Unmanned aerial vehicle intelligent control method and system based on ant colony algorithm

Abstract

The invention relates to the technical field of unmanned aerial vehicle intelligent control, and discloses an unmanned aerial vehicle intelligent control method and system based on an ant colony algorithm, and the method comprises the steps: obtaining real-time environment data containing an obstacle position and wind speed changes, and a standardization processing result; calculating a pheromone accumulation speed update value and an initial interval according to the data to obtain adjusted interval data; generating a path sequence and optimizing a formation in combination with the interval data and a standardized result to obtain an optimized path and an enhanced correlation value; and transmitting a control instruction based on the associated value, feeding back and adjusting, and outputting an unmanned aerial vehicle cooperative control result. According to the method, dynamic cooperation of multiple unmanned aerial vehicles in a complex environment can be realized, and the operation safety and stability are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for unmanned aerial vehicles (UAVs), and in particular to an intelligent control method and system for UAVs based on ant colony algorithm. Background Technology

[0002] The precision and dynamic adaptability of UAV collaborative control directly affect mission execution efficiency and operational safety in complex environments. Efficient interval optimization and path planning are key to ensuring the stability of multi-UAV collaboration and avoiding collision risks. The swarm intelligence technology inherent in UAVs is of great significance for promoting the large-scale application of UAVs in scenarios such as logistics delivery, environmental monitoring, and smart agriculture.

[0003] In one existing technology, multi-UAV control mainly adopts a cooperative mode based on static path planning. It formulates a control scheme by pre-setting flight routes and fixed UAV intervals, and makes local adjustments based on basic environmental data collected by sensors. Information such as obstacle positions and wind speed changes is simply filtered and stored.

[0004] However, because static collaborative modes are based on preset parameters and rely on fixed logic adjustments, they lack dynamic integration and in-depth analysis of real-time environmental data. This makes it difficult to quickly respond to sudden obstacles or wind speed fluctuations, and it cannot adapt to real-time collaborative optimization of drone spacing and paths. When dynamic conditions such as sudden obstacles or rapid wind speed changes occur in the drone's operating environment, it can easily lead to formation chaos or mission interruption. Therefore, existing technologies are insufficient to meet the dynamic adaptation requirements of intelligent drone control in complex scenarios. Summary of the Invention

[0005] This invention provides an intelligent control method and system for unmanned aerial vehicles (UAVs) based on ant colony algorithm, which solves the problem that existing technologies cannot meet the dynamic adaptation requirements of intelligent control of UAVs in complex scenarios.

[0006] In a first aspect, the present invention provides an intelligent control method for unmanned aerial vehicles (UAVs) based on an ant colony algorithm, comprising: Real-time environmental data, including obstacle locations and wind speed changes, is collected and processed to obtain an updated value for pheromone accumulation speed. The initial interval distance of each UAV is calculated based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, the interval is expanded by fusing real-time data of obstacle position and wind speed change to obtain the adjusted interval data. A preliminary path sequence is generated based on the adjusted interval data, and node coordinates are obtained. The processing results of the real-time environmental data are integrated using the ant colony algorithm to determine the dynamic feasibility of the preliminary path sequence. If the dynamic feasibility meets the preset feasibility threshold, update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration. By adaptively adjusting the formation configuration and integrating real-time environmental data, a compensation value for insufficient response speed is obtained. If it exceeds the preset compensation threshold, an optimized path sequence is obtained through path planning and balancing. Based on the processing results of the optimized path sequence fused with the real-time environmental data, a correlation compensation between the path sequence and the environmental feedback is generated. If the correlation exceeds the preset correlation threshold, the formation parameters are adjusted and wind speed change monitoring is incorporated to obtain an enhanced path sequence correlation value. Integrate and enhance path sequence correlation values ​​and pheromone accumulation speed update values, transmit preliminary collaborative control commands and obtain initial feedback data to determine the preliminary fusion state of path sequence and command transmission; Based on the initial fusion state, the final control command is transmitted, and through adaptive updates and environmental feedback loops, the continuous adjustment state of task execution is determined, and the UAV cooperative control result is output.

[0007] Secondly, the present invention provides an intelligent control system for unmanned aerial vehicles (UAVs) based on an ant colony algorithm, comprising: Data processing module: Collects real-time environmental data including obstacle location and wind speed changes, and processes it to obtain the pheromone accumulation speed update value; Interval adjustment module: Calculates the initial interval distance of each drone based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, it integrates real-time data of obstacle position and wind speed change to expand the interval and obtain the adjusted interval data. Path analysis module: Generates a preliminary path sequence and obtains node coordinates based on the adjusted interval data. It then uses an ant colony algorithm to integrate the processing results of the real-time environmental data to determine the dynamic feasibility of the preliminary path sequence. Formation configuration module: If the dynamic feasibility meets the preset feasibility threshold, update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration; Path optimization module: By adaptively adjusting the formation configuration and integrating real-time environmental data, it obtains the compensation value for insufficient response speed assessment. If it exceeds the preset compensation threshold, it obtains an optimized path sequence through path planning and balancing. The correlation enhancement module generates correlation compensation between the path sequence and environmental feedback based on the processing result of the optimized path sequence fused with the real-time environmental data. If the correlation exceeds the preset correlation threshold, the formation parameters are adjusted to incorporate wind speed change monitoring to obtain the enhanced path sequence correlation value. Fusion determination module: integrates the enhanced path sequence correlation value and the pheromone accumulation speed update value, transmits the initial collaborative control command and obtains the initial feedback data, and determines the initial fusion state of the path sequence and command transmission; Control output module: Based on the initial fusion state, transmit the final control command, and through adaptive update and environmental feedback loop, determine the continuous adjustment state of task execution and output the UAV collaborative control result.

[0008] Thirdly, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the ant colony algorithm-based intelligent control method for unmanned aerial vehicles as described in any one of the above claims.

[0009] Fourthly, the present invention also provides a computer-readable storage medium comprising a stored computer program, wherein the computer program, when running, controls the device where the computer-readable storage medium is located to execute the above-described ant colony algorithm-based intelligent control method for unmanned aerial vehicles.

[0010] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention collects obstacle position and wind speed change data in real time through sensors, and after noise reduction and standardization, it combines ant colony algorithm to dynamically calculate pheromone accumulation speed, which breaks through the limitations of traditional static environment data processing, realizes rapid capture of dynamic environment changes, provides accurate data support for subsequent UAV interval optimization and path planning, and effectively reduces collision risk; (2) The present invention calculates the initial interval of the UAV based on the pheromone accumulation speed update value, triggers the interval expansion by combining the preset interval threshold, and synchronously optimizes the path planning and formation configuration, so that the control scheme can adapt to the dynamic environment and the changes in the needs of multi-machine collaboration, avoid the efficiency decline or collaboration chaos caused by the interval adaptation imbalance, and improve the safety and efficiency of multi-machine operation in complex scenarios. (3) The present invention constructs a closed-loop control mechanism of “environmental data processing - interval path optimization - command feedback adjustment”. Through preliminary command feedback verification, final command dynamic correction and real-time environmental feedback loop, the formation parameters are integrated for adaptive update, ensuring that the UAV collaborative control is highly matched with the dynamic environment requirements, providing a complete technical solution for the large-scale application of multi-UAV clusters, and improving the success rate and stability of mission execution. Attached Figure Description

[0011] Figure 1 This is a schematic flowchart of the intelligent control method for unmanned aerial vehicles based on ant colony algorithm provided in the first embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an unmanned aerial vehicle (UAV) intelligent control system based on ant colony algorithm provided in the second embodiment of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] Reference Figure 1 The first embodiment of the present invention provides an intelligent control method for unmanned aerial vehicles based on ant colony algorithm, including the following steps: S1 collects real-time environmental data including obstacle location and wind speed changes, and processes it to obtain the pheromone accumulation speed update value; S2, calculate the initial interval distance of each drone based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, fuse the obstacle position and the wind speed change and expand the interval to obtain the adjusted interval data. S3, Generate a preliminary path sequence and obtain node coordinates for the adjusted interval data, and use the ant colony algorithm to integrate the processing results of the real-time environmental data to determine the dynamic feasibility of the preliminary path sequence; S4. If the dynamic feasibility meets the preset feasibility threshold, update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration. S5, by integrating the real-time environmental data with the adaptively adjusted formation configuration, a compensation value for insufficient response speed assessment is obtained. If it exceeds a preset compensation threshold, an optimized path sequence is obtained through path planning and balancing. S6. Based on the processing result of the optimized path sequence fused with the real-time environmental data, generate the correlation compensation between the path sequence and the environmental feedback. If it exceeds the preset correlation threshold, adjust the formation parameters to incorporate wind speed change monitoring to obtain an enhanced path sequence correlation value. S7, integrate the enhanced path sequence association value and the pheromone accumulation speed update value, transmit the initial collaborative control command and obtain the initial feedback data, and determine the initial fusion state of the path sequence and command transmission; S8. Based on the preliminary fusion state, transmit the final control command, determine the continuous adjustment state of task execution through adaptive update and environmental feedback loop, and output the UAV cooperative control result.

