Path planning and coverage optimization method for agricultural unmanned aerial vehicle spraying system
By using technical means such as intelligent path planning, coverage optimization and real-time data feedback in agricultural drone spraying systems, the existing system has solved the problems of low path planning efficiency, insufficient coverage optimization capabilities and lagging responses in dynamic adjustments, and achieved efficient, accurate and comprehensive coverage spraying operation results.
Patent Information
- Application Number
- CN202411735071.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-02-28
AI Technical Summary
The existing agricultural drone spraying systems have problems in low path planning efficiency, insufficient coverage optimization capabilities, lagging dynamic adjustment response and insufficient intelligence level, which is difficult to meet the demands of modern agriculture for accuracy and efficiency.
The path planning module uses the A* or Dijkstra algorithm to generate the shortest basic path, and dynamically adjusts it with the RRT* algorithm; the area coverage optimization module uses the coverage optimization algorithm and the genetic algorithm to optimize the path node layout; the real-time data feedback module processes environmental data through the Kalman filtering algorithm and dynamically adjusts the spray path; the spray correction module generates the spray path through the coverage difference analysis algorithm.
It significantly improves the efficiency, accuracy and coverage of spraying operations, dynamically adapts to complex farmland environments, reduces resource waste, and improves coverage uniformity and response speed.
Smart Images
Figure CN119533484A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of agricultural spraying equipment, and in particular to a path planning and coverage optimization method for an agricultural fuel-powered unmanned aerial vehicle spraying system. Background Art
[0002] With the continuous development of agricultural modernization, traditional manual spraying operations have problems such as low efficiency, serious waste of resources and uneven coverage, which makes it difficult to meet the needs of large-scale farmland operations for accuracy and efficiency. The introduction of drone spraying technology provides a new way to solve these problems. However, the current common agricultural drone spraying systems still have the following shortcomings: Inefficient path planning: Most existing drone spraying systems use fixed or preset paths, and cannot be dynamically adjusted according to the complexity of farmland terrain and the distribution of obstacles, resulting in reduced operating efficiency. At the same time, path planning that is not optimized with intelligent algorithms can easily lead to uneven spraying coverage, increasing the waste of pesticides or fertilizers.
[0003] Limited coverage optimization capabilities: In farmland with complex or irregular terrain, the existing drone spraying system is not able to optimize the area coverage and cannot effectively avoid missed or double spraying. In addition, the mechanism for generating the re-spraying path for incompletely covered areas is relatively simple, making it difficult to ensure the efficiency of the re-spraying operation.
[0004] Insufficient dynamic adjustment and real-time response: During farmland operations, environmental conditions (such as wind speed, temperature, humidity, etc.) may change in real time, which places higher demands on spraying volume and path planning. However, existing systems usually lack real-time monitoring and feedback mechanisms for environmental data during operations, resulting in the inability to dynamically adjust the spraying volume according to real-time environmental changes. In addition, there is a lack of real-time path adjustment functions, which affects the accuracy of spraying operations and resource utilization efficiency.
[0005] Insufficient intelligence level: The current spraying system does not fully utilize artificial intelligence algorithms for comprehensive optimization. In particular, in the combination of path planning, coverage optimization and dynamic adjustment, the existing technology has not yet provided an efficient and comprehensive solution. Systems that lack multi-algorithm collaborative optimization mechanisms have limited performance in dealing with complex operating scenarios and cannot meet the needs of agricultural intelligent development.
[0006] Therefore, there is an urgent need for an agricultural drone spraying system that integrates path planning, area coverage optimization, and real-time data feedback. Through the combination of intelligent algorithms, the efficiency, accuracy, and coverage of spraying operations can be improved to meet the operational requirements of modern agriculture. Summary of the invention
[0007] The purpose of the present invention is to overcome the problems of low path planning efficiency, insufficient coverage optimization capability, delayed dynamic adjustment response, etc. in the existing agricultural UAV spraying technology, and to provide a path planning and coverage optimization method for an agricultural fuel UAV spraying system. Through the organic combination of intelligent algorithms, high efficiency, accuracy and comprehensive coverage of spraying operations can be achieved.
