Autonomous collaborative operation method and system for multiple unmanned agricultural machines, and electronic equipment

Abstract

The invention discloses an autonomous collaborative operation method and system for multiple unmanned agricultural machines and electronic equipment, and belongs to the field of intelligent agricultural machines, and the method comprises the following steps: obtaining real-time communication data and position information between the unmanned agricultural machines based on a Beidou satellite system; farmland environment information and unmanned agricultural machine state information are collected based on various sensors; based on the farmland environment information and the unmanned agricultural machine state information, performing dynamic task allocation by adopting a competition-cooperation strategy to obtain a task allocation scheme; and based on the task allocation scheme, path planning is carried out by adopting an improved dung beetle algorithm, and an optimal autonomous collaborative operation path of the multiple unmanned agricultural machines is obtained. According to the invention, efficient autonomous collaborative operation of multiple unmanned agricultural machines in multiple plots in a large-scale farmland is realized, and the farmland operation efficiency is improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent agricultural machinery technology, and in particular relates to a method, system and electronic equipment for autonomous collaborative operation of multiple unmanned agricultural machines. Background Technology

[0002] With the development of technologies such as the Internet of Things, artificial intelligence, and BeiDou navigation, agricultural mechanization is gradually transforming into agricultural modernization, intelligence, and informatization. Traditional agriculture relies on manual operation or single-machine operation, which is inefficient and difficult to adapt to the needs of large-scale farmland. In addition, problems such as high labor costs, unstable operation quality, and resource waste are becoming increasingly prominent.

[0003] Multi-unmanned agricultural machinery collaborative operations can significantly improve operational efficiency and resource utilization through task allocation, path planning, and real-time communication, solving the problems of low efficiency and high cost in traditional agriculture. However, there are still many problems in related technologies in this field. Existing research mostly focuses on verification applications in small-scale experimental field environments, which are limited by the lack of heterogeneous constraint modeling and insufficient system scalability, making it difficult to adapt to the dynamic time-varying characteristics and spatial heterogeneity requirements of large-scale complex farmland scenarios. At the same time, multi-unmanned agricultural machinery collaborative operations rely on fixed network communication, lack self-organizing network capabilities, and cannot support real-time data interaction of large-scale nodes. Current task planning methods mostly adopt static deterministic models, failing to effectively integrate multi-source sensing data to construct a dynamic decision space, resulting in significantly limited autonomous fault tolerance capabilities under sudden working conditions. Traditional algorithms have slow convergence speeds when solving multi-unmanned agricultural machinery collaborative operations in large-scale farmland, making it difficult to meet the requirements of real-time operations. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention proposes a method, system, and electronic equipment for autonomous collaborative operation of multiple unmanned agricultural machines, thereby resolving the issues present in the prior art.

[0005] To achieve the above objectives, the present invention provides a method for autonomous collaborative operation of multiple unmanned agricultural machines, comprising:

[0006] Real-time communication data and location information between unmanned agricultural machines are obtained based on the BeiDou satellite system;

[0007] Based on the collection of farmland environmental information and unmanned agricultural machinery status information by multiple sensors;

[0008] Based on the farmland environment information and the status information of unmanned agricultural machinery, a task allocation scheme for autonomous collaborative operation of multiple unmanned agricultural machinery is obtained by using a competition-cooperation strategy for dynamic task allocation.

[0009] A fitness function is constructed, and based on the fitness function and the task allocation scheme, an improved dung beetle algorithm is used for path planning to obtain the optimal autonomous collaborative operation path for multiple unmanned agricultural machines.

[0010] Optionally, the farmland environmental information includes: weather change information, crop information, plot information, and obstacle information;

[0011] The status information of the unmanned agricultural machinery includes: real-time location information, operation status information, and fault diagnosis information.

[0012] Optionally, the process of obtaining a task allocation scheme by dynamically allocating tasks using a competition-cooperation strategy based on the farmland environment information and the status information of unmanned agricultural machinery includes:

[0013] Tasks are assigned based on farmland environmental information and operational needs.

[0014] Based on the aforementioned task, a competitive mechanism is used to evaluate the operational capabilities of each agricultural machine, resulting in a cost bid for each machine for the task.

