Agricultural robot computing task collaboration method and system

Abstract

The invention relates to the technical field of agricultural robots, in particular to an agricultural robot computing task collaboration method and system, and the method comprises the following steps: S1, environment perception and data collection: collecting environment perception data of a farmland operation region in real time; s2, generating a dynamic task priority queue: generating the dynamic task priority queue; s3, heterogeneous computing resource adaptation model construction: constructing a heterogeneous computing resource adaptation model, and outputting a task-resource matching matrix; s4, generating an initial task allocation scheme: generating the initial task allocation scheme; s5, environment interference monitoring and compensation: adjusting the initial task allocation scheme; and S6, task execution and queue updating: executing the adjusted initial task allocation scheme, and continuously updating the dynamic task priority queue. According to the method, the working efficiency can be remarkably improved while the working precision is ensured, the working time is shortened, the resource allocation is optimized, and the manual intervention is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of agricultural robots, and particularly to a method and system for collaborative computing tasks of agricultural robots. Background Art

[0002] With the advancement of agricultural modernization, intelligent agriculture has become an important means to improve production efficiency, reduce resource waste, and ensure crop health. Agricultural robots, as one of the core devices of intelligent agriculture, can perform precise farm operations, such as tasks like automatic irrigation, fertilization, and pest control. To improve operation efficiency and precision, agricultural robots need to perform task allocation and optimized scheduling based on real-time environmental data. This requires the system to have a strong environmental perception ability, be able to collect information such as soil humidity, crop growth status, and the positions of the robot cluster in real time, and reasonably allocate tasks based on this data to ensure the smooth progress of operations.

[0003] Existing agricultural robot task allocation methods often adopt fixed task allocation strategies and cannot dynamically adjust task priorities according to environmental changes or the real-time needs of the farmland. Most existing methods ignore the influence of environmental interference factors, such as communication delays, weather changes, etc., which may lead to delays or errors in task execution. At the same time, existing task scheduling methods usually lack optimized allocation of heterogeneous computing resources, and the matching between tasks and computing resources is not precise enough, resulting in waste of robot resources or unbalanced loads. In addition, existing technologies often do not implement a distributed decision-making mechanism for multi-robot collaboration and cannot fully utilize the collaborative operation potential of the robot cluster, resulting in low task completion efficiency and insufficient resource utilization.

[0004] The purpose of the present invention is to provide a method and system for collaborative computing tasks of agricultural robots, reduce the influence of interference on task execution, improve the real-time performance and precision of task scheduling, thereby enhancing the operation efficiency of agricultural robots, reducing resource waste, and ensuring the stability and continuity of operations. Summary of the Invention

[0005] The present invention provides a method and system for collaborative computing tasks of agricultural robots.

[0006] A method for collaborative computing tasks of agricultural robots includes the following steps:

[0007] S1, environmental perception and data collection: Real-time collect environmental perception data in the farm operation area, including soil humidity distribution data, crop growth status data, and real-time position coordinates of the robot cluster;

[0008] S2, Generation of Dynamic Task Priority Queue: Generate a dynamic task priority queue based on soil humidity distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters;

[0009] S3, Construction of Heterogeneous Computing Resource Adaptation Model: According to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, construct a heterogeneous computing resource adaptation model and output a task-resource matching matrix;

[0010] S4, Generation of Initial Task Allocation Scheme: Based on the output task-resource matching matrix, use a distributed consensus mechanism to generate an initial task allocation scheme;

[0011] S5, Environmental Interference Monitoring and Compensation: Continuously monitor the farmland environmental interference factors in real time. When it is detected that the communication delay parameter exceeds the preset threshold, start the environmental interference compensation algorithm to adjust the initial task allocation scheme;

[0012] S6, Task Execution and Queue Update: Execute the adjusted initial task allocation scheme and continuously update the dynamic task priority queue.

[0013] Optionally, the environmental perception and data collection in S1 include:

[0014] S11, Collection of Soil Humidity Distribution Data: Real-time collect the soil humidity data of each area of the farmland through soil humidity sensors;

[0015] S12, Collection of Crop Growth Status Data: Use crop growth monitoring equipment (growth sensors) to collect crop growth status data in real time and adjust the monitoring parameters according to different growth stages of the crops;

[0016] S13, Collection of Robot Cluster Location: Real-time collect the location information of the robot cluster through a positioning system (GPS).

