a weather station installed inside and / or outside the target planting area, a second environmental sensor installed on the inspection robot, a first camera installed on the inspection robot, a third environmental sensor installed on the mobile execution terminal, and a second camera installed on the mobile execution terminal. Crop data includes first crop data collected by the first camera and / or second crop data collected by the second camera, and environmental data includes one or more of the following: first environmental data collected by the first environmental sensor, second environmental data collected by the second environmental sensor, third environmental data collected by the third environmental sensor, and fourth environmental data collected by the weather station.

[0027] For example, the first camera may include one or more of the following: an RGB camera, a color depth (RGBD) camera, a thermal infrared camera, a hyperspectral camera, etc. Correspondingly, the first crop data may include one or more of the following: RGB images acquired by an RGB camera, RGB images acquired by an RGBD camera, depth images acquired by an RGBD camera, thermal infrared images acquired by a thermal infrared camera, hyperspectral images acquired by a hyperspectral camera, etc. The RGB camera may be, for example, a high-resolution RGB camera with at least 12 million effective pixels. Details of the target crop, such as color, texture, and small lesions, can be clearly identified from the RGB images acquired by the high-resolution RGB camera. The hyperspectral camera of the inspection robot can dynamically adjust imaging parameters through an adaptive spectral selection algorithm to optimize the acquisition quality of crop data. For example, the second camera may include one or more of the following: an RGB camera, a color depth (RGBD) camera, a thermal infrared camera, a hyperspectral camera, etc. Correspondingly, the second crop data may include one or more of the following: RGB images acquired by an RGB camera, RGB images acquired by an RGBD camera, depth images acquired by an RGBD camera, thermal infrared images acquired by a thermal infrared camera, hyperspectral images acquired by a hyperspectral camera, etc. Both the inspection robot and the mobile execution terminal can carry one or more cameras to collect crop data of the target crop.

[0028] For example, the first environmental sensor may include a soil sensor buried in the target planting area and / or a meteorological sensor fixedly installed within the target planting area. For example, in a greenhouse or polytunnel planting scenario, each greenhouse or polytunnel may deploy 6-8 sets of the first environmental sensors according to a "uniform distribution" principle, wherein the installation height of the meteorological sensor can be in the range of 1.5-2m, and the burial depth of the soil sensor can be in the range of 10-15cm. The soil sensor may include one or more of the following: soil temperature sensor, soil moisture sensor, soil conductivity sensor, soil nutrient content sensor, etc. Each of the soil sensors may exist individually or multiple sensors may be integrated together; for example, the soil temperature sensor and the soil moisture sensor may be implemented using an integrated soil temperature and humidity sensor. The meteorological sensor may include one or more of the following: air temperature sensor, air humidity sensor, light sensor, carbon dioxide concentration sensor, etc. Each of the meteorological sensors may exist individually or multiple sensors may be integrated together; for example, the air temperature sensor and the air humidity sensor may be implemented using an integrated air temperature and humidity sensor. For example, the first environmental data may include one or more of the following: soil temperature collected by a soil temperature sensor, soil moisture collected by a soil moisture sensor, soil conductivity collected by a soil conductivity sensor, soil nutrient content collected by a soil nutrient content sensor, air temperature collected by an air temperature sensor, air humidity collected by an air humidity sensor, light intensity collected by a light sensor, and air carbon dioxide concentration collected by a carbon dioxide concentration sensor. The second environmental sensor may include a meteorological sensor mounted on the inspection robot. This meteorological sensor may include one or more of the following: air temperature sensor, air humidity sensor, light intensity sensor, and carbon dioxide concentration sensor. Similarly, each of these meteorological sensors may exist individually or multiple sensors may be integrated together. For example, the second environmental data may include one or more of the following: air temperature collected by an air temperature sensor, air humidity collected by an air humidity sensor, light intensity collected by a light sensor, and air carbon dioxide concentration collected by a carbon dioxide concentration sensor. The third environmental sensor may include a meteorological sensor mounted on a mobile execution terminal. This meteorological sensor may include one or more of the following: air temperature sensor, air humidity sensor, light intensity sensor, and carbon dioxide concentration sensor. Similarly, each of these meteorological sensors may exist individually or multiple sensors may be integrated together. For example, the third environmental data may include one or more of the following: air temperature collected by an air temperature sensor, air humidity collected by an air humidity sensor, light intensity collected by a light sensor, and air carbon dioxide concentration collected by a carbon dioxide concentration sensor. The fourth environmental data is collected by a weather station.Weather stations can be set up in relatively open areas within the target planting area, or around the target planting area. For example, the fourth environmental data may include one or more of the following: air temperature, air humidity, light intensity, air carbon dioxide concentration, weather conditions, sunshine duration, precipitation, etc.

[0029] The data acquisition module 120 can support the automatic acquisition of planting data from environmental sensors, weather stations, inspection robots, etc., and can convert the acquired data into one or more preset data formats, such as JPG, JSON, CSV, etc.