[0015] Based on the preset allocation rules, the cost bids of each agricultural machine are optimized and selected to obtain preliminary task allocation results;

[0016] A collaborative mechanism is used to adjust the real-time job status of the preliminary task allocation results, resulting in a dynamically updated final task allocation scheme.

[0017] Optionally, in the process of dynamic task allocation using a competition-cooperation strategy, the lowest cost is used as the allocation criterion, where the objective function of the allocation criterion is:

[0018]

[0019] In the formula, m represents the number of unmanned agricultural machines required, x represents the task to be completed, and cost is... ij It is the cost for agricultural machinery i to complete task j; p ij This means that agricultural machinery i is assigned to task j.

[0020] Optionally, the process of collaboratively adjusting the real-time operation status of the preliminary task allocation results using a cooperative mechanism also includes quantifying the efficiency of unmanned agricultural machinery and evaluating the unmanned agricultural machinery itself based on its efficiency before making a bid.

[0021] The expression for calculating the efficiency of unmanned agricultural machinery is as follows:

[0022]

[0023] In the formula, e m To improve the efficiency of unmanned agricultural machinery; The service quality 's' of the unmanned agricultural machinery 'm' represents the revenue; 'u' represents the operating time efficiency of the unmanned agricultural machinery; cost(n) m ,c m ,fm ) represents the cost of the unmanned agricultural machinery m, which includes the completed workload n, energy consumption c, and failure rate f.

[0024] Optionally, the fitness function is:

[0025]

[0026] In the formula, fit is used to define the fitness function, t ij d represents the total time required for all agricultural machines i to complete all tasks j; ik This represents the sum of distances k that all agricultural machines i travel between tasks.

[0027] Optionally, based on the fitness function and the task allocation scheme, the process of using the improved dung beetle algorithm to perform path planning to obtain the optimal autonomous collaborative operation path for multiple unmanned agricultural machines includes:

[0028] Based on the task allocation scheme, the basic parameters are initialized, and the position of the unmanned agricultural machinery is initialized using the Logistic chaotic operator.

[0029] Construct a fitness function that aims to complete the maximum number of tasks in the shortest job time;

[0030] An initial path is generated based on the fitness function, and the fitness is evaluated.

[0031] Under the improved variable spiral search strategy, the initial path is iteratively updated based on the dung beetle's rolling ball, foraging, and stealing behavior model to obtain a new generation of candidate paths;

[0032] By repeatedly simulating the behavior of dung beetles and continuously updating their positions until the termination condition is met, the optimal autonomous collaborative operation path for multiple unmanned agricultural machines is obtained.

[0033] The present invention also provides a multi-unmanned agricultural machinery autonomous collaborative operation system for implementing the method described above. The system includes: a wireless communication module, an information acquisition module, a task allocation module, and a path planning module.

[0034] The wireless communication module is used for data transmission between unmanned agricultural machines and between unmanned agricultural machines and the control center.

[0035] The information acquisition module is used to collect farmland environmental information and real-time unmanned agricultural machinery status information transmitted by the wireless communication module;

[0036] The task allocation module dynamically allocates tasks based on the wireless communication module and the information acquisition module;

[0037] The path planning module, based on the information collection module and the task allocation module, uses an improved dung beetle algorithm to plan the optimal autonomous collaborative operation path for multiple unmanned agricultural machines.

[0038] The present invention also provides an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the steps of the method.

[0039] The present invention also provides a computer program product, including computer program instructions that are executed by a processor of the steps of the method.

[0040] Compared with the prior art, the present invention has the following advantages and technical effects:

[0041] This application adopts the high bandwidth transmission characteristics of 4G / 5G, which can realize data transmission and communication between unmanned agricultural machines and between unmanned agricultural machines and the control center. Compared with the traditional single communication mode, it can effectively solve the signal attenuation and interference problems in complex farmland environments. At the same time, the mutual communication between multiple unmanned agricultural machines can better synchronize the completion of operations and the status parameters of unmanned agricultural machines, thereby effectively saving resources and improving operational efficiency.

