Intelligent agricultural machinery collaborative operation and efficiency optimization system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent agriculture, in particular to an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things, which comprises an edge execution calculation module, a multi-source sensing monitoring module, a cloud general control processing module and a fault emergency and task continuing module. Through an innovative cloud-edge-end three-level collaborative architecture, cloud global optimization scheduling, edge side real-time calculation control and terminal accurate execution are deeply fused, and the architecture guarantees stability and low delay of data transmission in a complex farmland environment by utilizing multiple communication network technologies; the edge calculation module not only realizes local data fusion and real-time path planning, but also ensures the continuity of a calculation task when the agricultural machine moves through an integrated service migration management function, and the system realizes a closed loop from global intelligent decision to local transient response on the whole; and the real-time performance, the self-adaptability and the overall operation stability of collaborative operation of the agricultural machinery group are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of smart agriculture technology, specifically to an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things. Background Technology

[0002] With the acceleration of agricultural modernization, agricultural machinery automation, intelligence, and collaborative operations have become important development directions for improving agricultural production efficiency and reducing reliance on manual labor. Traditional agricultural machinery operations rely heavily on manual scheduling and single-machine control, making it difficult to achieve multi-machine collaboration, global optimization, and dynamic response. This results in problems such as low operational efficiency, insufficient resource utilization, and weak ability to cope with emergencies. In recent years, the deep integration of technologies such as the Internet of Things, edge computing, and artificial intelligence with agricultural equipment has promoted the research and application of intelligent agricultural machinery systems. These systems aim to achieve autonomous collaboration and optimized operational efficiency of agricultural machinery groups through information perception, intelligent decision-making, and collaborative control.

[0003] For example, a collaborative operation system for multiple intelligent agricultural machines and supply and transportation agricultural machines, with application number CN202410888538.4 and publication date 20241029, includes a data acquisition module, a data processing module, a collaborative operation module, and a control module. The data processing module calculates the preparation coefficient Qi of the multiple intelligent agricultural machines, the coordination coefficient Lz of the multiple intelligent agricultural machines, the preparation coefficient Ql of the supply and transportation agricultural machines, the coordination coefficient Ly of the supply and transportation agricultural machines, and the communication coordination coefficient Arj. The collaborative operation module uses the formula for the communication coordination coefficient Arj to simulate the optimal coordination coefficient. The control module uses the preparation coefficient Qi of the multiple intelligent agricultural machines, the coordination coefficient Lz of the multiple intelligent agricultural machines, the preparation coefficient Ql of the supply and transportation agricultural machines, and the coordination coefficient Ly of the supply and transportation agricultural machines to adjust the optimal preparation time, enabling the multiple intelligent agricultural machines and the supply and transportation agricultural machines to have the advantage of effective collaboration.

[0004] For example, a multi-objective optimization task scheduling method based on agricultural machinery operation efficiency, with application number CN202311634006.X and publication date of 20240223, includes the following steps: Step S1, Task decomposition: Decompose the agricultural machinery operation tasks of the farm into several sub-tasks; Step S2, Model building: Based on the information of the sub-tasks, build a multi-objective optimization model including operation efficiency; Step S3, Solve the model: Solve the model using NSGA. Algorithm II performs a series of solutions on the model established in step S2 to obtain a set of optimal solutions for the agricultural machinery task scheduling scheme; Step S4, Solution Decoding: After solving the model, the obtained set of optimal solutions is decoded; Step S5, Scheduling Scheme Execution: The agricultural machinery executes the sub-tasks assigned to it according to the decoded task scheduling scheme; By combining the characteristics of agricultural machinery operation efficiency, multi-objective optimization model, and NSGA... Algorithm II achieves multi-objective optimization scheduling of agricultural machinery tasks, effectively improving the operational efficiency of agricultural machinery.

[0005] Traditional and existing intelligent agricultural machinery systems are relatively weak in terms of secure data storage, reliable traceability, and visual monitoring and decision support. This results in a lack of efficient task succession and emergency path planning mechanisms when dealing with sudden agricultural machinery failures. In complex farmland environments, it is difficult to ensure the continuous stability of communication links and the reliability of equipment. Furthermore, the edge computing task migration and multimodal perception data fusion capabilities are insufficient. Global optimization scheduling lacks the support of digital twin models that interact with the physical world in real time. The efficiency of task dynamic reallocation algorithms needs to be improved, and there is a general lack of data-driven predictive maintenance capabilities for key components.