[0030] The model deployment module 130 deploys an artificial intelligence (AI) model. The model deployment module 130 is communicatively connected to the voice interaction terminal 110 and the data acquisition module 120, respectively. It uses the AI ​​model to analyze user intent and core entities, and generates decision instructions based on the analysis results and a preset knowledge base. These decision instructions are then sent to the execution equipment to control the equipment to perform corresponding tasks. Additionally, it generates question-and-answer results based on planting data and outputs these results via the voice interaction terminal 110. The execution equipment includes one or more of the following: inspection robots, mobile execution terminals, irrigation equipment, fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, pesticide application equipment, and plant protection robots.

[0031] In one embodiment, the model deployment module 130 can be implemented using a collaborative processing architecture of cloud servers and edge nodes, meaning that the parsing of user intent and the generation of decision instructions can be distributed across different processing units. In another embodiment, the model deployment module 130 can be implemented using edge nodes alone, meaning the cloud server can be eliminated, and the processing of planting data, model analysis, and generation of decision instructions can all be completed on the edge nodes. Edge nodes can be, for example, local servers. The former embodiment has stronger computing power, supporting the training of complex models and the long-term storage of large-scale data, and is more suitable for scenarios with large-scale planting (e.g., ≥50 greenhouses) and a surge in data processing volume, balancing local real-time performance with the need for cloud computing power expansion. The latter embodiment has limited computing power but can completely operate independently of the cloud, with lower data transmission latency and faster response speed, making it more suitable for scenarios with poor network conditions and small-scale planting (e.g., ≤100 acres). Most existing agricultural planting systems rely entirely on cloud servers for knowledge question answering and task generation, posing a risk of planting data leakage, and cloud response is affected by network conditions, failing to meet the real-time decision-making needs of scenarios such as pest and disease control and sudden environmental changes. The model deployment module 130 can adopt a collaborative processing architecture of cloud server plus edge node or be implemented using edge node alone. This can reduce the dependence on cloud server, reduce the risk of planting data leakage, and has low decision response latency, which can better meet the needs of real-time control.

[0032] In one embodiment, the question-and-answer agricultural planting system may further include a decision execution module, whose hardware implementation is a control center (middle layer). The model deployment module 130 can send decision instructions to the execution equipment through the decision execution module to control the execution equipment to perform the corresponding work tasks. In another embodiment, the decision execution module can be omitted, and the model deployment module 130 can directly send decision instructions to the execution equipment to control the execution equipment to perform the corresponding work tasks. For example, the model deployment module 130 can communicate with the execution equipment to send decision instructions via a wireless communication protocol (such as 5G). This approach is more suitable for scenarios with concentrated greenhouse distribution and good communication environment. Its advantages are that it can simplify the architecture of the agricultural planting system and reduce hardware costs. Its disadvantages are that it lacks centralized monitoring and fault redundancy mechanisms for the execution equipment, and the equipment coordination and stability are slightly reduced. The solution of using the decision execution module as the middle layer helps to realize centralized monitoring and fault redundancy mechanisms for the execution equipment, and helps to improve equipment coordination and stability.

[0033] For example, the AI ​​model can be constructed using one or more algorithms such as Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Random Forest. Preferably, CNN, LSTM, and Random Forest algorithms are integrated to construct a multi-dimensional AI model. For example, the AI ​​model can include one or more of the following: a crop growth status (or growth vigor) monitoring model, a pest and disease prediction model, a water and fertilizer requirement model, and an environmental adaptation model. The crop growth status monitoring model is used to analyze the growth status of the target crop based on at least a portion of the planting data to determine crop growth status information. For example, the crop growth status monitoring model can accurately analyze growth indicators such as leaf area index, fruit development progress, and chlorophyll content based on high-resolution images, hyperspectral images, and environmental data of the target crop. These growth indicators can reflect the growth status of the target crop (e.g., growth stage, presence of abnormalities such as yellowing disease, fruit cracking, and powdery mildew). The pest and disease prediction model is used to predict pests and diseases based on at least a portion of the planting data to determine pest and disease prediction information. For example, pest and disease prediction models can combine hyperspectral images of the target crop, environmental data (such as air temperature, air humidity, and air carbon dioxide concentration), weather conditions, and historical disease data of the target planting area to construct time-series prediction models, providing early warnings of pest and disease risk and spread at predetermined times (e.g., 3-7 days). Water and fertilizer demand models are used to analyze water and fertilizer requirements based on at least some data from the planting data to determine water and fertilizer needs. For example, water and fertilizer demand models can calculate precise irrigation amounts, fertilizer types (nitrogen, phosphorus, and potassium ratios), and fertilizer application times based on soil data (such as soil moisture and soil nutrient content), the target crop's growth stage (i.e., growth cycle, such as flowering period), and meteorological data (such as light intensity, weather conditions, and precipitation). Environmental adaptation models are used to analyze environmental adaptation based on at least some data from the planting data to determine environmental adaptation information. For example, environmental adaptation models can analyze the growth stages of a target crop based on crop data and dynamically match preset suitable environmental data (i.e., environmental adaptation information) for the target crop at different growth stages, such as suitable air temperature, air humidity, light intensity, and air carbon dioxide concentration for the target crop at the current growth stage. Environmental adaptation information can provide a basis for decision-making in environmental control in planting scenarios such as greenhouses.