[0042] The competitive-cooperative strategy dynamically allocates tasks based on real-time work progress and the status of unmanned agricultural machinery. Compared with the traditional static task allocation method, this strategy can achieve a balance between global resource optimization and local efficiency improvement in complex farmland environments. At the same time, in case of emergencies, the unmanned agricultural machinery can complete autonomous and collaborative work tasks within a specified time through dynamic task allocation.

[0043] An autonomous collaborative operation algorithm for multiple unmanned agricultural machines based on an improved dung beetle algorithm is adopted to plan the optimal autonomous collaborative operation path for unmanned agricultural machines. Compared with traditional manual planning or simple algorithm planning, it can effectively reduce the operation time of multiple unmanned agricultural machines in multiple plots, enabling unmanned agricultural machines to complete collaborative operation tasks with the shortest path, thereby greatly improving operation efficiency.

[0044] The method of this application enables efficient autonomous collaborative operation of multiple plots and multiple unmanned agricultural machines in large-scale farmland, which improves farmland operation efficiency, effectively reduces labor costs and resource waste, and enables autonomous perception decision-making and autonomous collaborative operation in complex farmland environments, and more rationally plans operation paths and dynamically allocates tasks. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0046] Figure 1 This is a schematic diagram of the system structure according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0048] Figure 3 This is a schematic diagram of the electronic device structure according to an embodiment of the present invention. Detailed Implementation

[0049] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0050] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0051] To address the aforementioned technical challenges, this application provides a method, system, and electronic device for autonomous collaborative operation of multiple unmanned agricultural machines. The method employs a competition-cooperation strategy for dynamic task allocation based on real-time operation progress, unmanned agricultural machine status, and unforeseen events. It utilizes an autonomous collaborative operation algorithm for multiple unmanned agricultural machines based on an improved dung beetle algorithm to plan the optimal autonomous collaborative operation path. Through multiple iterations and training in simulation environments and on vehicle-mounted embedded platforms, this method effectively improves task allocation efficiency and resource utilization.

[0052] Example 1

[0053] like Figure 1 The diagram shown is a schematic of the system structure in this embodiment, including: a wireless communication module, an environmental perception module, a path planning module, and a task allocation module; wherein the wireless communication module is used for data transmission between unmanned agricultural machines and between the agricultural machines and the control center; the information acquisition module collects farmland environmental information and real-time unmanned agricultural machine status information transmitted by the wireless communication module; the task allocation module performs dynamic task allocation based on the wireless communication module and the information acquisition module; and the path planning module plans the optimal autonomous collaborative operation path for multiple unmanned agricultural machines based on the information acquisition module and the task allocation module.

[0054] First, wireless communication modules are used for data transmission and communication, including 4G / 5G communication and satellite communication. The 4G / 5G communication ensures data transmission and remote control between multiple unmanned agricultural machines and between the machines and the control center, further enabling real-time monitoring and remote control of the machines' status. Satellite communication, by receiving location information from the BeiDou satellite system, achieves high-precision positioning of the unmanned agricultural machines, ensuring accurate operation in the fields. As one of the world's four major satellite navigation systems, the BeiDou Navigation Satellite System (BDS) possesses significant advantages such as high-precision positioning, high reliability, and autonomous controllability. It can provide real-time and accurate geographic coordinate information for the unmanned agricultural machines. In complex farmland environments, the unmanned agricultural machines achieve centimeter-level positioning accuracy with the help of the BeiDou system, effectively avoiding obstacles, accurately executing tasks, and improving operational efficiency and quality.

[0055] The information acquisition module is used to collect and organize relevant information such as farmland environmental information and unmanned agricultural machinery status information. The farmland environmental information includes weather change information, crop information, plot information, and obstacle information. Weather change information helps predict abnormal weather (such as heavy rain, strong winds, high temperatures, hail, and other extreme weather) to further rationally schedule unmanned agricultural machinery operation time. Crop information includes crop type (such as wheat, corn, rice, etc.), growth stage (such as germination period, growth period, maturity period, etc.), crop distribution (including planting area boundaries, changes in planting density, etc.), plot information (including the boundaries and area of ​​the farmland plots to be operated on), and obstacle information (including the location and size of obstacles such as power poles, towers, ditches, and trees in the farmland). This information is the foundation for multi-unmanned agricultural machinery task allocation and path planning.

