Multi-unmanned aerial vehicle cooperative full-coverage path planning optimization method and system

Abstract

The invention relates to the field of unmanned aerial vehicle intelligent scheduling, and discloses a multi-unmanned aerial vehicle cooperative full-coverage path planning optimization method and system, and the method comprises the steps: carrying out the continuous grid scanning of an inspection region based on the local terrain complexity, the obstacle density and the unmanned aerial vehicle sensing coverage radius; the real-time electric quantity, the load capacity, the speed limitation and the communication link state of each unmanned aerial vehicle are evaluated, and a cooperative capacity matrix is constructed; based on a variable-size operation unit and a matching map, introducing an adaptive collaborative optimization model, and predicting track conflict, coverage redundancy and energy consumption deviation through multiple factors; the coverage information and the abnormal state of the unmanned aerial vehicle are collected in real time in the inspection process, and when missing or repeated coverage is found, incremental replanning is executed for the affected local operation unit; and based on the coverage data and energy consumption statistics after each round of inspection is completed, calculating a coverage efficiency index and omission risk distribution in combination with real-time and accumulated data. The method has the advantages of improving the inspection efficiency and reducing the omission risk.

Description

Technical Field

[0001] This invention relates to the field of intelligent scheduling of unmanned aerial vehicles (UAVs), specifically to a method and system for multi-UAV collaborative full-coverage path planning optimization. Background Technology

[0002] When multiple drones collaborate to perform full-coverage tasks, pre-defined fixed routes or path planning methods based on simple heuristic algorithms are typically used. However, in practical applications, problems arise such as overlapping routes, coverage blind spots, and uneven power distribution among drones in complex terrain or environments with obstacles. For example, in large greenhouse farms, traditional path planning methods often ignore the impact of the canopy layout on the flight path, leading to duplicate coverage or missed areas during drone inspections. Simultaneously, if the power consumption of multiple drones differs significantly, some drones may return to base prematurely, resulting in decreased coverage efficiency. Even with dynamic environments, complex terrain, and multi-drone collaborative full-coverage tasks, achieving optimal coverage efficiency, minimal path conflicts, and balanced energy scheduling remains challenging. Therefore, designing a multi-drone collaborative full-coverage path planning optimization method and system to improve inspection efficiency and reduce the risk of omissions is essential. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method and system for multi-UAV collaborative full-coverage path planning optimization, which has the advantages of improving inspection efficiency and reducing the risk of omissions, and solves the problems mentioned in the background technology.

[0004] To achieve the aforementioned goals of improving inspection efficiency and reducing the risk of omissions, this invention provides the following technical solution: a multi-UAV collaborative full-coverage path planning optimization method, comprising the following steps:

[0005] The inspection area is continuously scanned in a grid pattern, and variable-size operation units are automatically generated based on local terrain complexity, obstacle density, and drone sensor coverage radius.

[0006] Based on the variable-size work unit, the real-time power, load capacity, speed limit and communication link status of each UAV are evaluated, a collaborative capability matrix is ​​constructed, and a matching map of multiple UAVs and work units is generated by combining the coverage priority and reachability of the work unit.

[0007] Based on variable-size work units and matching maps, an adaptive collaborative optimization model is introduced to predict track conflicts, coverage redundancy and energy consumption deviations through multiple factors, and dynamically adjust the UAV path priority order, flight altitude and heading offset according to the prediction results.

[0008] During the inspection, the drone coverage information and abnormal status are collected in real time. When missing or duplicate coverage data is found, incremental replanning is performed on the affected local operation units. Combined with the adaptive collaborative optimization model, the task sequence, trajectory path and flight parameters of the drone are updated.

[0009] After each round of inspection, based on the coverage data recorded by the UAV during the inspection and the updated flight parameters, the coverage rate and missing area indicators are calculated, and mission planning and trajectory adjustment data are generated.

[0010] Preferably, the process of constructing the collaborative capability matrix is ​​as follows:

[0011] Based on variable-size operating units and real-time status information of each UAV, including remaining power, payload margin, maximum flight time, flight speed and communication signal quality, the mission execution capability of the UAV is quantitatively evaluated.

[0012] By combining the communication link topology between UAVs and the mission synchronization delay parameters, a collaborative capability matrix reflecting the degree of multi-UAV cooperation coupling is constructed.

[0013] During the matrix construction process, the complementary performance of different UAVs in collaborative execution is normalized by introducing energy consumption balance factors and task coupling coefficients.

[0014] Ultimately, a collaborative capability matrix is ​​formed to match the suitability of operational units with UAV missions.