[0034] In addition, AI models can also include pre-trained Natural Language Processing (NLP) models, such as Bidirectional Encoder Representations from Transformers (BERT) or similar models. Pre-trained NLP models can parse user intent and core entities to obtain parsing results. For example, the parsing result could be "Anomaly cause query; Target object: grapes; Location: Greenhouse No. 1; Symptom description: Leaves are yellowing".

[0035] The model deployment module 130 can generate standardized decision instructions based on the analysis results and a preset knowledge base. For example, the preset knowledge base can be used to store variety parameters, anomaly association rules, and water and fertilizer ratio schemes for preset crops. For instance, for grapes, the preset knowledge base can store grape-specific knowledge, including variety parameters (e.g., the growth stages of Shine Muscat, Summer Black, and Kyoho varieties, and preset suitable environmental data), anomaly association rules (e.g., leaf yellowing - nitrogen deficiency / yellowing disease), and water and fertilizer ratio schemes (nitrogen, phosphorus, and potassium ratios at different growth stages). The preset knowledge base supports both manual and automatic iterative updates. For example, the decision instructions can include core information such as the type of executing device, execution parameters (e.g., irrigation volume, ventilation duration, supplemental lighting intensity), execution duration, and priority.

[0036] For example, the decision execution module or model deployment module 130 can generate control signals based on decision instructions and send the control signals corresponding to the decision instructions (the sending of decision instructions is the sending of the control signals) to the execution equipment to control the execution equipment to perform the corresponding work tasks. The execution equipment includes one or more of the following: inspection robots, mobile execution terminals, irrigation equipment, fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, pesticide application equipment, and plant protection robots.

[0037] Those skilled in the art will understand that the main functions of an inspection robot include perception, reconnaissance, and diagnosis (problem detection), and its output is data and information. The main functions of a plant protection robot include execution, intervention, and handling (problem resolution), and its output is physical actions. Inspection robots may have sensors for data collection, including cameras and / or environmental sensors. Inspection robots possess autonomous movement and path planning capabilities, and can trigger close-range data collection on crop growth status and pest and disease conditions according to preset routes or instructions, and can simultaneously acquire environmental data at the collection points. Plant protection robots may have actuators, such as pumps, nozzles, and robotic arms, to perform one or more agricultural operations such as spraying (i.e., applying pesticides), weeding, pruning, and harvesting. Supplemental lighting equipment, ventilation equipment, and shading equipment can be referred to as environmental control equipment. Irrigation equipment, fertilization equipment, and pesticide application equipment can be implemented independently, or any two or more can be integrated together; for example, irrigation equipment and fertilization equipment can be integrated into a single water and fertilizer machine. Irrigation equipment, fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, and pesticide application equipment can be devices that are fixedly installed in the target planting area. Mobile execution terminals can include one or more of the following: drones, underwater robots, and rail robots. They can function as inspection robots and / or agricultural robots, possessing sensors for data collection and / or actuators for performing agricultural operations. Mobile execution terminals can supplement the insufficient coverage of ground-based inspection robots, fixed environmental sensors, and fixed execution equipment, enabling integrated aerial inspection, high-altitude agricultural operations, and large-scale data collection. They are suitable for planting areas such as large-scale open-field cultivation and complex terrain (e.g., mountains and hills).

[0038] The decision execution module or model deployment module 130 can convert standardized decision instructions into control signals recognizable by the execution equipment and send them to the corresponding execution equipment to control the execution equipment to perform corresponding tasks. Tasks may include data acquisition and / or agricultural operations. Through control signals, the inspection robot and / or mobile execution terminal can be controlled to collect data from target plots and / or target points, and / or one or more of the following—mobile execution terminal, irrigation and fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, pesticide application equipment, and plant protection robot—can be controlled to perform corresponding agricultural operations. The "plot" described herein refers to a logical plot, and each plot can be defined by a set of boundary coordinates. Each plot can be further divided into several points, each point can have a unique identifier and be bound to a single crop. For example, the decision execution module or model deployment module 130 can be designed to simultaneously send control signals to multiple types of execution equipment to support the collaborative execution of multiple types of execution equipment. For instance, after controlling the start of the supplemental lighting equipment in a greenhouse, the decision execution module or model deployment module 130 can simultaneously adjust the operating parameters of the greenhouse's ventilation equipment to maintain the temperature and humidity balance within the greenhouse.

[0039] In addition, the model deployment module 130 can generate question-and-answer results corresponding to user queries based on planting data, and send the results to the voice interaction terminal 110 for output. The information contained in the question-and-answer results can be distinguished according to the user's intent category. For example, for querying the cause of an anomaly, question-and-answer results containing the cause of the anomaly can be generated; for consulting on growth status, question-and-answer results containing crop growth status information can be generated.

[0040] For example, the model deployment module 130 may include a data verification unit and a model optimization unit. The data verification unit can verify the validity of the planting data, such as filtering outliers (e.g., abnormal jump points) in the planting data and filling in missing values ​​in the multimodal data. The model optimization unit can iteratively update the decision rules of the AI ​​model based on the verified valid data, such as adjusting the inspection frequency and optimizing the threshold for triggering environmental regulation.