[0017] Optionally, the generation of the dynamic task priority queue in S2 includes:

[0018] S21, Processing of Soil Humidity Data: According to the real-time collected soil humidity distribution data, calculate the soil humidity values of each farmland area through the bilinear interpolation algorithm to obtain the spatial distribution map of soil humidity;

[0019] S22, Processing of Crop Growth Status Data: Evaluate the health status of the crops based on the growth data collected by the crop growth status monitoring equipment and combined with the crop variety and growth stage;

[0020] S23, Calculation of Task Priority Parameters: Based on the soil humidity and crop growth status data, calculate the irrigation task priority parameter P irrigation, the priority parameter P of the fertilization task fertilization and the priority parameter P of the pest control task pestcontrol ;

[0021] S24, generation of the dynamic task priority queue: Based on the priority parameters of the irrigation, fertilization, and pest control tasks, construct a dynamic task priority queue Q tasks .

[0022] Optionally, the construction of the heterogeneous computing resource adaptation model in S3 includes:

[0023] S31, parsing of the dynamic task priority queue: Parse the generated dynamic task priority queue Q tasks ={T1,T2,...,T n}), where each task T i includes the task type (irrigation, fertilization, pest control) and the priority parameter P(T i ), and the task queue is sorted from high to low according to the priority;

[0024] S32, real-time position acquisition and calculation of the robot cluster: Real-time acquisition of the position coordinates of the robot cluster R ={(x1,y1),(x2,y2),...,(x m ,y m}), where m is the number of robots, and (x i ,y i ) is the position coordinate of the i-th robot in the real-time environment;

[0025] S33, construction of the heterogeneous computing resource adaptation model: Construct a heterogeneous computing resource adaptation model, and calculate the matching degree M(T i ,R j ) between each task and the computing resource (robot cluster) according to the task type, priority, and real-time position coordinates of the robot cluster;

[0026] S34, generation of the task-resource matching matrix: Generate a task-resource matching matrix M i ,R j ) according to the matching degree M(T task-resource .

[0027] Optionally, the generation of the initial task allocation plan in S4 includes:

[0028] S41, input of the task-resource matching matrix: Use the task-resource matching matrix M task-resource generated by the heterogeneous computing resource adaptation model as the input;

[0029] S42, optimization objective of the allocation plan: Define the optimization objective as maximizing the task priority and resource utilization rate;

[0030] S43, Information Exchange and Weight Update in the Consensus Mechanism: In the distributed consensus mechanism, each robot node R j sends messages to adjacent robot nodes according to its task matching degree and current local information, and the weight w j (t) of each robot node is updated in each iteration;

[0031] S44, Generation of the Initial Task Assignment Scheme: After the consensus mechanism is completed, an initial task assignment scheme x initial = {x ij} is generated by evaluating the task matching degree and the assigned weight of each robot, where x ij = 1 means that task T i is assigned to robot R j , otherwise x ij = 0.

[0032] Optionally, the environmental interference monitoring and compensation in S5 includes:

[0033] S51, Monitoring of Environmental Interference Factors: The communication delay parameter δ comm in the farmland environment is monitored in real time. By comparing the communication delay parameter δ comm with the delay threshold δ thresh , it is judged whether there is communication interference;

[0034] S52, Task Assignment Adjustment: In the case where the communication delay exceeds the threshold, the environmental interference compensation algorithm is started, and the impact of interference on task execution is reduced by adjusting the initial task assignment scheme.

[0035] Optionally, the monitoring of environmental interference factors in S51 includes:

[0036] S511, Collection of Environmental Interference Factors: The communication delay parameter δ comm in the farmland environment is monitored in real time through sensors, including the time interval for sending and receiving data;

[0037] S512, Comparison of Delay Parameter with Preset Threshold: After the communication delay parameter is collected, it is compared with the preset delay threshold δ thresh . When δ comm > δ thresh , it is determined that there is communication interference.

[0038] Optionally, the task assignment adjustment in S52 includes:

[0039] S521, Interference Judgment: Based on the results of environmental interference factor monitoring, it is judged whether there is communication interference;

[0040] S522, Start the environmental interference compensation algorithm: When it is determined that there is communication interference, start the environmental interference compensation algorithm to adjust the initial task allocation scheme, and the adjustment process is optimized based on the relationship between communication delay and task priority.

[0041] Optionally, the task execution and queue update in S6 include:

[0042] S61, Execute the adjusted task allocation scheme: After the environmental interference compensation algorithm is started, the adjusted initial task allocation scheme is executed, and the robot starts to execute tasks according to the adjusted task order and priority according to the new allocation scheme;

[0043] S62, Dynamic task priority queue update: During the task execution process, continuously monitor the farmland environment and operation progress, and update the dynamic task priority queue in real time. As tasks are completed and new environmental data is collected (such as soil humidity, crop growth status, etc.), automatically adjust the task priority parameters in the queue.