[0015] Preferably, the process of generating a matching map between multiple UAVs and operational units is as follows:

[0016] Based on the comprehensive execution index of each UAV in the collaborative capability matrix and the coverage priority of the operation unit, a two-layer matching mechanism is adopted to calculate the matching weight between UAVs and operation units, and form an initial matching relationship matrix.

[0017] Based on accessibility constraints, and considering the drone's energy reserves, flight radius, and communication signal coverage, a set of operational units that each drone can perform is selected.

[0018] Based on the screening results, the optimal matching algorithm is used to comprehensively solve the matching weights between the UAV and the operation unit to determine the matching relationship;

[0019] The matching results are transformed into a graph structure, recording the allocation relationship, spatial location association, and communication relay link information between UAVs and work units, forming a matching graph of multiple UAVs and work units.

[0020] Preferably, the process of predicting track conflicts, coverage redundancy, and energy consumption deviations through multiple factors is as follows:

[0021] Based on the allocation relationship between UAVs and work units in the matching map, the initial trajectory data of each UAV is extracted, and trajectory simulation analysis is performed according to the task execution order.

[0022] Based on the spatiotemporal coordinate information of the flight tracks, the intersection areas, adjacent flight areas and communication interference areas between flight tracks are identified, and a flight track interaction dataset is established.

[0023] Using spatiotemporal prediction calculation methods, the probability of conflict is estimated for each track interaction dataset, and the coverage redundancy index is calculated by combining the spatial overlap area between operation units and the sensor coverage radius.

[0024] Based on the drone's flight speed, attitude changes, turning frequency, and load change parameters, the energy consumption rate is calculated, and an energy consumption deviation index is generated.

[0025] By integrating the probability of flight path conflict, coverage redundancy index, and energy consumption deviation index with weights, a multi-factor prediction dataset is formed.

[0026] Preferably, the process of dynamically adjusting the UAV path priority, flight altitude, and heading offset based on the prediction results is as follows:

[0027] Based on a multi-factor prediction dataset, the path priority score of each UAV in the collaborative coverage task is calculated.

[0028] An adaptive path adjustment model is introduced to perform flight altitude level separation for UAVs with high risk of track conflict, and adjust the heading offset according to the area density.

[0029] By combining real-time communication link bandwidth and feedback latency, the path execution order is dynamically corrected;

[0030] The effectiveness of path adjustments is periodically evaluated through a rolling update mechanism, and the optimization direction is corrected based on coverage redundancy and energy consumption balance indicators.

[0031] Preferably, the process of performing incremental replanning for the affected local work units is as follows:

[0032] When coverage omissions, duplicate coverages, or communication interruptions are detected in real time, the set of job units corresponding to the abnormal area is locked.

[0033] Based on the adaptive collaborative optimization model, the spatial dependence and trajectory interaction relationship between the affected unit and its neighboring units are analyzed to determine the scope of the replanning impact.

[0034] An incremental planning algorithm is used to reallocate local tracks without affecting the overall task allocation structure;

[0035] Simultaneously, the scheduling table is updated and the local path cost function is recalculated by combining the remaining battery power of the drone with the task priority.

[0036] Preferably, the process of generating mission planning and trajectory adjustment data is as follows:

[0037] Summarize the coverage information and flight status data collected by the drone sensors during the inspection process, including heading, speed, flight altitude, turning frequency and load parameters;

[0038] Spatial mapping and statistics are performed on the coverage information to calculate the actual coverage rate and the indicators of uncovered or overlapping areas for each work unit.

[0039] Process the UAV flight data and calculate the energy consumption index and flight path status parameters of each UAV during the inspection cycle;

[0040] The coverage index, the missing area index, and the drone flight parameters are integrated to form a structured data set.

[0041] A multi-drone collaborative full-coverage path planning and optimization system, including:

[0042] Area scanning module: Performs continuous grid scanning of the inspection area and generates variable-size work units by combining terrain and obstacle information;

[0043] Capability matching module: Evaluates the battery power, load, speed and communication status of each UAV, constructs a collaborative capability matrix and generates a matching map between UAVs and operational units;

[0044] Collaborative optimization module: Based on the matching map, an adaptive collaborative optimization model is introduced to predict track conflicts and energy consumption deviations, and dynamically adjust path priorities and flight parameters;

[0045] Dynamic replanning of blocks: During the inspection process, missed or duplicate covered areas are detected, and the task sequence and path are incrementally replanned.

[0046] Results evaluation module: After the inspection is completed, the coverage rate and omission indicators are calculated, and mission planning and track adjustment data are generated.