[0006] In view of this, there is an urgent need to design an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things to solve the above problems. Summary of the Invention

[0007] The purpose of this invention is to provide an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things to address the aforementioned shortcomings in the prior art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: An IoT-based intelligent agricultural machinery collaborative operation and efficiency optimization system includes an edge execution computing module, a multi-source sensing and monitoring module, a cloud-based central control and processing module, and a replacement execution module. The edge execution computing module, the multi-source sensing and monitoring module, the cloud-based central control and processing module, and the replacement execution module are connected through an IoT communication network. The IoT communication network is a wireless wide-area communication network that supports hybrid networking, including at least 5G, 4G and NB-IoT cellular networks; The edge execution computing module and the multi-source sensing and monitoring module integrate an industrial-grade IoT card, which has a wide operating temperature range of -40℃ to 85℃ and an IP67 or higher protection rating. It is used to establish and maintain a stable, low-latency communication link with the cloud-based central control processing module in complex farmland environments.

[0009] The edge execution computing module is deployed on or near the agricultural machinery to perform local fusion and preprocessing of the sensed data, and to perform agricultural machinery collaborative control, local path planning and safety monitoring based on real-time information; It should be noted that: The edge execution computing module includes a local data fusion and preprocessing unit, a real-time collaborative control and path planning unit, a security threshold monitoring and primary response unit, and a service migration management unit, wherein: The local data fusion and preprocessing unit is used to clean, spatiotemporally register and fuse heterogeneous data from the multi-source sensing and monitoring module to generate comprehensive agricultural machinery pose information. The real-time collaborative control and path planning unit is used to run an improved ant colony algorithm or a dynamic path optimization algorithm, and to perform local collision avoidance, formation maintenance and dynamic path replanning based on real-time obstacles and the positions of adjacent agricultural machinery. The safety threshold monitoring and primary response unit is used to preset multi-level operating thresholds for engine temperature, oil pressure, and operating speed, and to trigger local alarms or protective operations when the monitored data exceeds the limits. The service migration management unit is used to achieve seamless migration of computing tasks and status data when the movement of agricultural machinery causes the switching of its associated edge computing nodes, based on the task-service correlation model. The multi-source sensing and monitoring module is used to collect data on the operating conditions of agricultural machinery, the status of the farmland environment, high-precision positioning information, and operation quality. It should be noted that: The multi-source sensing and monitoring module includes an agricultural machinery operating condition sensing unit, a farmland environment sensing unit, a high-precision positioning and navigation unit, a multi-modal data fusion unit, and an operation quality monitoring unit, wherein: The agricultural machinery operating condition sensing unit is used to collect the operating status parameters of the core components of agricultural machinery through engine speed sensor, oil pressure sensor and vibration sensor; The farmland environment sensing unit is used to collect soil moisture and crop growth data through soil temperature and humidity sensors, nutrient sensors and multispectral cameras deployed in the field. The high-precision positioning and navigation unit integrates a Beidou / GPS RTK positioning module and an inertial measurement unit to provide agricultural machinery with centimeter-level real-time positioning and heading information; The operation quality monitoring unit is used to evaluate the operation quality indicators of sowing uniformity or harvesting loss rate in real time through a grain loss sensor or a visual sensor. The multimodal data fusion unit uses the Kalman filter algorithm to perform spatiotemporal alignment and fusion processing on the perception data of lidar, camera and millimeter-wave radar to improve the accuracy of obstacle recognition and terrain perception. The cloud-based central control and processing module is used to perform collaborative task scheduling and initial path planning for the agricultural machinery group based on the data uploaded by the multi-source sensing and monitoring module, and to perform operational efficiency analysis and optimization decisions based on the digital twin model. It should be noted that: The cloud-based central control and processing module includes an intelligent collaborative scheduling engine unit, a digital twin and performance optimization unit, and a blockchain-based secure evidence storage and traceability unit, wherein: The intelligent collaborative scheduling engine unit adopts a multi-objective optimization algorithm to comprehensively consider agricultural machinery operation efficiency, energy consumption and equipment lifespan, and performs global task decomposition, agricultural machinery assignment and initial path planning. The multi-objective optimization algorithm adopted by the intelligent collaborative scheduling engine unit is an improved NSGA-II algorithm. The digital twin and efficiency optimization unit is used to construct a three-dimensional virtual model that is mapped to physical farmland and agricultural machinery in real time, and to construct an agricultural machinery energy consumption prediction and operation quality assessment model based on historical and real-time data, and generate a precision operation prescription map for variable fertilization and sowing. The blockchain secure evidence storage and traceability unit adopts a consortium blockchain architecture to perform tamper-proof distributed evidence storage of the hash values ​​of key agricultural machinery operation data and task instructions. The blockchain secure evidence storage and traceability unit also realizes trusted data sharing across entities through smart contracts. The fault emergency and task succession module is used to dynamically reallocate unfinished tasks when agricultural machinery failure or operation interruption is detected, plan emergency paths for replacement agricultural machinery, and schedule backup resources to ensure the continuous execution of operation tasks. It should be noted that the fault emergency response and task succession module includes a fault diagnosis and dynamic task reallocation unit, an emergency path replanning and operation coordination unit, and a backup resource scheduling and task handover unit, wherein: The fault diagnosis and dynamic task reallocation unit is used to determine the fault status of agricultural machinery in real time based on multi-source sensing data, and reallocate unfinished work areas according to the remaining agricultural machinery operation capacity in the cluster. The fault diagnosis and dynamic task reallocation unit adopts an improved K-means++ algorithm that incorporates a penalty factor; The emergency path replanning and operation coordination unit is used to treat the location of the faulty agricultural machinery as a dynamic obstacle when agricultural machinery fails, to plan a conflict-free path for the replacement agricultural machinery, and to ensure seamless connection of the operation area. The backup resource scheduling and task handover unit is used to manage the backup agricultural machinery resource library, automatically schedule backup agricultural machinery to be put into operation when needed, and achieve seamless connection of operation data and status between the faulty agricultural machinery and the replacement agricultural machinery.