[0008] The path planning and coverage optimization method of the agricultural drone spraying system of the present invention is characterized by comprising the following steps: A. Information collection: obtain geographic information of the operating area and generate zoning information of the operating area; B. Path planning: Generate the shortest basic path from the starting point to the target area based on the partition information through the path planning module; C. Coverage optimization: Use the regional coverage optimization module to optimize the basic path and adjust the path node layout to ensure full coverage of the operation area; D. Dynamic adjustment: Collect operating environment information through real-time data feedback module and dynamically adjust the spraying path according to environmental changes; E. Re-spraying correction: When it is detected that the working area is not fully covered, a re-spraying path is automatically generated to complete the re-spraying operation.
[0009] Preferably, the path planning module uses the A* algorithm or the Dijkstra algorithm to generate the shortest basic path, and calculates the optimal path of the node through the cost function, wherein the cost function is: f(n)=g(n)+h(n); Among them, g(n) is the actual cost from the starting point to the current node, and h(n) is the heuristic estimate from the current node to the target node.
[0010] Preferably, the path planning module further adopts the RRT* algorithm for dynamically adjusting the path, and generates new path nodes by the following formula: ; in, is a random sampling point of distance in the existing path tree; The nearest node; η is the step size factor.
[0011] Preferably, the area coverage optimization module adjusts the path nodes through a coverage optimization algorithm, and the coverage calculation formula is: ; in, Indicates the area covered by spraying. Indicates the total area of the working area.
[0012] Preferably, the area coverage optimization module uses a genetic algorithm to optimize the path node distribution, and its fitness function is: ; in, and is the weight coefficient, coverage is the path coverage, and distance is the total length of the path.
[0013] Preferably, the environmental parameters collected by the real-time data feedback module include wind speed, humidity and temperature, and the system combines the Kalman filter algorithm to smooth the environmental data to ensure the accuracy of feedback when dynamically adjusting the path.
[0014] Preferably, the real-time data feedback module is further combined with a variable spraying amount adjustment module to dynamically adjust the spraying amount according to the crop density and health status of the operating area.
[0015] Preferably, a coverage difference analysis algorithm is used by detecting the uncovered area in the supplementary spraying correction step to automatically generate a supplementary spraying path based on the uncovered area.
[0016] Preferably, the path optimization result of the system is displayed in a real-time path on a user monitoring and operation terminal, and the terminal allows the user to manually adjust the path and spraying parameters through remote input.
[0017] Preferably, the path planning module collects environmental change data in real time through sensors, and when an obstacle or environmental condition change is detected, the RRT* algorithm is automatically called to dynamically adjust the path to generate an obstacle-free smooth path.
[0018] The agricultural fuel drone spraying system of the present invention achieves high efficiency, accuracy and comprehensive coverage of spraying operations through path planning, area coverage optimization, real-time data feedback and dynamic adjustment, and the synergy of user terminals. The specific technical effects are as follows: Intelligent path planning The present invention uses the path planning module to call the A* or Dijkstra algorithm to generate the shortest path from the starting point to the target area, significantly reducing flight time and fuel consumption and improving operation efficiency. During the operation, if an obstacle or complex terrain is detected, the path planning module calls the RRT* algorithm to generate a real-time dynamic path through dynamic sampling and path smoothing adjustment to ensure the continuity and safety of the spraying operation. This module combines static planning with dynamic adjustment algorithms to greatly improve the flexibility and adaptability of path planning, and is particularly suitable for complex terrain and dynamic operation environments.