[0014] In step S1, real-time environmental data including obstacle location and wind speed changes is collected and processed to obtain the pheromone accumulation speed update value, including: S11, spatial coordinate data is obtained through obstacle position sensor, and wind speed direction and magnitude data are obtained through wind speed sensor to form the real-time environmental dataset containing obstacle position and wind speed changes; S12, if the data points in the real-time environment dataset are complete and without missing data points, the real-time environment dataset is denoised and standardized to obtain a standardized real-time environment data set. S13, Based on the standardized real-time environmental data set, combined with the dynamic input adjustment parameters of environmental variables, the parameter adjustment results are obtained. The obstacle position and wind speed change data in the standardized real-time environmental data are iteratively calculated to obtain the pheromone accumulation intermediate value. S14, perform a weighted calculation on the intermediate value of pheromone accumulation, and integrate the parameter adjustment results to obtain the updated value of pheromone accumulation speed.

[0015] In step S11, spatial coordinate data is obtained through obstacle position sensors, and wind speed direction and magnitude data are obtained through wind speed sensors to form the real-time environmental dataset containing obstacle position and wind speed changes.

[0016] It should be noted that the obstacle location sensors are either lidar or ultrasonic sensors. LiDAR has a ranging accuracy of ±2 mm and can accurately collect the three-dimensional spatial coordinates of static obstacles, such as supports and pipes in smart agricultural greenhouses, and dynamic obstacles, such as temporarily placed agricultural tools. The coordinates are recorded in millimeters, and each obstacle is assigned a unique number for data tracking. The wind speed sensor uses an ultrasonic anemometer with a measurement range of 0-30 m / s and an accuracy of ±0.1 m / s. It can simultaneously acquire wind speed magnitude and direction, with north as 0° and increasing clockwise. For example, northeast is recorded as 45° and south as 180°. The formation of the real-time environmental dataset involves associating the two types of data by collection timestamp (accurate to the millisecond level). Each data entry contains "obstacle number - three-dimensional coordinates - wind speed magnitude - wind speed direction - collection time," ensuring a one-to-one correspondence between obstacle location and corresponding environmental wind speed. This avoids data matching errors due to time deviations and provides a complete and correlated original data foundation for subsequent processing.

[0017] In step S12, if the data points in the real-time environment dataset are complete and without missing data points, the real-time environment dataset is denoised and standardized to obtain a standardized real-time environment data set.

[0018] It should be noted that the data point integrity judgment checks whether each data point contains the core fields of "obstacle number, 3D coordinates, wind speed, wind direction, and acquisition time." Data is considered complete if there are no empty values ​​or missing items, thus avoiding the impact of missing data on the accuracy of subsequent calculations. The noise reduction process uses a mean filtering method, taking the arithmetic mean of five consecutive acquisitions before and after a data point as the final value. This eliminates small data fluctuations caused by sensor jitter or instantaneous interference, such as coordinate fluctuations within ±0.5 mm and wind speed fluctuations within ±0.2 m / s, ensuring data stability. The standardization process maps coordinate data and wind speed data to the 0-1 interval respectively, eliminating the impact of dimensional differences on algorithm iteration: coordinate data uses the boundary of the UAV's operating area as the extreme value, and converts the original coordinate values ​​into standardized values ​​in the 0-1 interval through maximum-minimum normalization; wind speed data uses the sensor's maximum range of 30 m / s as the maximum value, and converts the original wind speed values ​​into standardized values ​​in the 0-1 interval using a formula. The final standardized real-time environmental data set retains the correlation between data and meets the consistency requirements of subsequent algorithms for input data.

[0019] In step S13, based on the standardized real-time environmental data set and combined with the dynamic input adjustment parameters of environmental variables, the parameter adjustment results are obtained. The obstacle position and wind speed change data in the standardized real-time environmental data are iteratively calculated to obtain the pheromone cumulative intermediate value.

[0020] It should be noted that the parameter for dynamically adjusting the environmental variable input is the "pheromone evaporation coefficient." This coefficient is the core parameter in the ant colony algorithm that controls the pheromone decay rate. The default initial value is set to 0.5, and it needs to be dynamically adjusted according to the wind speed standardization value: when the wind speed standardization value is ≥0.1 (i.e., the actual wind speed is ≥3 m / s), the coefficient increases to 0.7. Because the environment is highly dynamic under high wind speed, accelerating pheromone evaporation can help the algorithm converge to the suitable path quickly. When the wind speed standardization value is <0.1 (i.e., the actual wind speed is <3 m / s), the coefficient remains at 0.5. The environment is stable under low wind speed, and the slow pheromone evaporation can ensure the stability of path planning. This adjustment logic is set based on the actual need that "the higher the wind speed, the stronger the environmental dynamism, and the need to improve the algorithm response speed." The iterative calculation simulates the path selection logic of an ant colony foraging. Standardized obstacle locations are considered "obstacle nodes," and open areas are considered "feasible nodes." In each iteration, the next feasible node is selected based on the node's pheromone concentration and wind speed influence factor. The number of iterations is set to 20, representing a balance between computational efficiency and path accuracy. Too few iterations may lead to path non-convergence, while too many will increase computation time. 20 iterations ensure a high probability of path convergence to the global optimum within the time delay required for real-time control. After each iteration, the cumulative pheromone amount at each feasible node is calculated, which is the median pheromone accumulation value.

[0021] In step S14, the pheromone accumulation intermediate value is weighted and calculated, and the parameter adjustment result is fused to obtain the pheromone accumulation speed update value.

[0022] It should be noted that the weighted calculation includes "iteration adaptation weight" and "wind speed adaptation weight," with weight ratios of 0.6 and 0.4 respectively. This weighting is based on the priority that "the number of iterations determines path convergence, and wind speed determines environmental adaptability." Iteration adaptation has a more critical impact on pheromone accumulation speed, hence its higher weight. Specifically, the iteration adaptation weight is calculated based on the number of iterations; the more iterations, the stronger the path adaptability, and the higher the weight. The wind speed adaptation weight is calculated based on the standardized wind speed value; the lower the wind speed, the more stable the environment, and the higher the weight. After obtaining the basic pheromone accumulation speed through weighted calculation, it is corrected by incorporating the adjustment results of the pheromone evaporation coefficient. If the evaporation coefficient is adjusted (e.g., from 0.5 to 0.7), the basic speed is corrected according to the adjustment amount to ensure that the pheromone accumulation speed in high wind conditions adapts synchronously to the dynamic environmental requirements as the evaporation coefficient adjusts. Finally, an updated pheromone accumulation speed value is obtained, providing a basis for subsequent UAV interval distance calculations.

[0023] In step S2, the initial interval distance of each drone is calculated based on the pheromone accumulation speed update value. If it is lower than a preset interval threshold, the obstacle position and the wind speed change are fused and the interval is expanded to obtain adjusted interval data, including: S21, call the obstacle location data and wind speed change data in the standardized real-time environmental data set, combine them with the pheromone cumulative speed update value, calculate the initial interval distance between each UAV, and obtain the initial interval distance set; S22, if the pheromone accumulation speed update value is lower than the preset speed threshold, the interval verification mechanism is triggered to verify the initial interval distance set and obtain the verification result; S23, if any initial interval distance in the initial interval distance set in the verification result is lower than the preset interval threshold, calculate the expanded interval distance based on the obstacle position and wind speed change data in the standardized real-time environmental data set, and generate the adjusted interval data.