[0010] The system also includes a predictive maintenance module, wherein: The predictive maintenance module analyzes historical operating data of agricultural machinery based on a long short-term memory network model, predicts the remaining service life of key components, and generates maintenance warnings.

[0011] The system also includes a visual monitoring and decision support platform, wherein: The platform is used to display the location of agricultural machinery, operation progress, environmental parameters and early warning information in real time, and supports remote task issuance, parameter adjustment and issuance of variable operation control commands based on precision operation prescription maps.

[0012] In the above technical solution, the present invention provides an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things, which has the following beneficial effects: (1) This invention integrates cloud-edge-terminal three-level collaborative architecture to deeply integrate cloud-side global optimization scheduling, edge-side real-time computing control and terminal precision execution. This architecture uses a variety of communication network technologies to ensure the stability and low latency of data transmission in complex farmland environments. Furthermore, the edge computing module not only realizes local data fusion and real-time path planning, but its integrated service migration management function also ensures the continuity of computing tasks when agricultural machinery moves. The system as a whole realizes a closed loop from global intelligent decision-making to local instantaneous response, which greatly improves the real-time performance, adaptability and overall operational stability of agricultural machinery group collaborative operation.

[0013] (2) By integrating high-precision positioning, multimodal environmental sensors, working condition monitoring and operation quality detection units, this invention constructs a full-dimensional, high-precision perception system covering agricultural machinery, environment and operation effect. It uses data fusion technology to generate accurate information on farmland status and comprehensive status of agricultural machinery. At the same time, combined with a visualization monitoring platform, it provides users with a panoramic real-time view of agricultural machinery location, operation progress, environmental parameters and early warning information, and supports remote variable control based on precise operation prescription maps. This enables the operation process to be clearly seen, managed and accurately controlled, providing direct support for refined agricultural management.

[0014] (3) The cloud-based central control processing module of this invention constructs a virtual model that is mapped to the physical world in real time based on digital twin technology, and uses improved NSGA-II and other multi-objective optimization algorithms to perform global task scheduling and path planning, seeking the optimal solution among multiple objectives such as efficiency and energy consumption. In addition, by introducing consortium blockchain technology to store key operation data in an immutable manner, and using smart contracts to achieve trusted sharing, this design combines simulation optimization, intelligent decision-making and trusted traceability, which not only improves the scientificity and efficiency of the operation plan, but also ensures the credibility of the operation data and the traceability of the whole process.

[0015] (4) This invention integrates proactive fault emergency response and predictive maintenance mechanisms. The fault emergency response module can quickly diagnose agricultural machinery failures and dynamically reallocate tasks, plan emergency paths, and schedule backup resources to ensure seamless operation and minimize the impact of interruptions. Meanwhile, the predictive maintenance module analyzes historical operating data and uses models such as long short-term memory networks to predict the lifespan of key components and provide early warnings. These two mechanisms together realize the transformation from passive response to proactive prevention, significantly reducing the risk of unexpected downtime, extending equipment lifespan, and ensuring the highly reliable completion of large-scale collaborative operation tasks and the long-term sustainable operation of the system.

[0016] (5) Through innovative system architecture design and integration of multiple key technologies, a highly reliable, intelligent and collaborative intelligent agricultural machinery operation system has been constructed. This system has brought significant benefits in improving operation efficiency and quality, ensuring operation continuity and safety, realizing data credibility and process traceability, reducing operation and maintenance costs and optimizing user experience, and has powerfully promoted the in-depth development of smart agriculture. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0018] Figure 1 This is a schematic diagram of the system architecture provided for an embodiment of an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings.