[0019] Comprehensive coverage and efficient use of resources The regional coverage optimization module uses the coverage optimization algorithm to calculate the coverage of the spraying area in real time and optimize the node layout of the basic path to ensure full coverage of the operation area. For irregular terrain or large areas of farmland, the genetic algorithm is further called to balance the coverage efficiency and path length through iterative optimization of path nodes to avoid missed spraying and repeated spraying, while saving pesticide and fuel resources. The combination of coverage optimization algorithm and genetic algorithm significantly improves the accuracy and resource utilization of spraying operations.
[0020] Dynamic adjustment and real-time response The real-time data feedback module monitors environmental parameters (wind speed, humidity, temperature, etc.) through sensors, and combines the Kalman filter algorithm to smooth the environmental data to improve data accuracy and avoid data fluctuations affecting the path and spraying decisions. When environmental conditions change (such as increased wind speed), the system calls the dynamic adjustment algorithm (RRT*) of the path planning module to optimize the path in real time, and dynamically adjusts the spraying amount according to crop density and environmental parameters through the variable spraying module to ensure the stability of the spraying effect. This module realizes flexible operation and precise spraying in complex environments through the linkage of real-time data and algorithms.
[0021] Efficient generation of spraying paths The re-spraying path generation module generates a preliminary re-spraying path for uncovered areas by calling the path planning module, and adjusts the nodes of the path in combination with the coverage optimization algorithm. For re-spraying operations in complex areas, the genetic algorithm is further called to perform secondary optimization of the path nodes to ensure that the re-spraying path can cover the missed areas while reducing flight time and resource waste. The re-spraying efficiency of this module is significantly improved, ensuring full coverage of the operation area.
[0022] User convenience The user monitoring and operation terminal integrates path planning and coverage optimization results, provides visual dynamic path and spraying status display, and users can monitor the operation in real time. Under specific needs, users can manually adjust the spraying parameters or paths, and the system calls the corresponding algorithm (such as path planning or variable spraying) based on user input to achieve personalized adjustment of the operation.
[0023] The present invention significantly improves the accuracy and efficiency of agricultural drone spraying operations through intelligent collaboration of three modules: path planning, coverage optimization, and real-time adjustment. Compared with traditional spraying systems, the present invention can dynamically adapt to complex farmland environments, effectively solve the problems of resource waste, incomplete coverage, and insufficient response to environmental changes, and provides an efficient, flexible, and intelligent solution for modern agriculture. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 It is a block diagram of the management steps of this device. DETAILED DESCRIPTION
[0025] Embodiment 1: A path planning and coverage optimization method for an agricultural drone spraying system, characterized in that it comprises the following steps: A. Information collection: obtain geographic information of the operating area and generate zoning information of the operating area; B. Path planning: Generate the shortest basic path from the starting point to the target area based on the partition information through the path planning module; C. Coverage optimization: Use the regional coverage optimization module to optimize the basic path and adjust the path node layout to ensure full coverage of the operation area; D. Dynamic adjustment: Collect operating environment information through real-time data feedback module and dynamically adjust the spraying path according to environmental changes; E. Re-spraying correction: When it is detected that the working area is not fully covered, a re-spraying path is automatically generated to complete the re-spraying operation.
[0026] Embodiment 2: The difference between this embodiment and embodiment 1 is that this embodiment also includes the following features: The path planning module uses the A* algorithm or the Dijkstra algorithm to generate the shortest basic path and calculates the optimal path of the node through the cost function, where the cost function is: f(n)=g(n)+h(n); Among them, g(n) is the actual cost from the starting point to the current node, and h(n) is the heuristic estimate from the current node to the target node.
[0027] Preferably, the path planning module further adopts the RRT* algorithm for dynamically adjusting the path, and generates new path nodes by the following formula: ; in, is a random sampling point of distance in the existing path tree; The nearest node; η is the step size factor.
[0028] Preferably, the area coverage optimization module adjusts the path nodes through a coverage optimization algorithm, and the coverage calculation formula is: ; in, Indicates the area covered by spraying. Indicates the total area of the working area.
[0029] Preferably, the area coverage optimization module uses a genetic algorithm to optimize the path node distribution, and its fitness function is: ; in, and is the weight coefficient, coverage is the path coverage, and distance is the total length of the path.