[0047] Compared with existing technologies, this invention provides a method and system for multi-UAV collaborative full-coverage path planning optimization, which has the following beneficial effects:

[0048] This invention achieves efficient, full-coverage path planning for multiple UAVs in complex environments by introducing variable-size work unit partitioning and an adaptive collaborative optimization model. It can adaptively adjust task allocation and flight parameters based on terrain complexity, obstacle distribution, and real-time communication status, significantly improving the coverage integrity and energy efficiency of inspection tasks. During execution, it can dynamically detect missed and duplicate coverage areas and achieve real-time path optimization through an incremental replanning mechanism, effectively reducing the probability of flight path conflicts and redundant operations. By introducing a collaborative capability matrix and coverage priority evaluation, intelligent scheduling and load balancing are achieved among multiple UAVs, improving group collaboration efficiency. The coverage efficiency and risk distribution data generated after each round of inspection can be used to form a closed-loop optimization mechanism, enabling the path planning algorithm to have continuous learning and adaptive evolution capabilities. It is suitable for various complex application scenarios such as power line inspection, emergency search and rescue, and farmland monitoring, and possesses high reliability, flexibility, and environmental adaptability. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the method of the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0051] 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.

[0052] Example 1: Please refer to Figure 1 As shown, the multi-UAV collaborative full-coverage path planning optimization method in this embodiment of the invention includes the following steps:

[0053] S1: Perform continuous gridded scanning of the inspection area and automatically generate variable-size operation units based on local terrain complexity, obstacle density, and drone sensor coverage radius;

[0054] The process of automatically generating variable-size work cells in S1 is as follows:

[0055] Based on the geographic information data and 3D terrain information of the inspection area, the inspection area is divided into layers and grids according to the effective detection radius and resolution threshold of the UAV sensor. Based on the geographic information data and 3D terrain model of the inspection area, the spatial boundary, terrain undulation and obstacle distribution information of the inspection area are obtained. According to the effective detection radius and resolution threshold of the UAV sensor, the inspection area is divided into layers and grids, and the spatial coordinates, area, boundary information and corresponding height level of each grid unit are recorded.

[0056] The terrain complexity index and obstacle density function are used to dynamically adjust the size of each grid cell;

[0057] The size of the initially generated grid cells is optimized using terrain complexity index and obstacle distribution density. For areas with large terrain undulations or dense obstacles, the size of the grid cells is reduced to ensure coverage accuracy. For areas with flat terrain and sparse obstacles, the size of the grid cells can be appropriately increased to reduce the number of tasks. Each adjusted grid cell retains spatial boundary and terrain attribute information, and the relationship between adjacent cells is updated.

[0058] During the generation process, grid adjacency analysis and boundary fusion are used to output variable-size work units with spatial coordinate attributes, terrain labels, and coverage priority parameters. In the process of generating work units, the adjacency relationship between grid units is analyzed to identify possible boundary intersections or overlapping areas. By merging or correcting adjacent grid boundaries, it is ensured that the work units continuously cover the inspection area, while retaining spatial coordinates, terrain labels, and coverage priority parameters. Finally, a set of structured variable-size work units is formed, with each unit containing coordinates, size, terrain features, and coverage priority information for use in UAV mission planning and path generation.

[0059] S2: Based on the variable-size work unit, evaluate the real-time power, load capacity, speed limit and communication link status of each UAV, construct a collaborative capability matrix, and generate a matching map of multiple UAVs and work units by combining the coverage priority and reachability of the work unit.

[0060] The process of constructing the collaborative capability matrix in S2 is as follows:

[0061] Based on variable-size work units and real-time status information of each UAV, including remaining battery power, payload margin, maximum endurance, flight speed, and communication signal quality, the mission execution capability of the UAVs is quantitatively evaluated. Based on variable-size work units, real-time status information of each UAV is collected, including remaining battery power, payload margin, maximum endurance, flight speed, and communication signal quality parameters, to quantify the capability of each UAV in performing inspection tasks. The energy consumption, time, and communication resource consumption required for each UAV to perform tasks in different work units are calculated.

[0062] By combining the communication link topology and mission synchronization delay parameters between UAVs, a collaborative capability matrix reflecting the degree of multi-UAV cooperation coupling is constructed. Using the communication link topology and mission synchronization delay parameters between UAVs, the degree of cooperation coupling between UAVs is calculated. Each element in the matrix represents the cooperation strength between two UAVs when performing a mission, including communication dependency, synchronization delay and potential conflict probability, and the multi-UAV cooperation relationship is represented in a structured manner.

[0063] During matrix construction, the complementary performance of different UAVs in collaborative execution is normalized by introducing energy consumption balance factors and task coupling coefficients. In the matrix construction process, energy consumption balance factors and task coupling coefficients are introduced to normalize the collaborative capability data between UAVs. By standardizing the energy consumption and task coupling degree of different UAVs, the difference in dimensions is eliminated, so that the matrix can accurately reflect the complementary performance and relative capabilities of each UAV in collaborative tasks.