[0020] like Figure 1 As shown in the figure, the present invention provides an intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things, including an edge execution computing module, a multi-source sensing and monitoring module, a cloud-based central control and processing module, and a fault emergency and task succession module. The edge execution computing module, the multi-source sensing and monitoring module, the cloud-based central control and processing module, and the fault emergency and task succession module are connected through an Internet of Things communication network. The Internet of Things (IoT) communication network is a wireless wide-area communication network that supports hybrid networking, including at least 5G, 4G, and NB-IoT cellular networks; The edge execution computing module and the multi-source sensing and monitoring module integrate industrial-grade IoT cards, which have a wide operating temperature range of -40℃ to 85℃ and an IP67 or higher protection rating. These cards are used to establish and maintain a stable, low-latency communication link with the cloud-based central control processing module in complex farmland environments.

[0021] The edge execution computing module is deployed on or near the agricultural machinery to perform local fusion and preprocessing of the sensed data, and to perform agricultural machinery collaborative control, local path planning and safety monitoring based on real-time information; It should be noted that: The edge computing module includes a local data fusion and preprocessing unit, a real-time collaborative control and path planning unit, a security threshold monitoring and primary response unit, and a service migration management unit, among which: The local data fusion and preprocessing unit is used to clean, spatiotemporally register and fuse heterogeneous data from the multi-source sensing and monitoring module to generate comprehensive agricultural machinery pose information; The real-time collaborative control and path planning unit is used to run the improved ant colony algorithm or dynamic path optimization algorithm, and to perform local collision avoidance, formation maintenance and dynamic path replanning based on the real-time obstacle and adjacent agricultural machinery positions. The safety threshold monitoring and primary response unit is used to preset multi-level operating thresholds for engine temperature, oil pressure, and operating speed, and to trigger local alarms or protective operations when the monitored data exceeds the limits. The service migration management unit is used to achieve seamless migration of computing tasks and status data when the movement of agricultural machinery causes the switching of its associated edge computing nodes, based on the task-service correlation model.

[0022] The multi-source sensing and monitoring module is used to collect data on the operating conditions of agricultural machinery, the status of the farmland environment, high-precision positioning information, and operation quality. It should be noted that: The multi-source sensing and monitoring module includes an agricultural machinery operating condition sensing unit, a farmland environment sensing unit, a high-precision positioning and navigation unit, a multimodal data fusion unit, and an operation quality monitoring unit, among which: The agricultural machinery operating condition sensing unit is used to collect the operating status parameters of the core components of agricultural machinery through engine speed sensor, oil pressure sensor and vibration sensor; The farmland environment sensing unit is used to collect soil moisture and crop growth data through soil temperature and humidity sensors, nutrient sensors and multispectral cameras deployed in the field. The high-precision positioning and navigation unit integrates a Beidou / GPS RTK positioning module and an inertial measurement unit to provide agricultural machinery with centimeter-level real-time positioning and heading information; The operation quality monitoring unit is used to assess operation quality indicators such as sowing uniformity or harvest loss rate in real time through grain loss sensors or visual sensors. The multimodal data fusion unit uses the Kalman filter algorithm to perform spatiotemporal alignment and fusion processing on the perception data from lidar, cameras and millimeter-wave radar to improve the accuracy of obstacle recognition and terrain perception.

[0023] The cloud-based central control and processing module is used to perform collaborative task scheduling and initial path planning for agricultural machinery groups based on data uploaded by the multi-source sensing and monitoring module, and to perform operational efficiency analysis and optimization decisions based on the digital twin model. It should be noted that: The cloud-based central control and processing module includes an intelligent collaborative scheduling engine unit, a digital twin and performance optimization unit, and a blockchain-based secure evidence storage and traceability unit, among which: The intelligent collaborative scheduling engine unit adopts a multi-objective optimization algorithm to comprehensively consider agricultural machinery operation efficiency, energy consumption, and equipment lifespan, and performs global task decomposition, agricultural machinery assignment, and initial path planning. The multi-objective optimization algorithm used by the intelligent collaborative scheduling engine unit is an improved NSGA-II algorithm. The digital twin and efficiency optimization unit is used to construct a three-dimensional virtual model that maps to physical farmland and agricultural machinery in real time, and to build an agricultural machinery energy consumption prediction and operation quality assessment model based on historical and real-time data, generating a precision operation prescription map for variable fertilization and sowing. The blockchain secure evidence storage and traceability unit adopts a consortium blockchain architecture to perform tamper-proof distributed evidence storage of the hash values ​​of key agricultural machinery operation data and task instructions. The blockchain secure evidence storage and traceability unit also realizes trusted data sharing across entities through smart contracts.