[0024] In step S21, obstacle location data and wind speed change data from the standardized real-time environmental data set are called, and combined with the pheromone cumulative speed update value, the initial interval distance between each UAV is calculated to obtain the initial interval distance set.

[0025] It should be noted that the aforementioned access to standardized real-time environmental data involves extracting obstacle locations related to the drone's operating area from the set, such as the 3D coordinates of supports and pipes within a smart agricultural greenhouse, and wind speed variation data, such as the magnitude and direction of wind speed in different areas of the greenhouse, ensuring a high degree of data matching with the current computational scenario. The calculation of the initial interval distance is based on the pheromone accumulation rate update value. A higher pheromone accumulation rate indicates stronger adaptability to feasible paths in the environment, allowing drones to appropriately reduce intervals to improve operational efficiency. A lower pheromone accumulation rate indicates higher environmental complexity, requiring an initial increase in interval to ensure safety. The calculation simultaneously references obstacle distribution density and wind speed to ultimately obtain the initial interval distance between each drone, forming an initial interval distance set. The calculation method is based on the pheromone accumulation rate update value, combined with obstacle distribution density and wind speed, each assigned a weight, and the initial interval distance is calculated by multiplying the baseline interval by the sum of all coefficients. Each record in the set includes "drone number pair - initial interval distance - calculation basis (pheromone speed / obstacle density / wind speed)" to ensure traceability of the calculation.

[0026] In step S22, if the pheromone accumulation speed update value is lower than the preset speed threshold, the interval verification mechanism is triggered to verify the initial interval distance set and obtain the verification result.

[0027] It should be noted that the preset speed threshold is set to 1.2, without units. This threshold is based on the iterative rules of the ant colony algorithm and experience in drone operation safety. When the pheromone cumulative speed update value is lower than this threshold, it indicates strong environmental dynamism or poor path adaptability, and the initial interval distance may pose a safety risk, requiring verification. If it is higher than this threshold, it indicates a stable environment and good path adaptability, and the initial interval can be used directly. The interval verification mechanism compares each distance in the initial interval distance set with the "obstacle safety distance lower limit" and the "wind speed adaptability distance lower limit." The obstacle safety distance lower limit is 1.5 times the drone's diameter. For example, if the drone's diameter is 0.8 meters, the lower limit is 1.2 meters to avoid collisions between the drone and obstacles. The wind speed adaptability distance lower limit is set according to the wind speed. For example, the lower limit is 1.5 meters when the wind speed is 2 meters / second and 2 meters when the wind speed is 4 meters / second to prevent the wind speed from causing an unexpected reduction in the drone's spacing. After verification, a "qualified" or "unqualified" verification result is output. Unqualified indicates that there is a situation where the initial interval distance is lower than the corresponding lower limit.

[0028] In step S23, if any initial interval distance in the initial interval distance set in the verification result is lower than the preset interval threshold, the extended interval distance is calculated based on the obstacle position and wind speed change data in the standardized real-time environmental data set, and the adjusted interval data is generated.

[0029] It should be noted that the preset interval threshold is set at 1.8 meters. The setting rule is based on the maximum size of a common operational drone (e.g., wheelbase of 0.6 meters), and is determined by the formula that the threshold equals the maximum size of the drone multiplied by 3. In smart agricultural greenhouse operations, this threshold can avoid collisions without affecting spraying efficiency due to excessively large intervals. When any initial interval distance is lower than this threshold, the interval needs to be extended. The calculated extended interval distance is determined by integrating obstacle positions from standardized real-time environmental data. The closer the obstacle, the greater the extension. For example, 0.5 meters is extended when the distance to the obstacle is 1 meter, and 0.3 meters is extended when the distance is 2 meters. Wind speed data is also considered; the greater the wind speed, the greater the extension. For example, 0.4 meters is extended when the wind speed is 3 meters / second, and 0.6 meters is extended when the wind speed is 5 meters / second. The extension amount for each initial interval is determined, and the extended distance must simultaneously meet three conditions: "not lower than the preset interval threshold", "not lower than the lower limit of the obstacle safety distance", and "not lower than the lower limit of the wind speed adaptation distance". Finally, the expanded interval distances were organized into adjusted interval data, with the data format being "UAV number pair - adjusted interval distance - expansion basis (obstacle distance / wind speed)", providing a safe interval basis for subsequent path planning.

[0030] In step S3, a preliminary path sequence is generated for the adjusted interval data and node coordinates are obtained. The processing results of the real-time environmental data are integrated using the ant colony algorithm to determine the dynamic feasibility of the preliminary path sequence, including: S31, Based on the standardized real-time environment data set and combined with the adjusted interval data, a preliminary path sequence is generated, and the coordinates of key nodes in the preliminary path sequence are extracted to obtain a set of node coordinates; S32, Calculate the straight-line distance between each node and the obstacle position in the standardized real-time environment data set based on the node coordinate set; S33, if the straight-line distance is lower than the preset distance threshold, the wind speed change data in the standardized real-time environmental data set is integrated to update the node coordinate risk level and obtain the dynamic environmental data set; S34, adjust the node distribution of the preliminary path sequence according to the dynamic environment data set, simulate the flight state of the UAV on the adjusted path, determine whether the preset feasibility threshold is met, and obtain the path sequence data corresponding to the adjusted interval data.

[0031] In step S31, based on the standardized real-time environmental data set and combined with the adjusted interval data, a preliminary path sequence is generated, and the coordinates of key nodes in the preliminary path sequence are extracted to obtain a set of node coordinates.

[0032] It should be noted that, in this invention, unless otherwise stated, the processing result of the real-time environmental data refers to the data set obtained after processing the real-time environmental data and used for subsequent path planning and collaborative control. Its core includes, but is not limited to, the pheromone accumulation speed update value and the standardized real-time environmental data set.

[0033] The generation of the preliminary path sequence uses obstacle locations in standardized real-time environmental data as "prohibited areas" and open areas as "feasible areas." Combined with adjusted interval data, this ensures that the spacing between drones along the path meets safety requirements. A continuous path from the start to the end of the operation is planned using path planning logic. The path is divided according to the rule of "setting a key node every 5 meters." The 5-meter interval accurately depicts the path direction without increasing computational load due to too many nodes, adapting to the accuracy requirements of drone operations. Extracting the coordinates of key nodes involves recording the three-dimensional coordinates of each key node in millimeters, consistent with the coordinate system of the standardized real-time environmental data. Each node is assigned a unique number (e.g., ND-001, ND-002), ultimately forming a set of node coordinates. Each record in the set contains "node number - three-dimensional coordinates - corresponding operation stage (e.g., start / middle / end)," providing a node basis for subsequent feasibility assessment.

[0034] In step S32, the straight-line distance between each node and the obstacle position in the standardized real-time environmental data set is calculated based on the node coordinate set.

[0035] It should be noted that the calculated straight-line distance is performed for each node in the node coordinate set, calculating the spatial straight-line distance between each node and the three-dimensional coordinates of all obstacles in the standardized real-time environmental data set. The distance result is rounded to two decimal places (in meters) to ensure accuracy meets safety assessment requirements. After the calculation is completed, an "obstacle distance list" is created for each node, containing "node number - obstacle number - straight-line distance," which visually presents the spatial relationship between each node and surrounding obstacles, avoiding safety hazards caused by missing obstacle distances.

[0036] In step S33, if the straight-line distance is lower than a preset distance threshold, the wind speed change data in the standardized real-time environmental data set is fused to update the node coordinate risk level and obtain a dynamic environmental data set.