[0056] The status information of unmanned agricultural machinery includes real-time location information, operation status information, and fault diagnosis information transmitted by the wireless communication module. Real-time location information can ensure the precise operation of unmanned agricultural machinery in the field. Operation status information includes the speed information, operation progress, and task completion status of unmanned agricultural machinery, which is used for the management and planning of multi-machine collaborative operation. Fault diagnosis information can help to quickly locate faulty agricultural machinery and quickly update tasks, ensuring that the operation tasks are completed within the specified time.

[0057] After the above process is completed, the farmland environment information and unmanned agricultural machinery status information collected by the information collection module are used as input for task allocation. During the real-time operation of the unmanned agricultural machinery, a competition-cooperation strategy is adopted for dynamic task allocation. For example, if an unmanned agricultural machinery malfunctions or its task status changes, other unmanned agricultural machinery can take over its task based on the principle of proximity and the maximum task completion amount to ensure the continuity and efficiency of the operation, thereby obtaining an optimal dynamic task allocation scheme.

[0058] The path planning module adopts a multi-machine collaborative autonomous operation algorithm based on the improved dung beetle algorithm. It takes the basic information from the information collection module and the task from the task planning module as input, and through continuous updates and iterations, plans an optimal multi-unmanned agricultural machinery autonomous collaborative operation path.

[0059] In this case, the performance evaluation metrics for the path planning module are: shortest path length and shortest operation time. After the fitness function is established, the path planning module generates initial agricultural machinery operation paths. In each iteration, the fitness function not only evaluates the geometric characteristics of the path but also comprehensively considers multiple dimensions such as operation efficiency and resource consumption. The fitness value of each path is calculated based on the fitness function; a higher fitness value indicates a better degree of optimization. This process involves iteratively adjusting the paths through an optimization algorithm to achieve dual optimization of path length and operation time.

[0060] Example 2

[0061] like Figure 2 The diagram shown is a flowchart of the method in this embodiment. This embodiment provides a method for autonomous collaborative operation of multiple unmanned agricultural machines. The method includes the following steps: acquiring real-time communication data and location information between unmanned agricultural machines based on the BeiDou satellite system; collecting farmland environmental information and unmanned agricultural machine status information based on multiple sensors; using a competition-cooperation strategy to dynamically allocate tasks based on the farmland environmental information and unmanned agricultural machine status information to obtain a task allocation scheme for autonomous collaborative operation of multiple unmanned agricultural machines; constructing a fitness function; and using an improved dung beetle algorithm to perform path planning based on the fitness function and the task allocation scheme to obtain the optimal autonomous collaborative operation path for multiple unmanned agricultural machines.

[0062] Specifically, 4G / 5G and satellite communication are used to ensure data transmission between unmanned agricultural machines (UAVs) and the control center, while high-precision positioning of the UAVs is achieved through BeiDou satellite communication. The 4G / 5G communication ensures data transmission and remote control between multiple UAVs and between the UAVs and the control center. Leveraging the high bandwidth and low latency of 4G / 5G networks, the system can achieve real-time monitoring and precise control of the UAV status, ensuring efficient transmission and execution of work commands. Furthermore, 4G / 5G communication supports concurrent connections of large-scale devices, providing a stable network foundation for collaborative operation of multiple UAVs, enabling efficient cooperation in complex farmland environments.

[0063] The aforementioned satellite communication can receive location information from the BeiDou satellite system to achieve high-precision positioning of unmanned agricultural machinery, ensuring accurate operation of the machinery in farmland. As one of the world's four major satellite navigation systems, the BeiDou Navigation Satellite System (BDS) has significant advantages such as high-precision positioning, high reliability, and autonomous controllability. It can provide unmanned agricultural machinery with real-time and accurate geographic coordinate information. In complex farmland environments, unmanned agricultural machinery can achieve centimeter-level positioning accuracy with the help of the BeiDou satellite system, thereby effectively avoiding obstacles in the farmland, accurately executing operational tasks, and improving operational efficiency and quality.

[0064] Afterwards, it collects farmland environmental information such as weather and plots, as well as unmanned agricultural machinery status information transmitted by the wireless communication module, including real-time location information, operation status information, fault information, etc.