[0024] The fault emergency and task succession module is used to dynamically reallocate unfinished tasks when agricultural machinery failure or operation interruption is detected, plan emergency paths for replacement agricultural machinery, and schedule backup resources to ensure the continuous execution of operation tasks. It should be noted that the fault emergency response and task succession module includes a fault diagnosis and dynamic task reallocation unit, an emergency path replanning and operation coordination unit, and a backup resource scheduling and task handover unit, wherein: The fault diagnosis and dynamic task reallocation unit is used to determine the fault status of agricultural machinery in real time based on multi-source sensing data, and reallocate unfinished work areas according to the remaining agricultural machinery operation capacity in the cluster. The fault diagnosis and dynamic task reallocation unit adopts an improved K-means++ algorithm that incorporates a penalty factor; The emergency route replanning and operation coordination unit is used to treat the location of the faulty agricultural machinery as a dynamic obstacle when agricultural machinery fails, to plan a conflict-free route for the replacement agricultural machinery, and to ensure seamless connection of the operation area. The standby resource scheduling and task handover unit is used to manage the standby agricultural machinery resource library, automatically schedule standby agricultural machinery to be put into operation when needed, and achieve seamless connection of operation data and status between the faulty agricultural machinery and the replacement agricultural machinery.

[0025] The system also includes a predictive maintenance module, in which: The predictive maintenance module analyzes historical operating data of agricultural machinery based on a long short-term memory network model to predict the remaining service life of key components and generate maintenance warnings.

[0026] The system also includes a visual monitoring and decision support platform, in which: The platform is used to display the location of agricultural machinery, operation progress, environmental parameters and early warning information in real time, and supports remote task issuance, parameter adjustment and variable operation control command issuance based on precision operation prescription map.

[0027] In one embodiment of this invention, the multimodal data fusion unit achieves deep fusion of lidar point cloud data, camera image data, and millimeter-wave radar detection data by employing a combination of Kalman filtering and deep learning. Specifically, this includes: first, aligning the timestamps and unifying the spatial coordinates of the multi-source sensing data; then, using a feature extraction network based on deep learning (such as PointNet++, YOLO, etc.), extracting key features such as obstacles, crop rows, and field ridges from the point cloud and images respectively; finally, fusing and estimating the extracted feature information using extended Kalman filtering (EKF) or unscented Kalman filtering (UKF) algorithms to generate a high-precision real-time 3D situation map of the farmland containing information such as obstacle location, type, speed, crop row lines, and work boundaries. This method effectively solves the problem of limited sensing capabilities of a single sensor in complex farmland environments, improving obstacle recognition accuracy to over 95%, and providing a reliable environmental model for precise navigation and obstacle avoidance of agricultural machinery.

[0028] In another embodiment of the present invention, the service migration management unit achieves seamless migration of computing services based on a task-service correlation model and Docker container technology. Its workflow is as follows: First, the unit continuously monitors the location of the agricultural machinery and the signal strength and latency of the edge node currently serving it; when it predicts or detects that the agricultural machinery is about to leave the coverage area of ​​the current node, or when the communication quality is lower than the threshold, the migration process is initiated. Secondly, it analyzes the correlation between all containerized services currently running on the agricultural machinery (such as path planning service, data preprocessing service, obstacle avoidance decision service) and the current task (such as "harvesting area A"), and packages highly correlated services and their runtime states (including memory state, configuration files, intermediate calculation results) into "migration groups". Next, the migration group data is pre-pushed or synchronized to the target edge node via 5G or high-speed local area network; Finally, the service is quickly started and restored the status the moment the agricultural machinery connects to the target node network. The end-to-end latency of the entire process is controlled within 200 milliseconds, ensuring the continuity of agricultural machinery operation and a seamless user experience.

[0029] In another embodiment of the present invention, the intelligent cooperative scheduling engine unit employs an improved NSGA-II algorithm that enhances the traditional NSGA-II algorithm by introducing local search operators (such as simulated annealing or tabu search) to solve multi-objective optimization problems in agricultural machinery cooperative operations. The algorithm solution process is as follows: First, a mathematical model is constructed for the problem of cooperative agricultural machinery operation. The objective function set typically includes minimizing the total operation time, minimizing the total energy consumption (fuel consumption / electricity consumption), maximizing the uniformity of operation quality, and equalizing the load of each agricultural machine. The constraints include the maximum speed of the agricultural machinery, the operating width, the turning radius, the farmland boundary, and the agronomic requirements for operation. Then, the improved NSGA-II algorithm is used to solve the problem: in each generation of evolution, in addition to crossover and mutation operations to generate offspring population, a local search is performed on some excellent individuals in the population with a certain probability to find better solutions in their neighborhood. Finally, through fast non-dominated sorting and crowding calculation, the next generation of the population is selected. After multiple iterations, a Pareto optimal solution set that achieves the best trade-off among multiple objectives is finally obtained. Schedulers can select a solution from this set based on their actual preferences, thereby achieving comprehensive optimization in multiple dimensions such as efficiency, energy consumption, and quality.