[0037] It should be noted that the preset distance threshold is set to 2 meters (adding a safety redundancy of 0.5-1.0 meters). This is based on the drone's body size and safety redundancy setting. For example, if the drone's diameter is 0.8 meters, a distance of 2 meters can reserve sufficient avoidance space. When the straight-line distance between a node and any obstacle is lower than this threshold, the node faces a collision risk and the risk level needs to be updated. The node coordinate risk level is divided into three levels: "low risk," "medium risk," and "high risk." Low risk is a distance ≥ 2 meters, with no avoidance requirement; medium risk is 1.5 meters ≤ distance < 2 meters, requiring cautious flight; and high risk is a distance < 1.5 meters, requiring emergency adjustment. When updating the risk level, wind speed change data from standardized real-time environmental data is simultaneously integrated. The higher the wind speed, the higher the risk level at the same distance. For example, at a distance of 1.8 meters, a wind speed of 2 meters / second is medium risk, while a wind speed of 4 meters / second is high risk. Finally, the "node number - 3D coordinates - risk level - associated obstacle / wind speed data" are organized into a dynamic environmental data set to provide a risk basis for path adjustment.

[0038] In step S34, the node distribution of the preliminary path sequence is adjusted according to the dynamic environment data set, the flight state of the UAV on the adjusted path is simulated, and it is determined whether the preset feasibility threshold is met, so as to obtain the path sequence data corresponding to the adjusted interval data.

[0039] It should be noted that the adjustment of the node distribution in the initial path sequence is aimed at medium- and high-risk nodes in the dynamic environment dataset. This is achieved by "translating nodes" to move them away from obstacles and into areas with lower wind speeds, with a movement range of 0.5-2 meters. This ensures that the risk level of the nodes is reduced to low risk after the movement. Alternatively, nodes can be added, while low-risk nodes remain unchanged. The simulated flight state is based on the adjusted node distribution, simulating the drone flying at a normal operating speed (e.g., 3 m / s) to check for issues such as "node spacing not meeting the adjusted interval data" or "sudden changes in flight direction exceeding 30°," as excessively large angles can easily lead to drone instability. The preset feasibility threshold is set to "95%" to balance operational safety and real-time performance, ensuring that most nodes comply to reduce risk without increasing computational time. Specifically, if the percentage of nodes meeting the safety conditions of interval compliance, angle compliance, and low risk level during simulated flight is ≥95%, the path sequence is dynamically feasible. Otherwise, the node distribution needs to be readjusted. Finally, the feasible adjusted path sequences are compiled into path sequence data, providing a path basis for subsequent formation configuration.

[0040] It should be noted that the core of the ant colony algorithm for path planning and integration lies in simulating the pheromone communication mechanism of ants. In one implementation, the drone... At the node Select the next node probability The calculation formula is as follows: in, For path ( , The concentration of pheromones on the surface; For heuristic information, nodes are usually taken as the basis. and The reciprocal of the distance between; For drones At the node The set of all possible successor nodes; and The parameters used to control the importance of pheromones and heuristic information typically take the value of [value missing]. =1, =2~5.

[0041] After all drones have completed one path construction, the pheromones along the path are updated according to the following rules: in, The pheromone volatility coefficient, with a value of (0, 1); For all drones on the path ( , The sum of the pheromone increments left on the surface.

[0042] In step S4, if the dynamic feasibility meets the preset feasibility threshold, the drone spacing optimization parameters are updated and iterated to determine the adaptively adjusted formation configuration, including: S41, if the dynamic feasibility of the adjusted preliminary path sequence meets the preset feasibility threshold, call the standardized real-time environmental data set to filter out the obstacle location and wind speed change data around the path. S42, Based on the coordinate range of the standardized real-time environmental data set, the UAV flight area is spatially discretized and divided into equal-sized grid units to obtain a spatial grid data set; S43, for the spatial grid data set, iteratively calculate the UAV spacing optimization parameters. If the formation response time corresponding to the optimization parameters is lower than the preset response threshold, adjust the parameter weights to obtain the first set of optimization parameters. S44, based on the first set of optimized parameters, the relative position coordinates of the UAV are adjusted by geometric transformation to form the adaptive adjusted formation configuration.

[0043] In step S41, if the dynamic feasibility of the adjusted preliminary path sequence meets the preset feasibility threshold, the standardized real-time environmental data set is called to filter out the location of obstacles and wind speed change data around the path.

[0044] It should be noted that the preset feasibility threshold is the 95% set in step S34. When the path sequence meets this threshold, it indicates that the path is basically feasible and can proceed to the formation configuration stage. The filtering of data around the path involves using the adjusted preliminary path sequence as the center and filtering obstacle positions and wind speed changes within a 5-meter radius on both sides. This ensures that subsequent formation configuration optimizes only key environmental factors around the path, improving computational efficiency. The filtered data is formatted as "path node number - surrounding obstacle position / wind speed - distance from path," providing accurate environmental data for spatial discretization processing.

[0045] In step S42, the UAV flight area is spatially discretized into equal-sized grid cells based on the coordinate range of the standardized real-time environmental data set, resulting in a spatial grid data set.

[0046] It should be noted that the spatial discretization involves dividing the UAV flight area into uniformly sized grid cells of 1m × 1m × 1m, using the coordinate extreme values ​​of standardized real-time environmental data as boundaries. A 1m grid accurately depicts environmental details, such as small obstacles and local wind speed differences, without causing a surge in computational complexity due to excessively fine grids. Each grid cell is labeled with three key pieces of information: "whether it contains obstacles," "average wind speed," and "corresponding path node." Cells containing obstacles are marked as "unavailable," while those without obstacles are marked as "available." The average wind speed is the average of all wind speed data within the grid. The corresponding path node is marked to indicate whether the grid overlaps with the path sequence. Finally, the information from all grid cells is compiled into a spatial grid dataset, formatted as "grid number-availability-average wind speed-corresponding path node," providing a spatialized environmental model for iterative spacing optimization parameters.

[0047] In step S43, for the spatial grid data set, the UAV spacing optimization parameters are iteratively calculated. If the formation response time corresponding to the optimization parameters is lower than the preset response threshold, the parameter weights are adjusted to obtain the first set of optimization parameters.

[0048] It should be noted that the UAV spacing optimization parameters include "lateral spacing coefficient" and "vertical spacing coefficient," both initially set to 1.0. The iterative calculation is based on a spatial grid data set: within the available grid area, parameters are adjusted according to average wind speed and path node distribution, and the formation response time is calculated after each iteration. The preset response threshold is set to 0.8 seconds (considering the latency of the entire sensing-computation-communication-execution chain, an inherent system latency margin of 100-200ms should be added), set based on the real-time control requirements of UAVs. An excessively long response time can lead to lag in environmental adaptation. If the formation response time after iteration is lower than this threshold, it indicates good parameter adaptability; if it is higher, the parameter weights need to be adjusted, such as increasing the weight of the longitudinal spacing coefficient, prioritizing the optimization of response speed along the path direction until the response time meets the threshold requirement. Finally, the parameters that meet the conditions are organized into a first set of optimized parameters, with the set format being "parameter type-parameter value-response time-adapted grid range".

[0049] In step S44, based on the first set of optimized parameters, the relative position coordinates of the UAV are adjusted by geometric transformation to form the adaptive adjusted formation configuration.

[0050] It should be noted that the geometric transformation includes two methods: translation and rotation. Translation adjusts the relative positions of the drones perpendicular to and along the path based on the horizontal and vertical spacing coefficients to ensure that the spacing meets the parameter requirements. Rotation adjusts the overall angle of the formation based on the wind speed and direction around the path. For example, if the wind speed is northeast, the formation is rotated by 15° to reduce wind resistance. After adjustment, it must be ensured that the relative position coordinates of all drones meet three conditions: "not exceeding the available grid area", "distance from obstacles ≥ preset distance threshold", and "spacing meets the first set of optimization parameters". Finally, the adjusted relative position coordinates of the drones are organized into an adaptive adjusted formation configuration data set, with the format "drone number - relative path coordinates - formation role (e.g., leader / left wing / right wing)", providing a suitable formation basis for subsequent path optimization.

[0051] In step S5, the adaptively adjusted formation configuration is fused with the real-time environmental data to obtain a compensation value for insufficient response speed assessment. If the compensation value exceeds a preset threshold, an optimized path sequence is obtained through path planning balancing, including: S51, associate the adaptively adjusted formation configuration data set with the standardized real-time environment data set, and calculate the response delay of the formation under different wind speeds and obstacle distributions; S52, based on the difference between the response delay and the preset response standard, calculate the compensation value for the insufficient response speed assessment; S53, if the compensation value exceeds the preset compensation threshold, the node spacing and turning angle of the initial path sequence are balanced and adjusted to generate an optimized path sequence set.