[0044] An agricultural robot computing task collaboration system for implementing the above-mentioned agricultural robot computing task collaboration method, including the following modules:

[0045] Environmental perception and data acquisition module: Real-time acquisition of environmental perception data in the farmland operation area, including soil humidity distribution data, crop growth status data, and real-time position coordinates of the robot cluster;

[0046] Dynamic task priority queue generation module: Generate a dynamic task priority queue based on soil humidity distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters;

[0047] Heterogeneous computing resource adaptation model construction module: According to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, construct a heterogeneous computing resource adaptation model and output a task-resource matching matrix;

[0048] Initial task allocation scheme generation module: Based on the output task-resource matching matrix, use a distributed consensus mechanism to generate an initial task allocation scheme;

[0049] Environmental interference monitoring and compensation module: Real-time monitoring of farmland environmental interference factors. When the detected communication delay parameter exceeds the preset threshold, start the environmental interference compensation algorithm to adjust the initial task allocation scheme;

[0050] Task execution and queue update module: Execute the adjusted initial task allocation scheme and continuously update the dynamic task priority queue.

[0051] Advantages of the present invention:

[0052] In the present invention, by combining the real-time collected soil humidity, crop growth status data and the location information of the robot cluster, the agricultural robot computing task collaboration method can accurately perceive the farmland environment and optimize the task priorities according to the real-time changes. This dynamic adjustment based on environmental data enables the agricultural robot to execute tasks such as irrigation, fertilization, and pest control more precisely, avoiding resource waste caused by the failure to timely reflect the changes in crop growth stages or soil humidity. In addition, by dynamically adjusting the task priorities, the system can significantly improve the working efficiency while ensuring the operation accuracy, reduce the operation time, optimize the resource allocation, and reduce the manual intervention.

[0053] In the present invention, through the heterogeneous computing resource adaptation model, the task priorities, the robot cluster location and the computing resources are matched to ensure that each task can be executed by the most suitable robot, thereby maximizing the utilization rate of resources. By constructing a task-resource matching matrix and optimizing the task allocation through a distributed consensus mechanism, the system can achieve the most reasonable resource allocation in the robot cluster, avoiding uneven task allocation and excessive waste of computing resources, and enhancing the collaborative operation efficiency of the robots. This optimization method improves the intelligent level of agricultural production and ensures that high-priority tasks can be completed in a timely and efficient manner.

[0054] In the present invention, by introducing an environmental interference monitoring and compensation mechanism, especially under the influence of environmental uncertainty factors such as communication delays, the interference factors can be detected in a timely manner, and the task allocation scheme can be adjusted through a compensation algorithm to reduce the negative impact of the interference on the operation. It ensures that even in the case of high communication delays or large environmental changes, the tasks of the agricultural robot can still be successfully completed in the order of priorities, avoiding operation delays and resource waste caused by communication problems or environmental interference. At the same time, dynamically updating the task priority queue ensures that subsequent tasks can be reasonably arranged according to the latest environmental data and operation progress, improving the operation stability and continuity, and enhancing the overall reliability and adaptability of the agricultural robot system. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only those of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0056] Figure 1 It is a schematic flow chart of the collaboration method according to the embodiment of the present invention;

[0057] Figure 2 It is a schematic diagram of the system function modules according to the embodiment of the present invention. Detailed implementation manners

[0058] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted here that, in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments. For some well-known technologies, those skilled in the art can also adopt other alternative ways for implementation; moreover, the accompanying drawings are only for more specifically describing the embodiments and are not intended to specifically limit the present invention.

[0059] It should be pointed out that in the specification, when referring to "an embodiment", "embodiment", "exemplary embodiment", "some embodiments", etc., it indicates that the described embodiment may include specific features, structures or characteristics, but not necessarily every embodiment includes the specific feature, structure or characteristic. Additionally, when combining an embodiment to describe a specific feature, structure or characteristic, implementing such a feature, structure or characteristic in combination with other embodiments (whether explicitly described or not) should be within the knowledge of those skilled in the relevant art.

[0060] Generally, the terms can be understood at least in part from their use in the context. For example, at least in part depending on the context, the term "one or more" used herein can be used to describe any feature, structure or characteristic in a singular sense, or can be used to describe a combination of features, structures or characteristics in a plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey a set of exclusive factors, but instead, at least in part depending on the context, allowing for the existence of other factors that may not be explicitly described.

[0061] As Figure 1 shown, a collaborative method for agricultural robot computing tasks includes the following steps:

[0062] S1, Environmental perception and data collection: Real-time collect environmental perception data of the farm operation area, including soil moisture distribution data, crop growth status data, and real-time position coordinates of the robot cluster;

[0063] S2, Generation of a dynamic task priority queue: Generate a dynamic task priority queue based on the soil moisture distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters;

[0064] S3, Construction of a heterogeneous computing resource adaptation model: According to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, construct a heterogeneous computing resource adaptation model and output a task-resource matching matrix;

[0065] S4, Generation of an initial task allocation scheme: Based on the output task-resource matching matrix, generate an initial task allocation scheme using a distributed consensus mechanism;

[0066] S5, Environmental interference monitoring and compensation: Real-time monitor the environmental interference factors in the farmland. When the detected communication delay parameter exceeds the preset threshold, start the environmental interference compensation algorithm to adjust the initial task allocation plan.