[0041] By adopting the above technical solution, an AI model can be used as the central hub to establish a complete closed loop of "consultation – data collection – equipment control – feedback (i.e., outputting Q&A results)". The AI ​​model can combine a pre-set knowledge base, user intent, core entities, and planting data to generate decision-making instructions and Q&A results, achieving integrated and intelligent management of crop planting. This technical solution integrates voice interaction, knowledge-based Q&A, and equipment control functions through an AI model for the first time, completing the entire "consultation – data collection – equipment control – feedback" chain without human intervention. This effectively solves the problem of collaborative gaps in existing systems, reducing the response time for crop planting anomalies from hours to minutes. Furthermore, existing agricultural planting systems often rely on button and touchscreen operations, requiring growers to interrupt their work terminals during field operations, making them unsuitable for actual field use scenarios. The agricultural planting system according to this embodiment supports voice access, eliminating the need for touchscreen operation, adapting to field operation scenarios, and reducing the difficulty of use for growers, especially elderly ones.

[0042] According to an embodiment of the present invention, the decision instruction includes an inspection instruction corresponding to the inspection robot and / or mobile execution terminal, the data acquisition module includes the inspection robot and / or mobile execution terminal, crop data is collected by the inspection robot and / or mobile execution terminal, and environmental data is collected by one or more of the following devices: inspection robot, mobile execution terminal, weather station, and a first environmental sensor set up inside the target planting area; when the model deployment module generates question-and-answer results based on the planting data, it specifically generates question-and-answer results based on the crop data and environmental data collected after the inspection robot and / or mobile execution terminal performs the operation task corresponding to the inspection instruction.

[0043] As described above, the data acquisition module may include one or more of the following: a first environmental sensor installed inside the target planting area, a weather station installed inside and / or outside the target planting area, a second environmental sensor installed on the inspection robot, a first camera installed on the inspection robot, a third environmental sensor installed on the mobile execution terminal, and a second camera installed on the mobile execution terminal. Crop data includes first crop data collected by the first camera and / or second crop data collected by the second camera, and environmental data includes one or more of the following: first environmental data collected by the first environmental sensor, second environmental data collected by the second environmental sensor, third environmental data collected by the third environmental sensor, and fourth environmental data collected by the weather station.

[0044] The inspection robot and / or mobile execution terminal can collect crop data and / or at least some environmental data. Decision instructions include a first inspection instruction corresponding to the inspection robot and / or a second inspection instruction corresponding to the mobile execution terminal. The first inspection instruction controls the inspection robot to perform a first inspection task, and the second inspection instruction controls the mobile execution terminal to perform a second inspection task. The inspection task is a work task. For example, the first and second inspection instructions may each include one or more of the following: identification information of the inspection task (e.g., task ID), identification information of the target planting area (e.g., greenhouse number), inspection object, data collection indicators (distance from the target crop, height, number of images, etc.), inspection plot, inspection point, and priority of the inspection task. The task content of the first and second inspection tasks is the collection of crop data and / or at least some environmental data. In this case, when the model deployment module generates question-and-answer results based on the planting data, it can specifically generate the question-and-answer results based on the crop data and environmental data collected after the inspection robot and / or mobile execution terminal perform the work task corresponding to the inspection instruction. The crop data used to generate the question-and-answer results is crop data collected after the inspection robot and / or mobile execution terminal perform the work tasks corresponding to the inspection instructions. This data is entirely collected by the inspection robot and / or mobile execution terminal. The environmental data used to generate the question-and-answer results is environmental data collected after the inspection robot and / or mobile execution terminal perform the work tasks corresponding to the inspection instructions. This data may include only the environmental data collected by the inspection robot and / or mobile execution terminal, only the environmental data collected by the first environmental sensor and / or weather station, or both.

[0045] By adopting the above scheme, a closed-loop logic of "consultation – data collection (including inspection) – execution of equipment control – feedback (i.e., output of Q&A results)" can be further formed. The data obtained from the inspection can be used to generate Q&A results, thus forming automated control and result feedback from consultation to inspection, making it convenient for users to quickly understand crop information.

[0046] According to an embodiment of the present invention, the voice interaction terminal is specifically used for: performing voice recognition on voice-based questions; converting the voice recognition result into structured text; and extracting core entities and user intent from the structured text.

[0047] The voice interaction terminal may include a speech recognition unit and a text normalization unit. The speech recognition unit can recognize spoken questions. For example, the spoken questions may be in Mandarin Chinese, and a Mandarin-specific Automatic Speech Recognition (ASR) model can be used for speech recognition. This model can filter environmental noise from ventilation equipment, irrigation equipment, etc., achieving an accuracy rate of ≥95%. The text normalization unit can convert the speech recognition results into structured text, extract core entities (such as greenhouse number, grape variety, and information of interest) and user intent from the structured text, and upload them to the model deployment module.

[0048] By adopting the above technical solution, speech recognition and structured text conversion can be performed on voice-based questions, making it convenient to quickly and accurately extract core entities and user intent from structured text.