[0030] Preferably, the environmental parameters collected by the real-time data feedback module include wind speed, humidity and temperature, and the system combines the Kalman filter algorithm to smooth the environmental data to ensure the accuracy of feedback when dynamically adjusting the path.
[0031] Preferably, the real-time data feedback module is further combined with a variable spraying amount adjustment module to dynamically adjust the spraying amount according to the crop density and health status of the operating area.
[0032] Preferably, a coverage difference analysis algorithm is used by detecting the uncovered area in the supplementary spraying correction step to automatically generate a supplementary spraying path based on the uncovered area.
[0033] Preferably, the path optimization result of the system is displayed in a real-time path on a user monitoring and operation terminal, and the terminal allows the user to manually adjust the path and spraying parameters through remote input.
[0034] Preferably, the path planning module collects environmental change data in real time through sensors, and when an obstacle or environmental condition change is detected, the RRT* algorithm is automatically called to dynamically adjust the path to generate an obstacle-free smooth path.
[0035] Embodiment 3: Large-scale farmland spraying operations Scenario: Farmland area: 200 hectares.
[0036] Topographic features: Some areas are irregular, with small ponds and shrub obstacles.
[0037] Environmental conditions: wind speed 2~5 m / s, humidity 60%, temperature 25°C.
[0038] Spraying task: Pesticide spraying, coverage rate ≥ 98%, spraying volume 5 L / hectare.
[0039] Drone parameters: Fuel capacity: 20 L, Fuel consumption: 2 L / hour.
[0040] Spraying speed: 10 m / s, spraying width: 5 m.
[0041] Application of path planning module Initial path generation: The basic path is planned using the A* algorithm, generating the shortest path from the starting point to the target area, covering most of the target area.
[0042] Generate path nodes: 120 nodes.
[0043] Initial path length: 18 km.
[0044] Estimated flight time: 36 minutes.
[0045] Dynamic path adjustment: Three pond obstacles were detected on site, and the path planning module called the RRT* algorithm for dynamic adjustment to generate an obstacle-free smooth path.
[0046] Path length after dynamic adjustment: 18.8 km.
[0047] Adjust path nodes: add 10 new nodes.
[0048] Application of regional coverage optimization module Coverage optimization algorithm: The coverage is calculated based on the basic path, and the initial coverage is 96.5%. The coverage optimization algorithm is called to adjust the path nodes, and the coverage reaches 99.2% after optimization.
[0049] Before node optimization: the area covered by the missed spraying was 3.5 hectares.
[0050] After node optimization: the area of the missed spraying area was reduced to 0.8 hectares.
[0051] Genetic algorithm optimizes path nodes: While optimizing coverage, the total path length and flight time were balanced, and the total path length was reduced to 18.5 km after optimization.
[0052] Application of real-time data feedback module Environmental parameter monitoring and spraying adjustment: During the operation, sensors detected that the wind speed briefly increased to 5 m / s in some areas. The system combined Kalman filtering to smooth the data and dynamically adjust the spraying path.
[0053] Adjusted path nodes: 2 new nodes are added to cover areas affected by wind speed.
[0054] Dynamic adjustment of spraying volume: reduce spraying volume by 10% in areas with higher wind speeds, and maintain standard spraying volume in other areas.
[0055] Application of the spraying path generation module Re-spraying of uncovered areas: According to the coverage analysis generated after the operation was completed, it was found that the missed spraying area was 0.8 hectares. The path planning module and coverage optimization module were called to generate the supplementary spraying path, and the genetic algorithm was used to optimize the nodes.
[0056] Re-spraying path nodes: 15 new nodes were added.
[0057] Length of the re-spraying path: 1.2 km.
[0058] Re-spraying time: 7 minutes.
[0059] User operation and terminal feedback Users can monitor path adjustments, environmental parameters and spray coverage in real time through the terminal.