[0067] S6, Task execution and queue update: Execute the adjusted initial task allocation plan and continuously update the dynamic task priority queue.

[0068] Through the above content, the optimization and coordination of task allocation are effectively realized, ensuring that the robot can execute agricultural operations efficiently and accurately. By constructing a heterogeneous computing resource adaptation model and adopting a distributed consensus mechanism, tasks and resources can be intelligently matched, improving the operation efficiency and reducing resource waste. At the same time, the environmental interference monitoring and compensation mechanism ensure that under the influence of uncertain factors, the task allocation plan can be adjusted in a timely manner, guaranteeing the stability and continuity of the operation, improving the agricultural operation efficiency, reducing manual intervention, and enhancing resource utilization.

[0069] The environmental perception and data collection in S1 include:

[0070] S11, Soil moisture distribution data collection: Real-time collect the soil moisture data of each area in the farmland through soil moisture sensors.

[0071] S12, Crop growth status data collection: Use crop growth monitoring equipment (growth sensors) to real-time collect the crop growth status data and adjust the monitoring parameters according to different growth stages of the crops.

[0072] S13, Robot cluster position collection: Real-time collect the position information of the robot cluster through a positioning system (GPS).

[0073] Through the above content, by comprehensively using soil moisture sensors, crop growth monitoring equipment, and positioning systems, key information of the farmland operation area can be obtained in real-time and comprehensively. The soil moisture data helps to accurately master the soil moisture condition, the crop growth status data provides real-time feedback on the growth process of the crops, and the real-time collection of the robot cluster position ensures the precise scheduling and coordination of tasks. Through the real-time acquisition of these multi-dimensional data, the operation accuracy and efficiency are improved.

[0074] The generation of the dynamic task priority queue in S2 includes:

[0075] S21, Soil moisture data processing: According to the real-time collected soil moisture distribution data, calculate the soil moisture values of each farmland area through the bilinear interpolation algorithm to obtain the spatial distribution map of soil moisture, expressed as:

[0076]

[0077] Among them, (x, y) are the coordinates of the target point to be interpolated, H1, H2, H3, and H4 are known soil moisture data values, and (x1, y1), (x2, y2), (x3, y3), and (x4, y4) are the coordinates of four known points;

[0078] S22, Crop growth status data processing: According to the growth data collected by the crop growth status monitoring device, combined with the crop variety and growth stage, evaluate the health status of the crop, expressed as:

[0079] C(x, y, t) = α·NDVI(x, y, t) + β·age + γ·T;

[0080] Among them, C(x, y, t) is the crop health status, α, β, and γ are weight factors to be adjusted, NDVI(x, y, t) is the normalized difference vegetation index value at position (x, y) at time t, reflecting the photosynthesis and health status of the crop, age is the crop growth stage parameter, and T is the temperature data;

[0081] S23, Task priority parameter calculation: Based on the soil moisture and crop growth status data, calculate the irrigation task priority parameter P irrigation 、fertilization task priority parameter P fertilization and the priority parameter P pestcontrol of the pest control task, expressed as:

[0082]

[0083] Among them, k1, k2, and k3 are adjustment factors, and H threshold 、C threshold are the soil moisture threshold and the crop health status threshold respectively;

[0084] The soil moisture threshold H threshold 、the crop health status threshold C threshold are set based on historical data, specifically including:

[0085] Collect historical data: Collect farmland data for multiple seasons or periods, including soil moisture and crop health status data;

[0086] Threshold setting: Set the threshold according to the mean and standard deviation of historical data, expressed as:

[0087] H threshold = μ(H) - k1·σ(H);

[0088] C threshold = μ(C) - k2·σ(C);

[0089] Among them, μ(H) is the mean of the soil humidity data H, σ(H) is the standard deviation of the soil humidity data H, k1 and k2 are adjustment factors, μ(C) is the mean of the crop health status data C, and σ(C) is the standard deviation of the crop health status data C;

[0090] S24, Dynamic task priority queue generation: Based on the priority parameters of irrigation, fertilization, and pest control tasks, construct a dynamic task priority queue Q tasks , expressed as:

[0091] Q tasks = sort(P irrigation , P fertilization , P pestcontrol );

[0092] Among them, sort is to sort from high to low according to the priority to generate the final task priority queue;

[0093] Through the above content, the intelligent scheduling and priority optimization of agricultural robot tasks can be realized. This method dynamically adjusts the execution order of various tasks by real-time monitoring of the environmental changes in the farmland, ensuring that tasks such as irrigation, fertilization, and pest control can be preferentially executed according to actual needs, avoiding resource waste and lag in crop growth. By setting the thresholds of soil humidity and crop health status, the system can promptly respond to environmental changes, accurately allocate the operations of the robot cluster, improve agricultural production efficiency, reduce manual intervention, increase resource utilization rate, and ensure the healthy growth of crops.