[0049] According to an embodiment of the present invention, the information of concern includes abnormal phenomena. A preset knowledge base is used to store the variety parameters of preset crops, abnormal association rules, and water and fertilizer ratio schemes. The variety parameters include at least one crop variety of preset crops, the growth stage of each crop variety, and the suitable environmental data of each crop variety at different growth stages. The abnormal association rules are used to indicate the association between the abnormal phenomena of preset crops and the causes of abnormalities. The water and fertilizer ratio scheme includes the amount of water and the nitrogen, phosphorus and potassium ratio required by preset crops at different growth stages.

[0050] As mentioned above, a preset knowledge base can be used to store variety parameters, anomaly association rules, and water and fertilizer ratio schemes for preset crops. A preset crop can be one or more crops; that is, a dedicated preset knowledge base can be set up for each preset crop, or knowledge of multiple crops can be stored in the same preset knowledge base. For example, for grapes, the preset knowledge base can store grape-specific knowledge, including variety parameters (e.g., growth stages and suitable environmental data for Shine Muscat, Summer Black, and Kyoho varieties), anomaly association rules (e.g., yellowing leaves - nitrogen deficiency / yellowing disease), and water and fertilizer ratio schemes (nitrogen, phosphorus, and potassium ratios for different growth stages). Anomaly association rules can determine the cause of the anomaly corresponding to the user's inquiry (e.g., "grape leaves are yellowing"). Water and fertilizer ratio schemes can determine the water and fertilizer requirements of the target crop at the current growth stage, i.e., the required water volume and nitrogen, phosphorus, and potassium ratios. Existing general-purpose algorithm models in agricultural planting systems are not customized for crop variety characteristics (e.g., the differentiated needs of Shine Muscat, Summer Black, and Kyoho varieties) and growth stages. Suitable environmental data and water and fertilizer ratio schemes are highly generalized but lack accuracy, making it difficult to guarantee crop quality and yield. Using the above scheme, the variety parameters, abnormal association rules, and water and fertilizer ratio schemes of the preset crops can be stored in a preset knowledge base. Customized design of inspection indicators (i.e. suitable environmental data) and water and fertilizer ratio schemes can be carried out for different crop varieties and different growth stages. The adaptability is better than that of general algorithm models, which can reduce the incidence of pests and diseases by about 15%-20% and increase the fruit quality rate by about 10%-15%.

[0051] By adopting the above scheme, the preset crop variety parameters, anomaly association rules, and water and fertilizer ratio schemes can be stored in a preset knowledge base. The preset knowledge base facilitates data updates, and therefore, it is easy to expand the anomaly association rules and water and fertilizer ratio schemes to more accurately respond to user inquiries or more precisely control irrigation and fertilization equipment to perform operational tasks.

[0052] According to an embodiment of the present invention, the model deployment module generates question-and-answer results based on planting data in the following manner: anomaly identification is performed on crop data in the planting data using an image recognition algorithm to obtain anomaly identification results; and question-and-answer results are generated based on the anomaly identification results in response to user questions.

[0053] Anomalies in crop data within planting data can be identified using any existing or future image recognition algorithm. Anomalies refer to abnormalities in the growth state of the target crop, such as the presence of yellowing disease, fruit cracking, or powdery mildew. The input to the image recognition algorithm is the crop data, and the output is the anomaly identification result. For example, the image recognition algorithm could be a You Only Look Once (YOLO) model, a U-Net model, or specifically, the YOLOv8 model. Question-and-answer results tailored to user queries can be generated based on the anomaly identification results.

[0054] By adopting the above technical solution, it is possible to determine whether there are any abnormalities in the growth status of the target crop based on crop data, and to provide feedback on the abnormality identification results to the user so that the user can understand the situation of the target crop with abnormalities in a timely manner and take countermeasures.

[0055] According to an embodiment of the present invention, the model deployment module includes a cloud server and edge nodes that are communicatively connected to each other. The cloud server is used to deploy AI models and maintain a preset knowledge base. The edge nodes are communicatively connected to a voice interaction terminal and a data acquisition module, respectively, and are used to call the AI ​​model of the cloud server to parse user intent and core entities, and generate a first decision instruction based on the parsing results and the preset knowledge base called from the cloud server. The cloud server is equipped with an AI model, which is used to generate a second decision instruction based on the parsing results, generate question-and-answer results based on planting data, and output the question-and-answer results via the voice interaction terminal. The difficulty level of the task corresponding to the first decision instruction is lower than the difficulty level of the task corresponding to the second decision instruction.

[0056] As mentioned above, the model deployment module 130 can be implemented using a collaborative processing architecture of cloud server and edge nodes. For example, an architecture of "local edge node + small cloud server" can be constructed. The edge node can be mainly responsible for parsing user intent and core entities, generating first decision instructions (i.e., decision instructions corresponding to simple tasks), and controlling the operation of execution equipment in real time based on the first decision instructions. The cloud server can be mainly responsible for complex data analysis (e.g., generating second decision instructions corresponding to complex tasks, generating question-and-answer results based on planting data, etc.) and updating the preset knowledge base. The difficulty level of each task can be preset as needed. This solution is mainly applicable to scenarios with a large planting scale (e.g., ≥50 greenhouses) and a surge in data processing volume, balancing local real-time performance with the expansion needs of cloud computing power. The advantage is that it can share the load of the local server and support large-scale expansion. The disadvantage is that the communication stability between the cloud and the local machine needs to be ensured, and the data security is slightly lower than that of a fully local deployment.