[0060] The user manually adds 2 nodes to cover the special area (high-density crop area near the border), and the system calls the path planning algorithm to make adjustments.
[0061] Assignment Results Total operation time: 43 minutes (including re-spraying).
[0062] Spraying coverage: 99.5%.
[0063] Fuel consumption: about 2.3 L.
[0064] Total pesticide consumption: 202 L (including 3 L saved by optimization).
[0065] Summarize Path planning and dynamic adjustment effects: By combining the A* and RRT* algorithms, the path adjustment problem caused by obstacles and dynamic environmental changes is solved, ensuring flight safety and operational efficiency.
[0066] Coverage optimization effect: The coverage optimization algorithm combined with the genetic algorithm achieved high coverage (99.5%) and resource conservation (reduced fuel and pesticide consumption).
[0067] Dynamic adjustment and re-spraying efficiency: The real-time feedback module dynamically responds to changes in wind speed, and the supplementary spraying path generation module accurately supplements the uncovered areas to ensure the overall operation quality.
Claims
1. A path planning and coverage optimization method for an agricultural drone spraying system, characterized in that: The following steps are involved: A. Information collection: obtain geographic information of the operating area and generate zoning information of the operating area; B. Path planning: Generate the shortest basic path from the starting point to the target area based on the partition information through the path planning module; C. Coverage optimization: Use the regional coverage optimization module to optimize the basic path and adjust the path node layout to ensure full coverage of the operation area; D. Dynamic adjustment: Collect operating environment information through real-time data feedback module and dynamically adjust the spraying path according to environmental changes; E. Re-spraying correction: When it is detected that the working area is not fully covered, a re-spraying path is automatically generated to complete the re-spraying operation.
2. The path planning and coverage optimization method for an agricultural drone spraying system according to claim 1, characterized in that: The path planning module uses the A* algorithm or the Dijkstra algorithm to generate the shortest basic path and calculates the optimal path of the node through the cost function, where the cost function is: f(n)=g(n)+h(n); Among them, g(n) is the actual cost from the starting point to the current node, and h(n) is the heuristic estimate from the current node to the target node.
3. The path planning and coverage optimization method for an agricultural drone spraying system according to claim 1 or 2, characterized in that: The path planning module further uses the RRT* algorithm to dynamically adjust the path and generates new path nodes through the following formula: ； in, is a random sampling point of distance in the existing path tree; The nearest node; η is the step size factor.
4. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 3, characterized in that: The regional coverage optimization module adjusts the path nodes through the coverage optimization algorithm, and the coverage calculation formula is: ； in, Indicates the area covered by spraying. Indicates the total area of the working area.
5. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 4, characterized in that: The area coverage optimization module uses a genetic algorithm to optimize the path node distribution, and its fitness function is: ； in, and is the weight coefficient, coverage is the path coverage, and distance is the total length of the path.
6. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 5, characterized in that: The environmental parameters collected by the real-time data feedback module include wind speed, humidity and temperature. The system combines the Kalman filter algorithm to smooth the environmental data to ensure the accuracy of feedback when dynamically adjusting the path.
7. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 6, characterized in that: The real-time data feedback module is further combined with the variable spraying amount adjustment module to dynamically adjust the spraying amount according to the crop density and health status of the operating area.
8. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 7, characterized in that: By detecting the uncovered area in the re-spraying correction step and using a coverage difference analysis algorithm, a re-spraying path is automatically generated based on the uncovered area.
9. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 8, characterized in that: The path optimization results of the system are displayed in real-time on the user monitoring and operation terminal, and the terminal allows the user to manually adjust the path and spraying parameters through remote input.
10. The path planning and coverage optimization method for an agricultural drone spraying system according to any one of claims 1 to 9, characterized in that: The path planning module collects environmental change data in real time through sensors. When obstacles or changes in environmental conditions are detected, the RRT* algorithm is automatically called to dynamically adjust the path to generate an obstacle-free smooth path.