[0030] In another embodiment of the present invention, the three-dimensional virtual model of farmland and agricultural machinery constructed by the digital twin and efficiency optimization unit is not only used for visualization, but its core function is to perform "simulation-optimization" iteration. Its workflow is as follows: First, based on Geographic Information System (GIS) data, historical operation data, and real-time sensor data, a high-precision virtual farmland environment is established, including terrain, soil properties, crop distribution, and fixed obstacles. Secondly, the digital model of the physical agricultural machinery (including dynamic parameters, operating mechanism model, and energy consumption model) is placed into this environment. After the intelligent collaborative scheduling engine unit generates a preliminary operation plan (task allocation and reference path), the unit will perform a high-fidelity simulation of the plan in the digital twin environment. During the simulation, the system can accurately calculate the estimated operation time, energy consumption, component wear and tear, and potential risks of collisions or substandard operation quality for each piece of agricultural machinery under the plan. Based on simulation results, performance optimization models (such as machine learning-based regression models or physics-based simulation optimization algorithms) can automatically adjust parameters in the plan, such as fine-tuning the path, reallocating workload, or suggesting better agricultural machinery travel speeds, and then conduct simulation verification again. After several "simulation-evaluation-optimization" cycles, a precise work prescription map with better overall performance, verified in a virtual environment, is finally output and sent to actual agricultural machinery for execution, thereby significantly improving the scientific rigor and reliability of the first round of work planning.

[0031] In another embodiment of this invention, the predictive maintenance module based on Long Short-Term Memory (LSTM) networks is fundamentally designed to construct a health degradation prediction model for key agricultural machinery components (such as engines, transmissions, and hydraulic pumps). The implementation steps are as follows: First, in the data preparation phase, time-series data of historical operating conditions of the agricultural machinery are collected, including vibration spectrum, oil pressure, coolant temperature, load current, etc., and aligned with the time point when the component ultimately failed, forming a training dataset. Second, in the model training phase, a multi-layer LSTM network is constructed. Its input is multi-dimensional operating condition time-series data over a past time window (e.g., the past 24 hours), and its output is a prediction of the probability of the component failing or an estimated remaining useful life (RUL) within a future time period (e.g., the next 72 hours). The model is trained using historical data to learn implicit patterns related to component wear and performance degradation in the operating condition data. Finally, in the online prediction phase, the real-time uploaded operating condition data stream from the agricultural machinery is input into the trained LSTM model, and the model outputs the current component's health score and failure risk warning in real time. When the predicted failure probability exceeds a preset threshold, the system will automatically generate a maintenance work order, notifying maintenance personnel in advance to prepare spare parts and schedule maintenance time windows, thereby realizing the transformation from "post-failure maintenance" to "predictive maintenance" and effectively reducing unplanned downtime.

[0032] In another embodiment of the present invention, the visual monitoring and decision support platform adopts a micro-frontend architecture and WebGL technology to provide users with an immersive monitoring experience. The platform's core functions include: 1) A unified map of the entire field: Based on a GIS map, the system displays in real time icons (different icons distinguish types), real-time locations, driving trajectory arrows, and operation status (sowing, harvesting, idle, faulty, etc.) of all online agricultural machinery, and highlights completed, ongoing, and non-operational field areas with different colors. 2) Virtual and physical linkage monitoring: The platform integrates a digital twin 3D view, allowing users to switch to virtual farm mode with one click, view the operation process of agricultural machinery from any angle, and retrieve real-time video streams from the front-end camera of the agricultural machinery or fixed cameras in the field to achieve direct observation of the physical world. 3) Intelligent Early Warning Dashboard: Centrally displays alarm information at all levels generated by the safety threshold monitoring unit, predictive maintenance module, etc. (such as "Agricultural Machinery A engine temperature warning", "Agricultural Machinery B is expected to need maintenance in 48 hours"), and provides one-click location and details viewing functions; 4) Remote Decision Control Console: The platform provides a user-friendly interface, allowing users to select farmland areas and generate and issue work tasks with a single click. In emergencies, specific agricultural machinery can be directly selected on the map to remotely send control commands such as "emergency stop" and "return to base." For variable-operation tasks, users can view and confirm prescription maps generated by the digital twin and efficiency optimization unit, and directly issue them to designated variable-operation fertilizer applicators or seeders for execution. This platform significantly reduces the management complexity of large-scale collaborative agricultural operations and improves decision-making efficiency and accuracy.