[0094] The construction of the heterogeneous computing resource adaptation model in S3 includes:

[0095] S31, Dynamic task priority queue parsing: Parse the generated dynamic task priority queue Q tasks = {T1, T2,..., T n}, where each task T i includes the task type (irrigation, fertilization, pest control) and the priority parameter P(T i ), and the task queue is sorted from high to low according to the priority. Tasks with higher priorities will be allocated more computing resources;

[0096] S32, Real-time position acquisition and calculation of the robot cluster: Real-time acquire the position coordinates of the robot cluster R = {(x1, y1), (x2, y2),..., (x m , y m ), where m is the number of robots, and (x i , y i ) is the position coordinate of the i-th robot in the real-time environment;

[0097] S33, Heterogeneous Computing Resource Adaptation Model Construction: Construct a heterogeneous computing resource adaptation model, and calculate the matching degree M(T i ,R j ) between each task and the computing resources (robot cluster) according to the task type, priority, and the real-time position coordinates of the robot cluster, expressed as:

[0098]

[0099] where M(T i ,R j ) is the matching degree between each task T i and each robot R j , d(T i ,R j ) is the distance between task T i and robot R j , ∈ is a small constant to prevent division by zero error, C(R j ) is the remaining computing power or battery power of robot R j , and C max is the maximum computing power or battery power of the robot;

[0100]

[0101] where is the target area coordinate of task T i , is the position coordinate of robot R j ;

[0102] S34, Task-Resource Matching Matrix Generation: Generate a task-resource matching matrix M i ,R j ) according to the matching degree M(T task-resource between each task and the robot, expressed as:

[0103]

[0104] where M task - resource is a matrix of size n×m, n is the number of tasks, m is the number of robots, and the elements in the matrix represent the matching degree between tasks and robots;

[0105] Through the above, the precise matching of tasks and robot resources is achieved, maximizing the task execution efficiency. Multiple factors such as the priority of tasks, the distance between tasks and robots, and the remaining computing power or battery level of robots are considered to ensure that high-priority tasks are executed by robots with the most abundant resources and optimal positions. This not only improves the collaborative operation efficiency of agricultural robots but also optimizes resource allocation, reduces energy consumption and operation latency, and enhances the overall automation and intelligence level of agricultural production.

[0106] The generation of the initial task allocation plan in S4 includes:

[0107] S41, Task-Resource Matching Matrix Input: The task-resource matching matrix M generated by the heterogeneous computing resource adaptation model is used as the input. task-resource As the input;

[0108] S42, Optimization Goal of the Allocation Plan: Define the optimization goal as maximizing task priority and resource utilization rate, ensuring that high-priority tasks are executed by appropriate robots, and minimizing the imbalance of the burden between tasks and robots. The objective function f is expressed as: opt Expressed as:

[0109]

[0110] Among them, x ij is the allocation variable. When the task T i is assigned to the robot R j , x ij = 1, otherwise x ij = 0, P(T i ) is the priority of the task T i , n is the total number of tasks, and m is the total number of robots;

[0111] S43, Information Exchange and Weight Update in the Consensus Mechanism: In the distributed consensus mechanism, each robot node R j sends messages to adjacent robot nodes according to its task matching degree and current local information. The weight w j (t) of each robot node is updated in each round of iteration and is expressed as:

[0112]

[0113] Among them, w j (t) is the weight of the robot R j at the t-th round of iteration, w j (t + 1) is the weight of the robot R j at the (t + 1)-th round of iteration, α ′ is the adjustment factor, and x ij (t) is the task T in the t-th round of iteration iWhether it is assigned to robot R j ;

[0114] S44, Initial task assignment plan generation: After the consensus mechanism is completed, an initial task assignment plan x is generated by evaluating the task matching degree and assignment weight of each robot initial ={x ij}, where x ij =1 indicates that task T i is assigned to robot R j , otherwise x ij =0;

[0115] Through the above content, the flexibility and stability of task assignment can be effectively improved. This method fully considers the proximity relationship between robots, ensures that the task assignment is more in line with the actual environment, makes the task coordination more efficient and accurate. In addition, information sharing and cooperation between adjacent robots help to reduce the communication burden of the system, and at the same time can improve the real-time performance and response speed of local task scheduling. Through multiple rounds of iteration of the consensus mechanism, the balance of task priority and resource utilization is finally achieved.