[0065] Multi-sensor data fusion is used to monitor farmland environmental information, which includes weather change information, crop information, plot information, and obstacle information. Weather change information helps predict abnormal weather (such as rainstorms, strong winds, high temperatures, hail, and other extreme weather) and further rationally arrange the operation time of unmanned agricultural machinery. Crop information includes crop type (such as wheat, corn, rice, etc.), growth stage (such as germination period, growth period, maturity period, etc.), crop distribution (including the boundaries of planting areas, changes in planting density, etc.), plot information (including the boundaries and area of ​​the farmland plots to be operated), and obstacle information (including the location and size of obstacles such as power poles, power towers, ditches, trees, etc. in the farmland). This information is the basic preparation for the task allocation and path planning of multiple unmanned agricultural machinery.

[0066] The status information of unmanned agricultural machinery includes real-time location information, operation status information, and fault diagnosis information transmitted by the wireless communication module. Real-time location information can ensure the precise operation of unmanned agricultural machinery in the field. Operation status information includes the speed information, operation progress, and task completion status of unmanned agricultural machinery, which is used for the management and planning of multi-machine collaborative operation. Fault diagnosis information can help to quickly locate faulty agricultural machinery and quickly update tasks, ensuring that the operation tasks are completed within the specified time.

[0067] This embodiment employs a competition-cooperation strategy, using information from the information acquisition module as input for dynamic task allocation. Task allocation is based on the farmland and unmanned agricultural machinery information collected through the information acquisition module. Task information is transmitted to each agricultural machine via wireless communication. Furthermore, the competition-cooperation strategy can be used to dynamically allocate tasks based on real-time work progress, unmanned agricultural machinery status, and unforeseen events.

[0068] The competition-cooperation strategy is a dynamic resource allocation method based on evolutionary game theory. It integrates competition and cooperation mechanisms to optimize the operational efficiency and quality of multiple unmanned agricultural machines. In the collaborative operation of multiple unmanned agricultural machines, this strategy enables each machine to compete and cooperate based on its own status and environmental information, achieving efficient task allocation and execution. The process includes: issuing operational tasks based on farmland environmental information and operational needs; evaluating the operational capabilities of each machine using a competition mechanism based on the operational tasks to obtain each machine's cost offer for the operational tasks; optimizing the cost offer of each machine based on preset allocation rules to obtain preliminary task allocation results; and using a cooperation mechanism to collaboratively adjust the real-time operational status of the preliminary task allocation results to obtain a dynamically updated final task allocation scheme.

[0069] Specifically, firstly, operational tasks are released based on farmland environmental information and operational needs, including task type, required resources, and task completion time, which can be referred to as a "quotation." At this point, each unmanned agricultural machine (UAV) uses a competitive mechanism to assess the cost of completing the task (such as time and energy consumption) based on its own status and environmental information, and generates a "bid" to compete for task allocation. This bid represents the UAV's offer to complete the task; UAVs with lower bids are more competitive in task allocation and are more likely to be selected to perform the task.

[0070] Next, the unmanned agricultural machinery submits its generated bid to the task allocation module. Based on the bid and preset rules (such as lowest cost, highest quality, etc.), the task allocation module selects the unmanned agricultural machinery with the best bid and assigns it a task. The assigned unmanned agricultural machinery executes the task and reports the result back to the task allocation module upon completion. The task allocation module monitors the operational status and progress of each unmanned agricultural machinery in real time for dynamic adjustments.

[0071] During task execution, unmanned agricultural machines share real-time status and environmental information through a cooperative mechanism to complete tasks collaboratively. If a new task arises or the status of an existing task changes, the task allocation module will reassign tasks and adjust the path planning. For example, if an unmanned agricultural machine malfunctions or its task status changes, other unmanned agricultural machines can take over its task based on proximity and maximum task completion to ensure the continuity and efficiency of operations.

[0072] Among these measures, in order to quantify the effectiveness of unmanned agricultural machinery and facilitate its better self-evaluation before bidding:

[0073]

[0074] Among them, e m To improve the efficiency of unmanned agricultural machinery; The service quality s of the unmanned agricultural machinery m represents the revenue, and u is a specific indicator of service quality, representing the operating time efficiency of the unmanned agricultural machinery in this application; cost(n m ,c m ,f m ) represents the cost of the unmanned agricultural machinery m, which includes the completed workload n, energy consumption c, and failure rate f.