[0052] In step S51, the adaptively adjusted formation configuration data set is associated with the standardized real-time environment data set to calculate the response delay of the formation under different wind speeds and obstacle distributions.

[0053] It should be noted that the data association involves binding each drone role (such as leader, left wing) in the adaptively adjusted formation configuration to the wind speed and obstacle distribution data of the corresponding grid in the standardized real-time environmental data, ensuring a one-to-one correspondence between formation and environmental information. The calculated response delay simulates the formation adjustment process under different environmental scenarios: in wind speed scenarios, it calculates the time required for the formation to stabilize; in obstacle distribution scenarios, it calculates the time required for the formation to avoid obstacles. The response delay result is retained to two decimal places (in seconds), providing basic data for subsequent compensation value calculations.

[0054] In step S52, a compensation value for the insufficient response speed assessment is calculated based on the difference between the response delay and the preset response standard.

[0055] It should be noted that the preset response standard is set to 1.0 second, based on the dynamic environmental adaptation requirements of drones, and represents the ideal upper limit for response latency. The difference is "response latency - 1.0 second," and a positive difference indicates insufficient response speed. The calculated compensation value is obtained by multiplying the difference by an environmental impact coefficient. The higher the wind speed and the denser the obstacles, the larger the coefficient. For example, the coefficient is 1.5 when the wind speed is 6 m / s. Based on the rule in the smart agricultural greenhouse scenario that "the coefficient increases by 0.2 for every 2 m / s increase in wind speed and 0.3 for every 1 obstacle per square meter increase in obstacle density," and considering the impact of wind speed and obstacle density on the drone formation response speed, the coefficient is set to 1.4 for areas with dense obstacles. The larger the compensation value, the more serious the problem of insufficient response speed, and the more priority should be given to adjustment.

[0056] In step S53, if the compensation value exceeds the preset compensation threshold, the node spacing and turning angle of the initial path sequence are balanced and adjusted to generate an optimized path sequence set.

[0057] It should be noted that the preset compensation threshold is set to 0.5, based on a balance between response speed and path efficiency. Compensation values ​​exceeding this threshold will significantly impact environmental adaptability. When the compensation value exceeds this threshold, a path planning balancing operation must be performed. The balancing adjustment includes: node spacing adjustment, increasing the original 5-meter node spacing to 6-8 meters, reducing the number of nodes, decreasing the formation adjustment frequency, and improving response speed; and turning angle adjustment, reducing the original turning angle exceeding 30° to ≤25°, reducing the drone's attitude adjustment range and shortening response latency. After adjustment, the formation response latency needs to be re-simulated to ensure the compensation value drops below the threshold, while ensuring the path covers the work area and does not miss any target work points. Finally, the adjusted path sequences are compiled into an optimized path sequence set, with the format "path number - node coordinate sequence - adapted formation configuration - response latency".

[0058] In step S6, based on the processing result of fusing the optimized path sequence with the real-time environmental data, a correlation compensation between the path sequence and the environmental feedback is generated. If the correlation exceeds a preset threshold, the formation parameters are adjusted to incorporate wind speed change monitoring to obtain an enhanced path sequence correlation value, including: S61, Based on the processing results of the standardized real-time environmental data set, calculate the safety margin of each node and obstacle position in the optimized path sequence and the degree of adaptation with wind speed and direction to obtain the path environment adaptation parameters. S62, calculate the correlation compensation between the path sequence and the environmental feedback based on the deviation between the path environment adaptation parameters and the preset adaptation standard; S63, if the correlation compensation exceeds the preset correlation threshold, extract the wind speed change monitoring data from the standardized real-time environmental data set to obtain the adjusted formation parameters; S64, based on the adaptability of the adjusted formation parameters and the optimized path sequence, calculate the enhanced path sequence association value to obtain the enhanced path sequence association value.

[0059] In step S61, based on the processing results of the standardized real-time environmental data set, the safety margin of each node and obstacle position in the optimized path sequence and the degree of adaptation with wind speed and direction are calculated to obtain the path environment adaptation parameters.

[0060] It should be noted that the safety margin is "the straight-line distance between the node and the obstacle - a preset distance threshold." A positive result indicates safety redundancy, with a larger value indicating higher redundancy; a negative result indicates a safety risk. The adaptability is calculated based on the angle between the node's flight direction and the wind direction: 0° (tailwind) adaptability 1.0, 90° (crosswind) adaptability 0.7, 180° (headwind) adaptability 0.4. The closer the adaptability is to 1.0, the greater the wind speed's assist to flight and the smaller the drag. The safety margin and adaptability of each node are organized into path environment adaptability parameters, with the parameter format "node number - safety margin - adaptability," providing an environmental adaptability basis for associated compensation calculations.

[0061] In step S62, the correlation compensation between the path sequence and the environmental feedback is calculated based on the deviation between the path environment adaptation parameters and the preset adaptation standard.

[0062] It should be noted that the preset adaptation standards are "safety margin ≥ 0.5 meters, adaptation degree ≥ 0.6". A 0.5-meter safety margin ensures sufficient clearance for nodes, and an adaptation degree of 0.6 ensures that wind speed impact is within acceptable limits. The deviation is defined as "preset standard value - actual parameter value". A positive deviation indicates that the parameter does not meet the standard and requires compensation. The calculation of correlation compensation involves multiplying the safety margin deviation and adaptation degree deviation by their respective weights and then summing them. The safety margin has a weight of 0.6, and the adaptation degree has a weight of 0.4. Safety takes precedence over wind speed adaptation. A larger correlation compensation indicates poorer path-environment compatibility, requiring greater adjustments. The correlation compensation calculation result is rounded to two decimal places, has no unit, and is only used to assess the degree of adaptation deviation.

[0063] In step S63, if the correlation compensation exceeds the preset correlation threshold, wind speed change monitoring data is extracted from the standardized real-time environmental data set to obtain the adjusted formation parameters.

[0064] It should be noted that the preset association threshold is set to 0.4, which is a balance between path adaptability and adjustment costs. Association compensation exceeding this threshold will significantly affect operational stability. When association compensation exceeds this threshold, the formation parameters need to be adjusted. The extraction of wind speed change monitoring data involves obtaining the real-time wind speed change trend (e.g., increasing, decreasing, or stable) around the optimized path sequence within 10 minutes. If the wind speed is increasing, the lateral spacing of the formation needs to be increased (to reduce wind-induced interference); if the wind speed is decreasing, the longitudinal spacing can be appropriately decreased (to improve operational efficiency). The adjusted formation parameters must meet the requirements of "safety margin ≥ preset distance threshold" and "adaptability ≥ preset adaptation standard," and the parameter format is "parameter type - adjusted value - adapted wind speed trend."

[0065] In step S64, based on the adaptability of the adjusted formation parameters and the optimized path sequence, the enhanced path sequence association value is calculated to obtain the enhanced path sequence association value.

[0066] It should be noted that the calculated enhanced path sequence correlation value is obtained by scoring the adaptability of the adjusted formation parameters with the node distribution and environmental data of the optimized path sequence. The scoring range is 0-1.0: when the adaptability meets the following conditions, the correlation value is 0.9-1.0; when some conditions are met, the correlation value is 0.6-0.8; when none are met, the formation parameters need to be readjusted. Finally, the correlation values ​​corresponding to each optimized path sequence are organized into an enhanced path sequence correlation value set, with the set format being "path number - enhanced correlation value - adapted formation parameters - environmental adaptability", providing a highly adaptable path foundation for subsequent control command transmission.