[0116] The environmental interference monitoring and compensation in S5 include:

[0117] S51, Environmental interference factor monitoring: Real-time monitor the communication delay parameter δ in the farmland environment comm , and compare the communication delay parameter δ comm with the delay threshold δ thresh to determine whether there is communication interference;

[0118] S52, Task assignment adjustment: In the case where the communication delay exceeds the threshold, start the environmental interference compensation algorithm, and reduce the impact of interference on task execution by adjusting the initial task assignment plan;

[0119] Through the above content, the negative impact of environmental interference on task execution can be effectively reduced. This mechanism ensures that when the communication delay exceeds the threshold, the task assignment plan is adjusted in time to avoid resource waste and operation delay caused by communication problems, thereby improving the robustness and reliability of the system. By dynamically adjusting the task priority and execution order, the system can maintain a high operation efficiency under different environmental conditions and ensure the continuity and accuracy of agricultural robot operations.

[0120] The environmental interference factor monitoring in S51 includes:

[0121] S511, Environmental interference factor collection: Real-time monitor the communication delay parameter δ in the farmland environment through sensors comm , including the time interval for sending and receiving data;

[0122] S512. Compare the delay parameter with a preset threshold: After collecting the communication delay parameter, compare it with the preset delay threshold δ thresh for comparison. When δ comm > δ thresh , it is determined that there is communication interference;

[0123] The delay threshold δ thresh is set based on historical data, specifically including:

[0124] Collect historical communication delay data: Collect historical communication delay data over a period of time where t represents the time point, and the data set includes the delay information of multiple communication nodes in the farmland environment at different time periods;

[0125] Calculate the statistical characteristics of the delay data: According to the historical communication delay data calculate the statistical characteristics of the delay (mean μ δ and standard deviation σ δ ), expressed as:

[0126]

[0127] where S is the total number of historical data, is the communication delay data at the t-th moment;

[0128] Threshold setting: Based on the mean and standard deviation of the historical data, set the communication delay threshold δ thresh , expressed as:

[0129] δ thresh = μ δ + k·σ δ ;

[0130] where δ thresh is the set communication delay threshold, and k is the adjustment coefficient;

[0131] Through the above content, it is possible to accurately identify whether the communication delay exceeds the range tolerated by the system, thereby timely discovering communication interference problems, being able to dynamically adapt to changes under different environmental conditions, improving the sensitivity of the system to the instability of the communication network. By real-time monitoring and comparing with the preset threshold, the system can effectively avoid task execution problems caused by communication delays and ensure the timeliness and execution efficiency of task allocation.

[0132] The task assignment adjustment in S52 includes:

[0133] S521. Interference determination: Based on the results of environmental interference factor monitoring, determine whether there is communication interference;

[0134] S522. Start the environmental interference compensation algorithm: When it is determined that there is communication interference, start the environmental interference compensation algorithm to adjust the initial task allocation plan to reduce the impact of interference on task execution. The adjustment process is optimized based on the relationship between communication delay and task priority to ensure that high-priority tasks are executed first and the remaining tasks are reasonably allocated, expressed as:

[0135]

[0136] where x adjusted is the adjusted task allocation plan, x initial is the initial task allocation plan, β ′ is the interference compensation coefficient, controlling the sensitivity of task adjustment, δ comm is the current communication delay, δ thresh is the threshold of communication delay;

[0137] Through the above, by starting the environmental interference compensation algorithm, the task allocation can be adjusted in real time, thus effectively reducing the impact of communication interference on task execution. This adjustment mechanism ensures that high-priority tasks can be preferentially processed in a delayed environment, guaranteeing the timeliness and stability of task execution, being able to dynamically respond to the instability of the communication network, optimizing resource allocation and task scheduling, and improving the efficiency and reliability of agricultural robot operations. At the same time, this mechanism minimizes the job delays or resource waste caused by environmental interference through flexible task adjustment.

[0138] The task execution and queue update in S6 include:

[0139] S61. Execute the adjusted task allocation plan: After the environmental interference compensation algorithm is started, the adjusted initial task allocation plan is executed, and the robot starts to execute tasks according to the adjusted task order and priority according to the new allocation plan;

[0140] S62. Dynamic task priority queue update: During the task execution process, continuously monitor the farmland environment and operation progress, and update the dynamic task priority queue in real time. As tasks are completed and new environmental data is collected (such as soil humidity, crop growth status, etc.), automatically adjust the task priority parameters in the queue;

[0141] Through the above, by executing the adjusted task allocation plan and continuously updating the dynamic task priority queue, it is possible to respond in real time to the changes and fluctuations of environmental conditions that occur during task execution, ensuring that tasks are always executed in the optimal priority order, thereby improving resource utilization and task completion efficiency. As the task progresses and environmental data is continuously updated, the priority queue will be automatically adjusted, enabling subsequent tasks to be reasonably allocated according to the latest requirements and priorities.