[0033] Working Principle: During operation, the system first comprehensively collects farmland environmental data such as soil and weather conditions, agricultural machinery operating data such as engine and fuel consumption, and high-precision positioning data through a multi-source sensing and monitoring module. This data is stably transmitted to the cloud-based central control processing module and the near-end edge execution computing module via an integrated industrial-grade IoT card and a 5G / 4G / NB-IoT hybrid network. In the cloud, the intelligent collaborative scheduling engine unit, based on improved NSGA-II and other multi-objective optimization algorithms, comprehensively considers global operation tasks, agricultural machinery status, and farmland conditions to generate a preliminary collaborative operation task allocation scheme and a global reference. The path, digital twin, and efficiency optimization unit repeatedly simulates and optimizes these solutions in a 3D virtual model, ultimately generating precise variable-operation prescription maps and optimized operation instructions. Simultaneously, all key operation instructions and data summaries are securely stored and traced via a blockchain-based notarization and traceability unit for immutability. The optimized operation instructions are then sent to the edge execution computing module. This module's real-time collaborative control and path planning unit, combined with high-precision environmental perception information processed locally from the multimodal data fusion unit, runs an improved ant colony algorithm to perform localized, refined dynamic planning and real-time optimization of the global path. Obstacle avoidance control directly drives the agricultural machinery's execution, while safety threshold monitoring and the primary response unit work in parallel to ensure that the machinery's operating parameters are always within a safe range. During operation, if a machine malfunctions, the fault diagnosis and dynamic task redistribution unit determines the problem in real time. The fault emergency and task continuation module is immediately activated, and the dynamic task redistribution unit uses an improved K-means++ algorithm to quickly distribute the remaining tasks of the malfunctioning machine to other machines in the cluster. The emergency path replanning unit replans conflict-free paths for the replacement machine. The backup resource scheduling unit can call upon backup machines, and the service migration management unit ensures that computational tasks... The system ensures continuity of agricultural machinery or network operations. Furthermore, the predictive maintenance module analyzes historical operating data and uses an LSTM model to predict component failure risks in advance. The overall system operation status, including the location of agricultural machinery, operation progress, environmental parameters, and early warning information, is displayed to users in real time through a visual monitoring and decision support platform. This platform also supports remote monitoring and decision intervention. In this way, the system completes a closed loop from full-domain perception to cloud-based intelligent decision-making, to edge real-time control and precise terminal execution, and has the ability to perform fault emergency self-healing and reliable data traceability, thus achieving high efficiency and intelligence in agricultural machinery collaborative operations.

[0034] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. An intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things, comprising an edge execution computing module, a multi-source sensing and monitoring module, a cloud-based central control and processing module, and a fault emergency and task succession module, characterized in that: The edge execution computing module, multi-source sensing and monitoring module, cloud-based central control and processing module, and fault emergency and task succession module are connected through an Internet of Things (IoT) communication network. The edge execution computing module is deployed on or near the agricultural machinery to perform local fusion and preprocessing of the sensed data, and to perform agricultural machinery collaborative control, local path planning and safety monitoring based on real-time information; The multi-source sensing and monitoring module is used to collect data on the operating conditions of agricultural machinery, the status of the farmland environment, high-precision positioning information, and operation quality. The cloud-based central control and processing module is used to perform collaborative task scheduling and initial path planning for the agricultural machinery group based on the data uploaded by the multi-source sensing and monitoring module, and to perform operational efficiency analysis and optimization decisions based on the digital twin model. The fault emergency and task succession module is used to dynamically reallocate unfinished tasks when agricultural machinery failure or operation interruption is detected, plan emergency paths for replacement agricultural machinery, and schedule backup resources to ensure the continuous execution of operation tasks.

2. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The IoT communication network is a wireless wide-area communication network that supports hybrid networking, including at least 5G, 4G and NB-IoT cellular networks; The edge execution computing module and the multi-source sensing and monitoring module integrate an industrial-grade IoT card, which has a wide operating temperature range of -40℃ to 85℃ and an IP67 or higher protection rating. It is used to establish and maintain a stable, low-latency communication link with the cloud-based central control processing module in complex farmland environments.

3. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The edge execution computing module includes a local data fusion and preprocessing unit, a real-time collaborative control and path planning unit, a security threshold monitoring and primary response unit, and a service migration management unit, wherein: The local data fusion and preprocessing unit is used to clean, spatiotemporally register and fuse heterogeneous data from the multi-source sensing and monitoring module to generate comprehensive agricultural machinery pose information. The real-time collaborative control and path planning unit is used to run an improved ant colony algorithm or a dynamic path optimization algorithm, and to perform local collision avoidance, formation maintenance and dynamic path replanning based on real-time obstacles and the positions of adjacent agricultural machinery. The safety threshold monitoring and primary response unit is used to preset multi-level operating thresholds for engine temperature, oil pressure, and operating speed, and to trigger local alarms or protective operations when the monitored data exceeds the limits. The service migration management unit is used to achieve seamless migration of computing tasks and status data when the movement of agricultural machinery causes the switching of its associated edge computing nodes, based on the task-service correlation model.

4. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The multi-source sensing and monitoring module includes an agricultural machinery operating condition sensing unit, a farmland environment sensing unit, and a high-precision positioning and navigation unit, wherein: The agricultural machinery operating condition sensing unit is used to collect the operating status parameters of the core components of agricultural machinery through engine speed sensor, oil pressure sensor and vibration sensor; The farmland environment sensing unit is used to collect soil moisture and crop growth data through soil temperature and humidity sensors, nutrient sensors and multispectral cameras deployed in the field. The high-precision positioning and navigation unit integrates a Beidou / GPS RTK positioning module and an inertial measurement unit to provide agricultural machinery with centimeter-level real-time positioning and heading information.

5. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The multi-source sensing and monitoring module also includes a multi-modal data fusion unit and a work quality monitoring unit, wherein: The operation quality monitoring unit is used to evaluate the operation quality indicators of sowing uniformity or harvesting loss rate in real time through a grain loss sensor or a visual sensor. The multimodal data fusion unit uses the Kalman filter algorithm to perform spatiotemporal alignment and fusion processing on the perception data of lidar, camera and millimeter-wave radar to improve the accuracy of obstacle recognition and terrain perception.

6. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The cloud-based central control and processing module includes an intelligent collaborative scheduling engine unit, a digital twin and performance optimization unit, and a blockchain-based secure evidence storage and traceability unit, wherein: The intelligent collaborative scheduling engine unit uses a multi-objective optimization algorithm to comprehensively consider agricultural machinery operation efficiency, energy consumption, and equipment lifespan, and performs global task decomposition, agricultural machinery assignment, and initial path planning. The digital twin and efficiency optimization unit is used to construct a three-dimensional virtual model that is mapped to physical farmland and agricultural machinery in real time, and to construct an agricultural machinery energy consumption prediction and operation quality assessment model based on historical and real-time data, and generate a precision operation prescription map for variable fertilization and sowing. The blockchain-based secure evidence storage and traceability unit uses a consortium blockchain architecture to perform tamper-proof distributed evidence storage of the hash values ​​of key agricultural machinery operation data and task instructions.

7. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 6, characterized in that, The blockchain-based secure evidence storage and traceability unit also enables trusted data sharing across entities through smart contracts; The intelligent collaborative scheduling engine unit uses an improved NSGA-II algorithm for its multi-objective optimization.

8. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The fault emergency response and task succession module includes a fault diagnosis and dynamic task reallocation unit, an emergency path replanning and operation coordination unit, and a backup resource scheduling and task handover unit, wherein: The fault diagnosis and dynamic task reallocation unit is used to determine the fault status of agricultural machinery in real time based on multi-source sensing data, and reallocate unfinished work areas according to the remaining agricultural machinery operation capacity in the cluster. The fault diagnosis and dynamic task reallocation unit adopts an improved K-means++ algorithm that incorporates a penalty factor; The emergency path replanning and operation coordination unit is used to treat the location of the faulty agricultural machinery as a dynamic obstacle when agricultural machinery fails, to plan a conflict-free path for the replacement agricultural machinery, and to ensure seamless connection of the operation area. The backup resource scheduling and task handover unit is used to manage the backup agricultural machinery resource library, automatically schedule backup agricultural machinery to be put into operation when needed, and achieve seamless connection of operation data and status between the faulty agricultural machinery and the replacement agricultural machinery.

9. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The system also includes a predictive maintenance module, wherein: The predictive maintenance module analyzes historical operating data of agricultural machinery based on a long short-term memory network model, predicts the remaining service life of key components, and generates maintenance warnings.

10. The intelligent agricultural machinery collaborative operation and efficiency optimization system based on the Internet of Things according to claim 1, characterized in that, The system also includes a visual monitoring and decision support platform, wherein: The platform is used to display the location of agricultural machinery, operation progress, environmental parameters and early warning information in real time, and supports remote task issuance, parameter adjustment and issuance of variable operation control commands based on precision operation prescription maps.

Images