[0067] In step S7, the enhanced path sequence association value and the pheromone accumulation speed update value are integrated, preliminary cooperative control commands are transmitted and initial feedback data is obtained, and the preliminary fusion state of the path sequence and command transmission is determined, including: S71, integrate the enhanced path sequence association value and the pheromone accumulation speed update value to generate a preliminary cooperative control command including flight speed and formation spacing; S72, the preliminary cooperative control command is transmitted to each UAV execution module through a message queue, and the delay data and command integrity identifier of the command transmission are recorded to obtain the first command transmission set; S73, Initial feedback data is obtained from each UAV execution module. If the deviation between the initial feedback data and the preliminary collaborative control command exceeds a preset fusion threshold, the feedback data is smoothed using a weighted average method to obtain a first set of feedback data. S74, based on the first feedback data set and the first instruction transmission set, calculate the matching degree between the path sequence nodes and the actual position of the UAV, and determine the preliminary fusion state of the path sequence and instruction transmission.

[0068] In step S71, the enhanced path sequence association value and the pheromone cumulative speed update value are integrated to generate a preliminary cooperative control command that includes flight speed and formation spacing.

[0069] It should be noted that the integrated data combines the enhanced path sequence correlation value with the pheromone cumulative speed update value: when the enhanced correlation value is ≥0.9, the flight speed is set to 3.5 m / s (to improve operational efficiency); when the correlation value is 0.7-0.89, the speed is set to 3.0 m / s (to balance efficiency and stability); when the correlation value is <0.7, the speed is set to 2.5 m / s (prioritizing stability). The formation spacing directly adopts the lateral and longitudinal spacing values ​​from the adjusted formation parameters. The initial collaborative control command format is "UAV number - flight speed - lateral spacing - longitudinal spacing - target path number", ensuring that the command includes the core parameters of UAV operation.

[0070] In step S72, the preliminary collaborative control command is transmitted to each UAV execution module through a message queue, and the delay data and command integrity identifier of the command transmission are recorded to obtain the first command transmission set.

[0071] It should be noted that the message queue is an ordered data channel for command transmission, ensuring that commands are transmitted in the order of UAV numbers to avoid command confusion. The recorded latency data is the time from command issuance to execution module reception; the latency must be ≤0.1 seconds, set based on real-time control requirements, as excessive latency can easily lead to poor command synchronization. The command integrity flag checks whether the command received by the execution module contains all core fields: "flight speed, formation spacing, and target path number." If all fields are present, it is marked "complete"; otherwise, it is marked "incomplete." Finally, the "UAV number - command content - transmission latency - integrity flag" are organized into the first command transmission set, providing a basis for command transmission for subsequent fusion status judgment.

[0072] In step S73, initial feedback data is obtained from each UAV execution module. If the deviation between the initial feedback data and the preliminary collaborative control command exceeds a preset fusion threshold, the feedback data is smoothed using a weighted average method to obtain a first set of feedback data.

[0073] It should be noted that the initial feedback data includes the actual flight speed, actual formation spacing, and actual position coordinates of the UAV, with a feedback frequency of 10 times / second (to ensure real-time data accuracy). The preset fusion threshold is set to "±10%", based on the UAV control accuracy setting. For example, if the commanded flight speed is 3.0 m / s, a deviation exceeding ±0.3 m / s requires processing. If the deviation exceeds this threshold, it indicates a significant difference between the feedback data and the command, requiring a weighted average smoothing method. The average of the current feedback data and the previous two feedback data is calculated, with the weight decreasing over time. The smoothed feedback data is organized into a first feedback data set, with the set format being "UAV number - actual speed - actual spacing - actual position - deviation status - smoothing processing identifier".

[0074] In step S74, based on the first feedback data set and the first instruction transmission set, the matching degree between the path sequence nodes and the actual position of the UAV is calculated to determine the preliminary fusion state of the path sequence and instruction transmission.

[0075] It should be noted that the matching degree calculation is performed for each UAV, calculating the straight-line distance between its actual position and the corresponding node of the target path sequence. A distance ≤ 0.5 meters is considered a "match," and a distance > 0.5 meters is considered a "mismatch." The matching degree is calculated by dividing the number of matched UAVs by the total number of UAVs. The initial fusion status is divided into three levels: "Excellent," "Good," and "Poor." "Excellent" is defined as a matching degree ≥ 90%, command transmission latency ≤ 0.1 seconds, and completeness 100%; "Good" is defined as a matching degree 80%-89% or latency 0.1-0.2 seconds and completeness 100%; and "Poor" is defined as a matching degree < 80%, latency > 0.2 seconds, or completeness < 100%. The fusion status corresponding to each target path sequence is recorded to provide a status basis for subsequent final control command transmission.

[0076] In step S8, the final control command is transmitted based on the preliminary fusion state. Through adaptive updates and environmental feedback loops, the continuous adjustment state of task execution is determined, and the UAV cooperative control result is output, including: S81, based on the preliminary fusion state, combined with the optimized path sequence and the enhanced path sequence correlation value, generate a final collaborative control command containing real-time adjustment rules, transmit it to each UAV execution module, record the delay and integrity of the command transmission, and obtain a second command transmission set; S82, obtain real-time feedback data from each UAV execution module, and perform noise reduction and standardization processing on the feedback data triggered by the second instruction transmission set to obtain the first real-time feedback set; S83, if the first real-time feedback set does not match the preset formation threshold, the environmental feedback data of the standardized real-time environmental data set is fused, the formation parameters are adjusted, and the first fused data set is obtained; S84, update the UAV formation parameters according to the first fused data set, dynamically correct the flight command in combination with the first real-time feedback set, determine the continuous adjustment state of mission execution, and output the UAV cooperative control result including flight trajectory and formation configuration based on the continuous adjustment state.

[0077] In step S81, based on the preliminary fusion state and combined with the optimized path sequence and enhanced path sequence correlation value, a final collaborative control command containing real-time adjustment rules is generated and transmitted to each UAV execution module. The delay and integrity of the command transmission are recorded to obtain a second command transmission set.

[0078] It should be noted that the generation of the final collaborative control command is as follows: When the initial fusion status is "Excellent," the command retains the flight speed and formation spacing of the initial collaborative control command, with the addition of an adjustment rule of "updating position every 10 seconds"; when the status is "Good," the speed is reduced by 0.2-0.3 m / s, the formation spacing is increased by 5%-10%, and the rule of "updating position every 5 seconds" is added; when the status is "Poor," the speed is reduced by 0.5 m / s, the spacing is increased by 10%-15%, the rule of "updating position every 2 seconds" is added, and the optimized path sequence with a higher enhanced correlation value is used. The final command format is "UAV number - flight speed - formation spacing - target path number - real-time adjustment rule (update frequency / path switching condition)." The latency recorded after transmission must be ≤0.08 seconds and the integrity 100%, exceeding the requirements of the initial command, to ensure the accuracy of the final command. This is then organized into a second command transmission set, with the same format as the first command transmission set.

[0079] In step S82, real-time feedback data is obtained from each UAV execution module, and the feedback data triggered by the second instruction transmission set is denoised and standardized to obtain the first real-time feedback set.

[0080] It should be noted that the real-time feedback data acquisition frequency has been increased to 2 times / second, higher than the initial feedback, to ensure the real-time nature of continuous adjustments. The data includes the actual speed, spacing, position, and attitude angles (such as pitch and yaw angles) of the UAV. The noise reduction process uses median filtering, taking the median of three consecutive feedback data to eliminate instantaneous pulse interference. Standardization processing converts the position coordinates to a coordinate system consistent with the optimized path sequence, and retains one decimal place for speed and spacing data to ensure a uniform data format. The processed real-time feedback data is organized into a first real-time feedback set, with the set format being "UAV ID-Timestamp-Actual Speed-Actual Spacing-Actual Position-Attitude Angle-Data Status".

[0081] In step S83, if the first real-time feedback set does not match the preset formation threshold, the environmental feedback data of the standardized real-time environmental data set is fused, the formation parameters are adjusted, and a first fused data set is obtained.

[0082] It should be noted that the preset formation thresholds include "flight speed deviation ≤ ±0.2 m / s", "formation spacing deviation ≤ ±5%", "attitude angle deviation ≤ ±3°", and "position deviation ≤ ±0.3 m". Specifically: the flight speed deviation threshold is set based on the response accuracy of the UAV's power system; ±0.2 m / s balances energy consumption and trajectory stability. The formation spacing deviation of ±5% is based on the safety redundancy requirements of multi-UAV collaborative operations, avoiding collisions due to excessive spacing fluctuations without unduly limiting collaborative flexibility. The attitude angle deviation of ±3° is based on the accuracy of the UAV's attitude sensors and flight stability requirements, ensuring that changes in the aircraft's attitude remain within a controllable range. The position deviation of ±0.3 m is set in conjunction with the accuracy requirements of the operational scenario to ensure mission execution effectiveness. If any deviation exceeds the threshold, it is considered a mismatch.