[0142] As shown in Figure 2 the figure, an agricultural robot computing task collaboration system is used to implement the above-mentioned agricultural robot computing task collaboration method, including the following modules:

[0143] Environment perception and data acquisition module: Real-time acquisition of environment perception data in the farm operation area, including soil humidity distribution data, crop growth status data, and real-time position coordinates of the robot cluster;

[0144] Dynamic task priority queue generation module: Generate a dynamic task priority queue based on soil humidity distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters;

[0145] Heterogeneous computing resource adaptation model construction module: Construct a heterogeneous computing resource adaptation model according to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, and output a task-resource matching matrix;

[0146] Initial task allocation scheme generation module: Generate an initial task allocation scheme based on the output task-resource matching matrix by using a distributed consensus mechanism;

[0147] Environment interference monitoring and compensation module: Real-time monitor the farm environment interference factors. When it is detected that the communication delay parameter exceeds the preset threshold, start the environment interference compensation algorithm to adjust the initial task allocation scheme;

[0148] Task execution and queue update module: Execute the adjusted initial task allocation scheme and continuously update the dynamic task priority queue.

[0149] The present invention covers any alternatives, modifications, equivalent methods, and solutions made within the spirit and scope of the present invention. To enable the public to have a thorough understanding of the present invention, specific details are described in detail in the following preferred embodiments of the present invention. However, those skilled in the art can fully understand the present invention without these detailed descriptions. In addition, well-known methods, processes, procedures, components, and circuits are not described in detail to avoid unnecessary confusion to the essence of the present invention.

[0150] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A collaborative method for agricultural robot computing tasks, characterized in that It includes the following steps: S1, Environmental perception and data collection: Real-time collect environmental perception data of the farm operation area, including soil humidity distribution data, crop growth status data, and real-time position coordinates of the robot cluster; S2, Generation of dynamic task priority queue: Generate a dynamic task priority queue based on the soil humidity distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters; S3, Construction of heterogeneous computing resource adaptation model: According to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, construct a heterogeneous computing resource adaptation model and output a task-resource matching matrix; S4, Generation of initial task allocation plan: Based on the output task-resource matching matrix, use a distributed consensus mechanism to generate an initial task allocation plan; S5, Environmental interference monitoring and compensation: Real-time monitor farm environmental interference factors. When it is detected that the communication delay parameter exceeds the preset threshold, start the environmental interference compensation algorithm to adjust the initial task allocation plan; S6, Task execution and queue update: Execute the adjusted initial task allocation plan and continuously update the dynamic task priority queue.

2. The collaborative method for agricultural robot computing tasks according to claim 1, wherein The environmental perception and data collection in S1 include: S11, Soil humidity distribution data collection: Real-time collect soil humidity data of each area of the farm through soil humidity sensors; S12, Crop growth status data collection: Use crop growth monitoring equipment to real-time collect crop growth status data and adjust monitoring parameters according to different growth stages of the crops; S13, Robot cluster position collection: Real-time collect the position information of the robot cluster through a positioning system.

3. The collaborative method for agricultural robot computing tasks according to claim 1, wherein The generation of the dynamic task priority queue in S2 includes: S21, Soil humidity data processing: According to the real-time collected soil humidity distribution data, calculate the soil humidity values of each farm area through a bilinear interpolation algorithm to obtain a spatial distribution map of soil humidity; S22, Crop growth status data processing: According to the growth data collected by the crop growth status monitoring equipment, combined with the crop variety and growth stage, evaluate the health status of the crops; S23, Task priority parameter calculation: Based on soil humidity and crop growth status data, calculate the irrigation task priority parameter P irrigation , fertilization task priority parameter P fertilization and the priority parameter P of the pest control task pestcontrol ; S24, Generation of dynamic task priority queue: Based on the priority parameters of irrigation, fertilization, and pest control tasks, construct a dynamic task priority queue Q tasks .