[0075] To facilitate adjusting the weighting of various performance parameters when calculating the effectiveness of unmanned agricultural machinery, corresponding parameter data for all unmanned agricultural machinery are introduced, and the benefits or costs represented by each parameter are controlled between 0 and 1:

[0076]

[0077] Where δ1, δ2, δ3, and δ4 are the weighting factors for each performance parameter, which are mainly set according to the actual situation of unmanned agricultural machinery, and δ1+δ2+δ3+δ4=1, and σ is the total number of unmanned agricultural machinery. Let s be the total service quality of the total number of unmanned agricultural machines (k), u be a specific indicator of service quality representing the operating time efficiency of the unmanned agricultural machines, and n be the total service quality of the total number of unmanned agricultural machines (k). k ,c k ,f k It represents the total completed workload n, total energy consumption c, and total failure rate f of the total number of unmanned agricultural machines k.

[0078] The dynamic task planning model proposed in this embodiment uses the lowest cost as the allocation criterion:

[0079]

[0080] Where m represents the number of unmanned agricultural machines required, x represents the task to be completed, and cost is... ij It is the cost for agricultural machinery i to complete task j; p ij This means that agricultural machinery i is assigned to task j.

[0081] This paper designs a multi-machine collaborative autonomous operation algorithm based on an improved dung beetle algorithm. The algorithm continuously updates and iterates to plan the optimal collaborative operation path for multiple unmanned agricultural machines. Using information collected by the information acquisition module as input, the improved dung beetle algorithm simulates the optimization process in nature. It initializes the positions of the unmanned agricultural machines using a Logistic chaotic operator and updates their positions by simulating behaviors such as rolling a ball, dancing, foraging, and stealing. The algorithm searches for the optimal solution during the collaborative operation of multiple unmanned agricultural machines.

[0082] In the initial stage of path planning, the primary task is to construct a fitness function: to complete the maximum amount of tasks in the shortest operation time.

[0083]

[0084] In the formula, f it Used to define the fitness function, t ij d represents the total time required for all unmanned agricultural machines i to complete all tasks j; ik This represents the sum of distances k that all unmanned agricultural machines i travel between tasks.

[0085] After the fitness function is established, the path planning module generates initial paths. Subsequently, the system evaluates these paths one by one according to the constructed fitness function and accurately calculates the fitness value of each path. It should be noted that in this evaluation system, the higher the fitness value, the more significant the optimization of the corresponding path, that is, the more efficiently the path can meet the predetermined operational goals and task requirements.

[0086] Specifically, in this embodiment, an improved dung beetle algorithm is used to plan the optimal path.

[0087] 1. Initialize parameters: First, set the basic parameters: number of dung beetles N, maximum number of iterations pop_max, rolling step size α, turning angle ω0, deflection coefficient k, learning factor β, theft coefficient λ, initial chaotic value x0, etc.

[0088] In the solution space of the farmland, an initial position is randomly generated as a coordinate point for each unmanned agricultural machine:

[0089] X i (0)=(x i ,y i ).

[0090] Initialization speed of each unmanned agricultural machine: V i (0)=(v ix ,v iy ).

[0091] The method for initializing the position of unmanned agricultural machinery using the Logistic chaotic operator is as follows: x n+1 =r·x n (1-x n ).

[0092] Where, x n x represents the initial position of the unmanned agricultural machinery. n+1 x represents the position of the unmanned agricultural machinery after initialization using the Logistic chaotic operator. n ∈[0,1], r represents a positive integer related to the reproduction rate and mortality rate. The chaotic sequence is mapped to the value space of the optimization variable, thereby using the chaotic properties to determine the initial position of the unmanned agricultural machinery.

[0093] When the dung beetle performs the rolling ball behavior, its position is updated as follows when there are no obstacles in its direction of movement:

[0094]

[0095] Where n represents the current iteration number, indicating the position of the j-th dung beetle in the nth iteration; k is the deflection coefficient, within the range (0, 0.2]; B represents a fixed value within the range (0, 1); C is a natural coefficient, whose value may be -1 or 1, where 1 indicates no deviation and -1 indicates deviation from the initial path. X m This represents the worst-case position globally, and Δx is used to simulate changes in light intensity.