[0057] According to an embodiment of the present invention, the question-and-answer agricultural planting system further includes a decision execution module, which is communicatively connected to the model deployment module. The model deployment module sends decision instructions to the execution device through the decision execution module. The decision execution module is used to generate control signals based on the decision instructions and send the control signals to the execution device to control the execution device to perform the corresponding operation task.

[0058] As described above, the model deployment module can send decision instructions to the execution devices through the decision execution module to control the execution devices to perform the corresponding tasks. Using the decision execution module as an intermediate layer helps to achieve centralized monitoring and fault redundancy mechanisms for the execution devices, and improves device coordination and stability.

[0059] According to an embodiment of the present invention, the decision execution module is also used to monitor the operating status of the execution equipment and the execution progress of the task in real time. When it is determined that the execution equipment has malfunctioned based on the operating status, or when it is determined that the task execution is abnormal based on the execution progress, an alarm message is issued and / or the system automatically switches to a backup execution equipment.

[0060] The decision execution module may include a status monitoring unit for real-time monitoring of the operating status and task progress of the execution equipment. Operating status includes, for example, the flow rate of irrigation equipment, the battery level of the plant protection robot, and the working status of environmental control equipment. When the operating status indicates a malfunction in the execution equipment, or when the execution progress indicates an abnormal task execution, the status monitoring unit can immediately trigger an alarm, such as by pushing a notification to the manager's mobile terminal. This promptly reminds the manager to inspect and maintain the execution equipment, ensuring the normal execution of the task. When the operating status indicates a malfunction in the execution equipment, or when the execution progress indicates an abnormal task execution, the status monitoring unit can also automatically switch to a backup execution equipment. The backup execution equipment then takes over from the malfunctioning or abnormal execution equipment to perform subsequent tasks, ensuring the continuity of the task. For example, issuing alarm information and automatically switching to the backup execution equipment can be performed simultaneously.

[0061] According to an embodiment of the present invention, the decision instruction includes a job task; the decision execution module generates a control signal based on the decision instruction and sends the control signal to the execution device in the following manner: adding the job task to the task queue, sorting the job tasks in the task queue according to a preset priority rule, generating control signals for the job tasks in the task queue in sequence and sending them to the execution device, wherein, during sorting, if the priority of the first job task currently received is higher than that of the second job task most recently to be executed, the second job task is paused and the first job task is inserted in the task queue before the second job task.

[0062] The decision execution module may include an instruction scheduling unit. This unit adds tasks to a task queue, sorts them according to a preset priority rule, and generates control signals for the tasks in the queue in that order, sending them to the execution equipment. For example, the preset priority rule could be something like "emergency control (weight 0.6) > critical growth period (weight 0.3) > routine monitoring (weight 0.1)". The instruction scheduling unit supports the execution of high-priority tasks (hereinafter referred to as high-priority tasks) in the queue. When a high-priority task is generated, it can pause the execution of lower-priority tasks (hereinafter referred to as low-priority tasks). High-priority tasks could be, for example, emergency pest and disease control or sudden changes in environmental data. This scheduling scheme allows for timely responses to events requiring urgent handling by executing corresponding tasks in the queue, thus improving the management of the planting area.

[0063] According to an embodiment of the present invention, the task queue is a first-in-first-out queue, and the decision execution module is further configured to: merge similar work tasks of the same plot or adjacent plots; and / or, freeze the work task and record the breakpoint information of the work task after the execution of any work task is interrupted, and continue to execute the work task based on the breakpoint information when the preset recovery conditions are met; and / or, mark the execution progress of each work task in real time, the execution progress including pending execution, in execution, completed, and abnormally terminated.

[0064] For example, a First-In-First-Out (FIFO) queue can be used to store job tasks, which can effectively save storage space while ensuring the order of task execution. For example, the decision execution module can support merging similar job tasks on the same or adjacent plots. Similar job tasks include, for example, collecting image data of different crops on the same or adjacent plots using an inspection robot. This approach is particularly suitable for scenarios where the job task is inspection. This allows for dynamic adjustment of the number of job tasks to evenly distribute the working resources of the execution equipment, effectively saving equipment operation time.

[0065] For example, the decision execution module can support automatic resumption of execution after a task is interrupted. This approach is particularly suitable for scenarios where the task is an inspection task. For instance, there may be various reasons for an inspection task to be interrupted, such as the inspection point being temporarily inaccessible due to an obstacle, the inspection point being temporarily restricted, insufficient resources for the inspection robot (e.g., insufficient battery power), or manual suspension of the inspection task. When an inspection task is interrupted, it can be temporarily frozen and the breakpoint information recorded, and then resumed when preset recovery conditions are met. Freezing means keeping the inspection task in the task queue but not participating in sorting and distribution. Resumption includes allowing the inspection task to participate in sorting and distribution. Resumption can be triggered by, for example, the inspection point's path becoming accessible again, the restricted period for the inspection point ending, sufficient resources for the inspection robot being restored (e.g., battery power greater than or equal to a preset battery threshold), or manual unfreezing of the inspection task. Breakpoint information can include one or more of the following: interruption time, interruption location, interruption reason, progress, and on-site snapshot. Interruption time is the moment the interruption occurred. Interruption location is the specific location at the time of the interruption, such as a specific plot of land or point. Reasons for interruption may include, for example, sudden thunderstorms, drone disconnection, insufficient battery power, or the discovery of a major epidemic requiring urgent reporting. Progress can include a list of completed and uncompleted inspection plots, points, or items. On-site snapshots can include photos or videos of the on-site environment (such as weather conditions and crop status) at the time of interruption. This approach enables recovery from interrupted tasks and resumption of inspections, thus saving task execution costs and improving the robustness of the agricultural planting system.