4. The collaborative method for agricultural robot computing tasks according to claim 3, wherein The construction of the heterogeneous computing resource adaptation model in S3 includes: S31, Parsing of the dynamic task priority queue: Parse the generated dynamic task priority queue Q tasks ={T1, T2,..., T n}, where each task T i includes a task type and a priority parameter P(T i ), and the task queue is sorted from highest to lowest priority; S32, Real-time Position Acquisition and Calculation of Robot Cluster: Real-time acquisition of the position coordinates of the robot cluster R = {(x1, y1), (x2, y2),..., (x m , y m ), where m is the number of robots, and (x i , y i ) is the position coordinate of the i-th robot in the real-time environment; S33, Heterogeneous computing resource adaptation model construction: Construct a heterogeneous computing resource adaptation model, and calculate the matching degree M(T i ,R j ) between each task and computing resource according to the task type, priority, and real-time position coordinates of the robot cluster; S34, Task-Resource Matching Matrix Generation: Based on the matching degree M(T i , R j ) between each task and the robot, generate the task-resource matching matrix M task-resource .

5. A collaborative method for agricultural robot computing tasks according to claim 4, characterized in that, The generation of the initial task allocation plan in S4 includes: S41, Task-Resource Matching Matrix Input: The task-resource matching matrix M generated by the heterogeneous computing resource adaptation model is used as the input; task-resource as the input; S42, Allocation plan optimization objective: Define the optimization objective as maximizing task priority and resource utilization; S43, Information Exchange and Weight Update in the Consensus Mechanism: In the distributed consensus mechanism, each robot node R j sends messages to adjacent robot nodes according to its task matching degree and current local information, and the weight w j (t) of each robot node is updated in each round of iteration; S44, Initial task assignment plan generation: After the consensus mechanism is completed, an initial task assignment plan x is generated by evaluating the task matching degree and assignment weight of each robot. initial ={x ij}, where x ij =1 indicates that task T i is assigned to robot R j , otherwise x ij =0.

6. The collaborative method for agricultural robot computing tasks according to claim 5, characterized in that, The environmental interference monitoring and compensation in S5 includes: S51, Environmental interference factor monitoring: Real-time monitoring of the communication delay parameter δ in the farmland environment comm , based on the communication delay parameter δ comm and comparing it with the delay threshold δ thresh to determine whether there is communication interference; S52, Task allocation adjustment: In the case where the communication delay exceeds the threshold, start the environmental interference compensation algorithm. By adjusting the initial task allocation plan, reduce the impact of interference on task execution.

7. A collaborative method for agricultural robot computing tasks according to claim 6, characterized in that The environmental interference factor monitoring in S51 includes: S511, Environmental interference factor collection: The communication delay parameter δ in the farmland environment is monitored in real time through sensors comm , including the time interval for sending and receiving data; S512, Comparing the delay parameter with a preset threshold: After collecting the communication delay parameter, compare it with the preset delay threshold δ thresh for comparison. When δ comm > δ thresh , it is determined that there is communication interference.

8. The collaborative method for agricultural robot computing tasks according to claim 7, wherein The task allocation adjustment in S52 includes: S521, Interference determination: Based on the results of environmental interference factor monitoring, determine whether there is communication interference; S522, Start the environmental interference compensation algorithm: When it is determined that there is communication interference, start the environmental interference compensation algorithm to adjust the initial task allocation plan, and the adjustment process is optimized based on the relationship between communication delay and task priority.

9. A collaborative method for agricultural robot computing tasks according to claim 1, characterized in that The task execution and queue update in S6 includes: S61. Execute the adjusted task allocation plan: After the environmental interference compensation algorithm is started, the adjusted initial task allocation plan is executed, and the robot starts to execute tasks according to the adjusted task order and priority according to the new allocation plan; S62. Dynamically update the task priority queue: During the task execution process, continuously monitor the farmland environment and operation progress, and update the dynamic task priority queue in real time. As tasks are completed and new environmental data is collected, automatically adjust the task priority parameters in the queue.

10. An agricultural robot computing task collaboration system for implementing an agricultural robot computing task collaboration method according to any one of claims 1-9, characterized in that, It includes the following modules: Environmental perception and data acquisition module: Real-time collect environmental perception data of the farmland operation area, including soil moisture distribution data, crop growth status data, and real-time position coordinates of the robot cluster; Dynamic task priority queue generation module: Generate a dynamic task priority queue based on the soil moisture distribution data and crop growth status data, including irrigation task priority parameters, fertilization task priority parameters, and pest control task priority parameters; Heterogeneous computing resource adaptation model construction module: According to the generated dynamic task priority queue and the real-time position coordinates of the robot cluster, construct a heterogeneous computing resource adaptation model and output a task-resource matching matrix; Initial task allocation plan generation module: Based on the output task-resource matching matrix, use a distributed consensus mechanism to generate an initial task allocation plan; Environmental interference monitoring and compensation module: Real-time monitor the farmland environmental interference factors. When it is detected that the communication delay parameter exceeds the preset threshold, start the environmental interference compensation algorithm to adjust the initial task allocation plan; Task execution and queue update module: Execute the adjusted initial task allocation plan and continuously update the dynamic task priority queue.

Images