[0096] The location updates when dung beetles are foraging, reproducing, and stealing are as follows:

[0097]

[0098] The improved variable spiral search strategy is: β = e qn ×cos(2πq),

[0099] q is defined as a random number uniformly distributed between 0 and 1; meanwhile, h is affected by the cosine function and varies with the number of iterations; x * It is determined to be the current local optimum, while hq * and UQ * represents the upper and lower limits of the breeding region, respectively; meanwhile, g is a random vector of size 1×D that follows a normal distribution, and D represents the dimension of the optimization challenge.

[0100] The behavior of dung beetles is repeatedly simulated, and their positions are continuously updated until the termination condition is met: the maximum number of iterations is reached or the rate of change of the optimal solution is lower than a threshold. During the iteration process, the dung beetle with the highest fitness value is recorded, which is the optimal solution and corresponds to the optimal operation path for multi-unmanned agricultural machinery collaborative operation.

[0101] Dung beetles' behaviors, such as rolling balls, dancing, foraging, and stealing, help the algorithm explore diverse paths in complex farmland environments and find the shortest and safest work routes. Meanwhile, the improved location update strategy further enhances the algorithm's global search capability and convergence speed, ensuring the accuracy and efficiency of path planning.

[0102] The improved dung beetle algorithm supports real-time collaboration among multiple unmanned agricultural machines during operation. By dynamically adjusting the path, the agricultural machines can flexibly cooperate with other agricultural machines based on real-time environmental information and their own status, thereby improving overall operation efficiency and quality.

[0103] Example 3

[0104] like Figure 3The diagram shown is a schematic representation of the physical structure of the electronic device in this embodiment. The electronic device includes a computing device and an in-vehicle embedded platform. The computing device includes a processor 310, a memory 320, and a communication bus 330. The processor 310 and the memory 320 communicate with each other through the communication bus 330. The processor 310 can call logical instructions in the memory 320 to execute a method for autonomous collaborative operation of multiple unmanned agricultural machines. The memory 320 stores executable code. When the executable code is processed by the processor 310, it can enable the processor 310 to execute part or all of the methods described above.

[0105] The vehicle-mounted embedded platform in the electronic device includes an industrial PC (Jetson Nano), a Livox Mid-360, a camera, an IMU, a host computer, a mobile terminal, and a network terminal. The Jetson Nano serves as the core computing unit, providing powerful computing capabilities for processing complex algorithms and real-time data. The Livox Mid-360 is a high-performance LiDAR sensor primarily used for environmental perception and obstacle avoidance. The camera is used to collect visual information from farmland, supporting various shooting modes and functions. The IMU measures the motion status of the unmanned agricultural machinery. The host computer, as the core control unit of the vehicle-mounted embedded platform, is responsible for the overall system coordination and management. Mobile devices (such as tablets and smartphones) are used for remote monitoring and operation of the unmanned agricultural machinery. The network terminal is responsible for the connection and data transmission between the unmanned agricultural machinery and the 4G / 5G network. This vehicle-mounted embedded platform can support the testing and verification of the models and algorithms involved in the methods described above.

[0106] The electronic device includes at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the steps of a method for autonomous collaborative operation of multiple unmanned agricultural machines.

[0107] Furthermore, the method according to this application can also be implemented as a computer program or computer program product, which includes computer program code instructions for performing some or all of the steps in the method described above.

[0108] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for autonomous collaborative operation of multiple unmanned agricultural machines, characterized in that, Includes the following steps: Real-time communication data and location information between unmanned agricultural machines are obtained based on the BeiDou satellite system; Based on the collection of farmland environmental information and unmanned agricultural machinery status information by multiple sensors; Based on the farmland environment information and the status information of unmanned agricultural machinery, a task allocation scheme for autonomous collaborative operation of multiple unmanned agricultural machinery is obtained by using a competition-cooperation strategy for dynamic task allocation. A fitness function is constructed, and based on the fitness function and the task allocation scheme, an improved dung beetle algorithm is used for path planning to obtain the optimal autonomous collaborative operation path for multiple unmanned agricultural machines.

2. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 1, characterized in that, The farmland environmental information includes: weather change information, crop information, plot information, and obstacle information; The status information of the unmanned agricultural machinery includes: real-time location information, operation status information, and fault diagnosis information.

3. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 1, characterized in that, The process of obtaining a task allocation scheme by dynamically allocating tasks based on the farmland environment information and the status information of unmanned agricultural machinery using a competition-cooperation strategy includes: Tasks are assigned based on farmland environmental information and operational needs. Based on the aforementioned task, a competitive mechanism is used to evaluate the operational capabilities of each agricultural machine, resulting in a cost bid for each machine for the task. Based on the preset allocation rules, the cost bids of each agricultural machine are optimized and selected to obtain preliminary task allocation results; A collaborative mechanism is used to adjust the real-time job status of the preliminary task allocation results, resulting in a dynamically updated final task allocation scheme.

4. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 3, characterized in that, In the process of dynamic task allocation using a competition-cooperation strategy, the lowest cost is used as the allocation criterion, and the objective function for the allocation criterion is: In the formula, m represents the number of unmanned agricultural machines required, x represents the task to be completed, and cost is... ij It is the cost for agricultural machinery i to complete task j; p ij This means that agricultural machinery i is assigned to task j.

5. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 3, characterized in that, The collaborative adjustment process of the real-time operation status of the preliminary task allocation results using a cooperative mechanism also includes quantifying the efficiency of unmanned agricultural machinery and evaluating the unmanned agricultural machinery itself based on its efficiency before making a bid. The expression for calculating the efficiency of unmanned agricultural machinery is as follows: In the formula, e m To improve the efficiency of unmanned agricultural machinery; The service quality 's' of the unmanned agricultural machinery 'm' represents the revenue; 'u' represents the operating time efficiency of the unmanned agricultural machinery; cost(n) m ,c m ,f m ) represents the cost of the unmanned agricultural machinery m, which includes the completed workload n, energy consumption c, and failure rate f.

6. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 1, characterized in that, The fitness function is: In the formula, f it Used to define the fitness function, t ij d represents the total time required for all agricultural machines i to complete all tasks j; ik This represents the sum of distances k that all agricultural machines i travel between tasks.

7. The method for autonomous collaborative operation of multiple unmanned agricultural machines according to claim 6, characterized in that, Based on the fitness function and the task allocation scheme, the process of obtaining the optimal autonomous collaborative operation path for multiple unmanned agricultural machines using the improved dung beetle algorithm includes: Based on the task allocation scheme, the basic parameters are initialized, and the position of the unmanned agricultural machinery is initialized using the Logistic chaotic operator. Construct a fitness function that aims to complete the maximum number of tasks in the shortest job time; An initial path is generated based on the fitness function, and the fitness is evaluated. Under the improved variable spiral search strategy, the initial path is iteratively updated based on the dung beetle's rolling ball, foraging, and stealing behavior model to obtain a new generation of candidate paths; By repeatedly simulating the behavior of dung beetles and continuously updating their positions until the termination condition is met, the optimal autonomous collaborative operation path for multiple unmanned agricultural machines is obtained.

8. A multi-unmanned agricultural machinery autonomous collaborative operation system, characterized in that, The system for implementing the method as described in claim 1 includes: a wireless communication module, an information acquisition module, a task allocation module, and a path planning module; The wireless communication module is used for data transmission between unmanned agricultural machines and between unmanned agricultural machines and the control center. The information acquisition module is used to collect farmland environmental information and real-time unmanned agricultural machinery status information transmitted by the wireless communication module; The task allocation module dynamically allocates tasks based on the wireless communication module and the information acquisition module; The path planning module, based on the information collection module and the task allocation module, uses an improved dung beetle algorithm to plan the optimal autonomous collaborative operation path for multiple unmanned agricultural machines.

9. An electronic device, characterized in that, The electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the steps of the method according to any one of claims 1-7.

10. A computer program product comprising computer program instructions, characterized in that, When executed by a processor, the computer program instructions implement the steps of the method described in any one of claims 1-7.

Images