[0066] For example, the decision execution module can also mark the execution progress of each task in real time, such as marking it as "pending execution," "in execution," "completed," or "abnormally terminated." Interruption of task execution can be considered an abnormal termination. By marking the execution progress of each task in real time, the decision execution module can easily monitor the execution status of tasks and make timely adjustments, such as pausing tasks that have "abnormally terminated," to avoid interfering with other tasks.

[0067] According to an embodiment of the present invention, the decision execution module is further configured to receive execution feedback data from the execution device and send the execution feedback data to the model deployment module. The model deployment module is further configured to: iteratively train the AI ​​model based on the execution feedback data and newly collected planting data within a preset time period after the execution device performs the task.

[0068] The model deployment module supports continuous learning and transfer learning of the AI ​​model. The decision execution module receives execution feedback data from the execution equipment. The data acquisition module collects planting data in real time. After the execution equipment performs its tasks, the planting data and execution feedback data are fed back to the model deployment module in real time by the data acquisition module and the decision execution module, respectively, so that the model deployment module can use this data to iteratively train the AI ​​model. Transfer learning can quickly apply the knowledge of existing crop models to similar crops, reducing the data and time costs required for training new crop models by more than 60%. In addition, customized AI models can be developed for different crop varieties, growth stages, and planting area environments.

[0069] For example, the execution feedback data may include the actual execution parameters of the executing device. For irrigation equipment, actual execution parameters may include: actual operating time, actual water flow (i.e., actual irrigation volume), water pressure, valve opening degree, etc. For fertilization equipment, actual execution parameters may include: EC / pH value, etc. For ventilation equipment, actual execution parameters may include: actual fan speed, actual window opening angle, actual wind speed, actual ventilation time, etc. For supplemental lighting equipment, actual execution parameters may include: actual supplemental lighting intensity, actual operating time, etc. For shading equipment, actual execution parameters may include the actual opening / closing percentage of the shading curtain, etc. Each type of executing device has its own corresponding actual execution parameters, which can be set according to actual needs; these will not be elaborated upon here. The parameter type (first parameter type) of the actual execution parameters of each executing device may be consistent with the parameter type (second parameter type) of the execution parameters of that executing device included in the above decision instruction, or the first parameter type corresponding to each executing device may at least include the second parameter type corresponding to that executing device.

[0070] Existing general-purpose algorithm models lack a continuous learning mechanism and cannot dynamically update model parameters based on new data, leading to performance degradation after long-term use. However, the AI ​​model according to embodiments of the present invention combines transfer learning and continuous learning techniques to dynamically update model parameters, resulting in prediction accuracy and environmental adaptability that are superior to general-purpose algorithm models. Based on planting data and execution feedback data, the AI ​​model can dynamically adjust the inspection frequency of the inspection robot, environmental data, and water and fertilizer ratio schemes, avoiding redundant inspections and ineffective adjustments, thereby reducing energy consumption and labor costs by approximately 30% or more.

[0071] Although exemplary embodiments have been described herein with reference to the accompanying drawings, it should be understood that the above exemplary embodiments are merely illustrative and are not intended to limit the scope of the invention. Various changes and modifications can be made therein by those skilled in the art without departing from the scope and spirit of the invention. All such changes and modifications are intended to be included within the scope of the invention as claimed in the appended claims.

[0072] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0073] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed.

[0074] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0075] Similarly, it should be understood that, in order to streamline the invention and aid in understanding one or more of the various aspects of the invention, features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof in the description of exemplary embodiments of the invention. However, this approach should not be construed as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the corresponding claims, its inventive point lies in solving the corresponding technical problem with fewer features than all of those in a single disclosed embodiment. Therefore, the claims following the detailed description are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0076] Those skilled in the art will understand that, apart from the mutual exclusion of features, all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or apparatus so disclosed can be combined in any combination. Unless otherwise expressly stated, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0077] Furthermore, those skilled in the art will understand that although some embodiments described herein include certain features but not others included in other embodiments, combinations of features from different embodiments are intended to be within the scope of the invention and form different embodiments. For example, in the claims, any of the claimed embodiments can be used in any combination.