[0083] The fused environmental feedback data is extracted from standardized real-time environmental data, including the latest wind speed changes and obstacle location data. Formation parameters are adjusted as follows: when speed deviation exceeds limits, the speed is finely adjusted proportionally; when spacing deviation exceeds limits, the lateral / longitudinal spacing is adjusted based on environmental data; and when position deviation exceeds limits, the target position coordinates are adjusted. The adjusted parameters are then fused with the feedback data to form the first fused data set, formatted as "UAV ID - Adjusted Speed ​​- Adjusted Spacing - Adjusted Position - Reason for Deviation - Environmental Adaptation Basis".

[0084] In step S84, the UAV formation parameters are updated according to the first fused data set, the flight commands are dynamically corrected in combination with the first real-time feedback set, the continuous adjustment state of mission execution is determined, and the UAV cooperative control result including flight trajectory and formation configuration is output based on the continuous adjustment state.

[0085] It should be noted that the updated formation parameters are obtained by synchronizing the adjusted speed, spacing, and position from the first fused data set to all UAV execution modules. The dynamic correction of flight commands is based on real-time adjustment rules (e.g., updating every 1 second), taking into account UAV communication latency (approximately 0.5 seconds), environmental data sampling frequency (1 second / time), and operational response requirements (attitude adjustment must be completed within 2 seconds). Multiple sets of field tests have verified that this duration achieves an optimal balance between command transmission efficiency and timely environmental adaptation. Combined with cyclical environmental feedback data, the target parameters in the flight commands are corrected to ensure that the commands are always adapted to the current environment. The continuous adjustment state of task execution is divided into three levels: "stable," "fine-tuning," and "emergency adjustment." "Stable" is defined as all parameter deviations ≤ 50% of the threshold; "fine-tuning" is defined as 50% < deviation ≤ threshold; and "emergency adjustment" is defined as deviation > threshold. Ultimately, based on continuous adjustments, the system outputs collaborative control results for drones, including "complete flight trajectories of each drone (timestamp-position coordinates), changes in formation configuration throughout the process (timestamp-formation parameters), and environmental adaptation adjustment records (timestamp-adjustment reason-adjustment effect)," providing complete data support for mission review and subsequent optimization.

[0086] It should be noted that the dynamically corrected flight command is based on the deviation between the first real-time feedback set and the target path, and generates the correction amount using a proportional-integral-derivative (PID) control algorithm. Specifically, taking position deviation as an example, the calculation formula for its control quantity u(t) is as follows: in, for The deviation between the actual position of the drone and the target path nodes at any given time; , and These are the proportional, integral, and differential coefficients, which need to be tuned according to the UAV dynamics model, for example, by determining typical values ​​through the Ziegler-Nichols method or experimental trial and error.

[0087] In summary, this invention achieves end-to-end optimization of UAVs from environmental data acquisition to task execution through a dynamic collaborative control mechanism driven by ant colony algorithm. This effectively improves the collaborative accuracy and task execution stability of multiple UAVs in complex environments, providing strong support for the large-scale application of UAV swarms in scenarios such as agricultural plant protection and environmental monitoring.

[0088] Reference Figure 2 The second embodiment of the present invention provides an intelligent control system for unmanned aerial vehicles based on ant colony algorithm, comprising: The data processing module is used to collect real-time environmental data including obstacle locations and wind speed changes, and process it to obtain the pheromone accumulation speed update value; The interval adjustment module is used to calculate the initial interval distance of each drone based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, the obstacle position and the wind speed change are fused and the interval is expanded to obtain the adjusted interval data. The path analysis module is used to generate a preliminary path sequence and obtain node coordinates for the adjusted interval data, and to integrate the processing results of the real-time environmental data using the ant colony algorithm to determine the dynamic feasibility of the preliminary path sequence. The formation configuration module is used to update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration if the dynamic feasibility meets the preset feasibility threshold. The path optimization module is used to obtain a compensation value for insufficient response speed by integrating the real-time environmental data with the adaptively adjusted formation configuration. If the compensation value exceeds a preset compensation threshold, an optimized path sequence is obtained through path planning and balancing. The correlation enhancement module is used to generate correlation compensation between the path sequence and the environmental feedback based on the processing result of the optimized path sequence fused with the real-time environmental data. If the correlation exceeds the preset correlation threshold, the formation parameters are adjusted to incorporate wind speed change monitoring to obtain the correlation value of the enhanced path sequence. The fusion determination module is used to integrate the enhanced path sequence association value and the pheromone accumulation speed update value, transmit preliminary cooperative control instructions and obtain initial feedback data, and determine the preliminary fusion state of the path sequence and instruction transmission. The control output module is used to transmit the final control command according to the preliminary fusion state, determine the continuous adjustment state of task execution through adaptive update and environmental feedback loop, and output the UAV cooperative control result.

[0089] It should be noted that the ant colony algorithm-based UAV intelligent control system provided in this embodiment of the invention is used to execute all the process steps of the ant colony algorithm-based UAV intelligent control method in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0090] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a drone intelligent control program based on an ant colony algorithm. When the processor executes the computer program, it implements the steps in the various embodiments of the drone intelligent control method based on the ant colony algorithm described above, for example... Figure 1Step S11 is shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above system embodiments, such as a data processing module. Exemplarily, the computer program can be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, which describe the execution process of the computer program in the electronic device.

[0091] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0092] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0093] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0094] If the modules / units integrated into the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or system capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0095] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. An unmanned aerial vehicle intelligent control method based on an ant colony algorithm, characterized in that, include: Real-time environmental data, including obstacle locations and wind speed changes, is collected and processed to obtain an updated value for pheromone accumulation speed. The initial interval distance of each drone is calculated based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, the obstacle position and the wind speed change are fused and the interval is expanded to obtain the adjusted interval data. A preliminary path sequence is generated based on the adjusted interval data, and node coordinates are obtained. The processing results of the real-time environmental data are integrated using the ant colony algorithm to determine the dynamic feasibility of the preliminary path sequence. If the dynamic feasibility meets the preset feasibility threshold, update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration. By integrating the real-time environmental data with the adaptively adjusted formation configuration, a compensation value for insufficient response speed is obtained. If the value exceeds a preset compensation threshold, an optimized path sequence is obtained through path planning and balancing. Based on the processing result of the optimized path sequence fused with the real-time environmental data, the correlation compensation between the path sequence and the environmental feedback is generated. If it exceeds the preset correlation threshold, the formation parameters are adjusted to incorporate wind speed change monitoring to obtain an enhanced path sequence correlation value. Integrate the enhanced path sequence association value with the pheromone accumulation speed update value, transmit preliminary cooperative control instructions and obtain initial feedback data to determine the preliminary fusion state of the path sequence and instruction transmission; Based on the initial fusion state, the final control command is transmitted, and through adaptive updates and environmental feedback loops, the continuous adjustment state of task execution is determined, and the UAV cooperative control result is output.

2. The method according to claim 1, characterized in that, The process of collecting real-time environmental data, including obstacle locations and wind speed changes, and processing it to obtain pheromone accumulation speed update values ​​includes: Spatial coordinate data is obtained by using obstacle position sensors, and wind speed direction and magnitude data are obtained by using wind speed sensors to form the real-time environmental dataset containing obstacle position and wind speed changes. If the data points in the real-time environment dataset are complete and without missing data points, the real-time environment dataset is denoised and standardized to obtain a standardized real-time environment data set. Based on the standardized real-time environmental data set, combined with the dynamic input adjustment parameters of environmental variables, the parameter adjustment results are obtained. The obstacle position and wind speed change data in the standardized real-time environmental data are iteratively calculated to obtain the pheromone accumulation intermediate value. The pheromone accumulation intermediate value is weighted and calculated, and the parameter adjustment results are combined to obtain the pheromone accumulation speed update value.

3. The method according to claim 2, characterized in that, The process involves calculating the initial interval distance for each drone based on the pheromone accumulation speed update value. If this distance is lower than a preset interval threshold, the obstacle position and wind speed change are fused together, and the interval is expanded to obtain adjusted interval data, including: By calling the obstacle location data and wind speed change data in the standardized real-time environmental data set, and combining them with the pheromone cumulative speed update value, the initial interval distance between each UAV is calculated to obtain the initial interval distance set. If the pheromone accumulation speed update value is lower than the preset speed threshold, the interval verification mechanism is triggered to verify the initial interval distance set and obtain the verification result; If any initial interval distance in the initial interval distance set in the verification result is lower than the preset interval threshold, the extended interval distance is calculated based on the obstacle position and wind speed change data in the standardized real-time environmental data set, and the adjusted interval data is generated.

4. The method according to claim 3, characterized in that, The process of generating a preliminary path sequence and obtaining node coordinates based on the adjusted interval data, integrating the processing results of the real-time environmental data using an ant colony algorithm, and determining the dynamic feasibility of the preliminary path sequence includes: Based on the standardized real-time environment data set and combined with the adjusted interval data, a preliminary path sequence is generated, and the coordinates of key nodes in the preliminary path sequence are extracted to obtain a set of node coordinates. Based on the set of node coordinates, the straight-line distance between each node and the location of obstacles within the standardized real-time environmental data set is calculated. If the straight-line distance is lower than a preset distance threshold, the wind speed change data in the standardized real-time environmental data set is integrated to update the node coordinate risk level and obtain a dynamic environmental data set. The node distribution of the initial path sequence is adjusted based on the dynamic environment data set, the flight state of the UAV on the adjusted path is simulated, and it is determined whether the preset feasibility threshold is met, so as to obtain the path sequence data corresponding to the adjusted interval data.

5. The method according to claim 4, characterized in that, If the dynamic feasibility meets the preset feasibility threshold, the drone spacing optimization parameters are updated and iterated to determine the adaptively adjusted formation configuration, including: If the dynamic feasibility of the adjusted preliminary path sequence meets the preset feasibility threshold, the standardized real-time environmental data set is called to filter out the location of obstacles and wind speed change data around the path. Based on the coordinate range of the standardized real-time environmental data set, the UAV flight area is spatially discretized and divided into equal-sized grid units to obtain a spatial grid data set. For the spatial grid data set, the UAV spacing optimization parameters are iteratively calculated. If the formation response time corresponding to the optimization parameter is lower than the preset response threshold, the parameter weights are adjusted to obtain the first set of optimization parameters. Based on the first set of optimized parameters, the relative position coordinates of the UAVs are adjusted by geometric transformation to form the adaptive adjusted formation configuration.

6. The method according to claim 5, characterized in that, The process involves integrating the adaptively adjusted formation configuration with real-time environmental data to obtain a compensation value for insufficient response speed assessment. If this value exceeds a preset compensation threshold, an optimized path sequence is obtained through path planning and balancing, including: The adaptively adjusted formation configuration is associated with the standardized real-time environmental data set to calculate the response delay of the formation under different wind speeds and obstacle distributions. Based on the difference between the response delay and the preset response standard, the compensation value for the insufficient response speed assessment is calculated. If the compensation value exceeds the preset compensation threshold, the node spacing and turning angle of the initial path sequence are adjusted to generate an optimized path sequence set.

7. The method according to claim 6, characterized in that, The process of integrating the optimized path sequence with real-time environmental data to generate a correlation compensation between the path sequence and environmental feedback, and adjusting the formation parameters to incorporate wind speed change monitoring if the correlation exceeds a preset threshold, yields an enhanced path sequence correlation value, including: Based on the processing results of the standardized real-time environmental data set, the safety margin of each node and obstacle position in the optimized path sequence and the degree of adaptation with wind speed and direction are calculated to obtain the path environment adaptation parameters. Based on the deviation between the path environment adaptation parameters and the preset adaptation standard, the correlation compensation between the path sequence and the environment feedback is calculated. If the correlation compensation exceeds the preset correlation threshold, wind speed change monitoring data is extracted from the standardized real-time environmental data set to obtain the adjusted formation parameters; Based on the compatibility between the adjusted formation parameters and the optimized path sequence, the enhanced path sequence association value is calculated to obtain the enhanced path sequence association value.

8. The method according to claim 1, characterized in that, The process of integrating the enhanced path sequence association value and the pheromone accumulation speed update value, transmitting preliminary cooperative control commands and acquiring initial feedback data, and determining the preliminary fusion state of the path sequence and command transmission includes: By integrating the enhanced path sequence association value with the pheromone cumulative speed update value, a preliminary cooperative control command containing flight speed and formation spacing is generated; The initial collaborative control commands are transmitted to each UAV execution module through a message queue, and the latency data and command integrity identifier of the command transmission are recorded to obtain the first command transmission set. Initial feedback data is obtained from each UAV execution module. If the deviation between the initial feedback data and the preliminary collaborative control command exceeds a preset fusion threshold, the feedback data is smoothed using a weighted average method to obtain a first set of feedback data. Based on the first feedback data set and the first instruction transmission set, the matching degree between the path sequence nodes and the actual position of the UAV is calculated to determine the preliminary fusion state of the path sequence and instruction transmission.

9. The method according to claim 2, characterized in that, The process of transmitting final control commands based on the initial fusion state, determining the continuous adjustment state of task execution through adaptive updates and environmental feedback loops, and outputting UAV cooperative control results includes: Based on the initial fusion state, combined with the optimized path sequence and the enhanced path sequence correlation value, a final collaborative control command containing real-time adjustment rules is generated and transmitted to each UAV execution module. The delay and integrity of the command transmission are recorded to obtain a second command transmission set. Real-time flight data is obtained from each UAV execution module, and the feedback data triggered by the second command transmission set is denoised and standardized to obtain the first real-time feedback set; If the first real-time feedback set does not match the preset formation threshold, the environmental feedback data of the standardized real-time environmental data set is merged, the formation parameters are adjusted, and the first fused data set is obtained. Update the UAV formation parameters based on the first fused data set, dynamically correct flight commands in conjunction with the first real-time feedback set, determine the continuous adjustment state of mission execution, and output the UAV collaborative control result including flight trajectory and formation configuration based on the continuous adjustment state.

10. A drone intelligent control system based on ant colony algorithm, characterized in that, include: Data processing module: Collects real-time environmental data including obstacle location and wind speed changes, and processes it to obtain the pheromone accumulation speed update value; Interval adjustment module: Calculates the initial interval distance of each drone based on the pheromone accumulation speed update value. If it is lower than the preset interval threshold, it integrates the obstacle position and the wind speed change and expands the interval to obtain the adjusted interval data. Path analysis module: Generates a preliminary path sequence and obtains node coordinates based on the adjusted interval data, integrates the processing results of the real-time environmental data using the ant colony algorithm, and determines the dynamic feasibility of the preliminary path sequence; Formation configuration module: If the dynamic feasibility meets the preset feasibility threshold, update the drone spacing optimization parameters and iterate to determine the adaptively adjusted formation configuration; Path optimization module: By integrating the adaptively adjusted formation configuration with the real-time environmental data, it obtains a compensation value for insufficient response speed assessment. If the compensation value exceeds a preset threshold, it obtains an optimized path sequence through path planning and balancing. The correlation enhancement module generates correlation compensation between the path sequence and environmental feedback based on the processing result of the optimized path sequence fused with the real-time environmental data. If the correlation exceeds the preset correlation threshold, the formation parameters are adjusted to incorporate wind speed change monitoring to obtain the correlation value of the enhanced path sequence. Fusion determination module: integrates the enhanced path sequence association value and the pheromone accumulation speed update value, transmits preliminary cooperative control instructions and obtains initial feedback data, and determines the preliminary fusion state of the path sequence and instruction transmission; Control output module: Based on the preliminary fusion state, transmit the final control command, determine the continuous adjustment state of task execution through adaptive update and environmental feedback loop, and output the UAV collaborative control result.

Images