[0078] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some modules in the question-and-answer agricultural planting system according to embodiments of the present invention. The present invention can also be implemented as an apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0079] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0080] The above description is merely a specific embodiment of the present invention or an explanation of that embodiment. The scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A question-and-answer agricultural planting system, characterized in that, include: A voice interaction terminal is used to receive voice questions from users and determine core entities and user intentions based on the voice questions. The core entities include identification information of the target planting area, the variety of the target crop in the target planting area, and the user's attention information. The data acquisition module is used to collect planting data of the target planting area, including crop data of the target crop and / or environmental data of the target planting area; The model deployment module deploys an artificial intelligence model. The model deployment module is communicatively connected to the voice interaction terminal and the data acquisition module, respectively. It is used to use the artificial intelligence model to analyze the user intent and the core entity, generate decision instructions based on the analysis results and a preset knowledge base, send the decision instructions to the execution device to control the execution device to perform the corresponding work task, generate question and answer results based on the planting data, and output the question and answer results through the voice interaction terminal. The execution equipment includes one or more of the following: inspection robot, mobile execution terminal, irrigation equipment, fertilization equipment, supplemental lighting equipment, ventilation equipment, shading equipment, pesticide application equipment, and plant protection robot.

2. The question-and-answer agricultural planting system according to claim 1, characterized in that, The decision-making instructions include the inspection instructions corresponding to the inspection robot and / or the mobile execution terminal; the data acquisition module includes the inspection robot and / or the mobile execution terminal; the crop data is collected by the inspection robot and / or the mobile execution terminal; and the environmental data is collected by one or more of the following devices: the inspection robot, the mobile execution terminal, a weather station, and a first environmental sensor installed inside the target planting area. When generating question-and-answer results based on the planting data, the model deployment module specifically generates the question-and-answer results based on the crop data and the environmental data collected after the inspection robot and / or the mobile execution terminal execute the work task corresponding to the inspection instruction.

3. The question-and-answer agricultural planting system according to claim 1 or 2, characterized in that, The voice interaction terminal is specifically used for: Perform speech recognition on the voice question information; The speech recognition results are converted into structured text, and the core entities and user intent are extracted from the structured text.

4. The question-and-answer agricultural planting system according to claim 1 or 2, characterized in that, The information of concern includes abnormal phenomena. The preset knowledge base is used to store the variety parameters of preset crops, abnormal association rules, and water and fertilizer ratio schemes. The variety parameters include at least one crop variety of the preset crop, the growth stage of each crop variety, and the preset suitable environment data of each crop variety at different growth stages. The abnormal association rules are used to indicate the association between the abnormal phenomena of the preset crop and the abnormal causes. The water and fertilizer ratio scheme includes the amount of water and the nitrogen, phosphorus and potassium ratio required by the preset crop at different growth stages.

5. The question-and-answer agricultural planting system according to claim 1 or 2, characterized in that, The model deployment module generates question-and-answer results based on the planting data in the following manner: Anomaly identification results are obtained by using an image recognition algorithm to identify anomalies in the crop data within the planting data. The question-and-answer results are generated based on the anomaly identification results and in response to the user's question.

6. The question-and-answer agricultural planting system according to claim 1 or 2, characterized in that, The model deployment module includes a cloud server and edge nodes that are communicatively connected to each other. The cloud server is used to deploy the artificial intelligence model and maintain the preset knowledge base. The edge node is communicatively connected to the voice interaction terminal and the data acquisition module, respectively, and is used to call the artificial intelligence model of the cloud server to parse the user intent and the core entity, and generate a first decision instruction based on the parsing result and the preset knowledge base called from the cloud server; The cloud server is equipped with the artificial intelligence model, which is used to generate a second decision instruction based on the analysis results, generate a question-and-answer result based on the planting data, and output the question-and-answer result via the voice interaction terminal. The difficulty level of the task corresponding to the first decision instruction is lower than the difficulty level of the task corresponding to the second decision instruction.

7. The question-and-answer agricultural planting system according to claim 1 or 2, characterized in that, The question-and-answer agricultural planting system also includes a decision execution module, which is communicatively connected to the model deployment module. The model deployment module sends the decision instructions to the execution device through the decision execution module. The decision execution module is used to generate control signals based on the decision instructions and send the control signals to the execution device to control the execution device to perform the corresponding work tasks.

8. The question-and-answer agricultural planting system according to claim 7, characterized in that, The decision execution module is also used to monitor the operating status of the execution equipment and the execution progress of the task in real time. When it is determined that the execution equipment has malfunctioned based on the operating status, or when it is determined that the task is abnormal based on the execution progress, an alarm message is issued and / or the system automatically switches to a backup execution equipment.

9. The question-and-answer agricultural planting system according to claim 7, characterized in that, The decision instruction includes the work task; the decision execution module generates a control signal based on the decision instruction and sends the control signal to the execution device in the following manner: The task is added to the task queue, and the tasks in the task queue are sorted according to a preset priority rule. The task in the task queue generates the control signal in sequence and sends it to the execution device. During sorting, if the priority of the first task received is higher than that of the second task to be executed most recently, the second task is paused and the first task is inserted in the task queue before the second task.

10. The question-and-answer agricultural planting system according to claim 7, characterized in that, The task queue is a first-in-first-out queue, and the decision execution module is further used for: Combine similar work tasks on the same or adjacent plots; and / or, After any job task is interrupted, the job task is frozen and its breakpoint information is recorded. The job task is then resumed based on the breakpoint information when a preset recovery condition is met; and / or, The execution progress of each job task is marked in real time, including pending execution, in execution, completed, and abnormal termination.

Images