[0064] Finally, a collaborative capability matrix is ​​formed to match the suitability of work units and UAV tasks. The task execution capability indicators, normalized collaborative coupling data and energy consumption balance information are integrated to generate the final collaborative capability matrix. The matrix is ​​stored in the form of structured data, with each row representing a UAV and each column corresponding to a work unit. The matrix element values ​​reflect the suitability and collaborative capability of the UAV in each work unit task, and are used for matching UAVs with work units and path planning calculations.

[0065] The process of generating a matching map of multiple UAVs and work units in S2 is as follows:

[0066] Based on the comprehensive execution index of each UAV in the collaborative capability matrix and the coverage priority of the work unit, a two-layer matching mechanism is adopted to calculate the matching weight between UAVs and work units and form an initial matching relationship matrix. The comprehensive execution index of each UAV is extracted from the collaborative capability matrix. This index comprehensively considers parameters such as the UAV's remaining battery power, payload capacity, flight speed and communication status. For each work unit, the coverage priority parameter is obtained. Based on the two-layer matching mechanism, the first layer scores the compatibility between each UAV and each work unit, and the second layer adjusts the score in combination with the work unit priority to generate the initial matching weight matrix of UAV-work unit.

[0067] Based on accessibility constraints, and considering the drone's remaining energy, flight radius, and communication signal coverage, a set of work units that each drone can perform is selected. Accessibility constraints are introduced to assess whether each drone can reach and execute each work unit based on its remaining battery power, flight radius, and communication signal coverage. This is achieved through methods including spatial distance calculation, energy consumption prediction, and communication coverage detection. Work units exceeding flight radius or energy limits, or with insufficient signal, are excluded, ultimately forming a candidate set of work units for each drone's executable tasks.

[0068] Based on the screening results, the optimal matching algorithm is used to comprehensively solve the matching weights between UAVs and task units to determine the matching relationship. Within the screened candidate set, the matching weights between UAVs and task units are optimized. Using an optimal matching algorithm, such as the Hungarian algorithm or other iterative optimization methods, and comprehensively considering the weight matrix and the UAV's task carrying capacity, the algorithm calculates which UAV should be assigned to each task unit. The algorithm output ensures that all task units are assigned while balancing the UAV's task load.

[0069] The matching results are transformed into a graph structure, recording the allocation relationship, spatial location association, and communication relay link information between UAVs and work units, forming a matching graph of multiple UAVs and work units. The final matching results are then transformed into a graph data structure, where each node represents a UAV or work unit, and each edge represents task allocation and spatial association. The graph records the allocation information, spatial coordinates, and necessary communication link parameters of UAVs and work units, facilitating use in path planning and task scheduling. The generated graph provides structured data support for multi-UAV collaborative tasks.

[0070] S3: Based on variable-size work units and matching maps, an adaptive collaborative optimization model is introduced to predict track conflicts, coverage redundancy and energy consumption deviations through multiple factors, and dynamically adjust the UAV path priority order, flight altitude and heading offset according to the prediction results.

[0071] The process of predicting track conflicts, coverage redundancy, and energy consumption deviations through multiple factors in S3 is as follows:

[0072] Based on the allocation relationship between UAVs and task units in the matching graph, initial trajectory data for each UAV is extracted, and trajectory simulation analysis is performed according to the task execution order. According to the allocation results of UAVs and task units in the matching graph, initial trajectory data is generated according to the task execution order of each UAV. The trajectory data includes the spatial coordinates, flight altitude, speed, and timestamp information of the UAV in each task unit. Using trajectory simulation tools, the task sequence is converted into a continuous three-dimensional path for subsequent trajectory interaction analysis.

[0073] Based on the spatiotemporal coordinate information of flight tracks, we identify the intersection areas, adjacent flight areas, and communication interference areas between flight tracks, and establish a flight track interaction dataset. Using the spatiotemporal coordinate information of flight tracks, we perform spatial adjacency and intersection analysis on the trajectories between UAVs. By calculating the distance matrix between flight tracks, we identify intersection points and close-range flight areas, and at the same time, we assess potential interference areas based on communication link characteristics. We integrate this information to form the flight track interaction dataset.

[0074] Using spatiotemporal prediction calculation methods, the probability of conflict is estimated for each track interaction dataset. The coverage redundancy index is calculated by combining the spatial overlap area between work units and the sensor coverage radius. The spatiotemporal prediction algorithm is applied to the track interaction dataset, and the probability of conflict between different tracks in the future planning period is estimated by combining the UAV speed and mission execution time. At the same time, the coverage redundancy index is quantified based on the spatial overlap area of ​​adjacent work units and the sensor coverage radius to reflect the degree of repeated coverage of the mission area.

[0075] Based on the parameters of the UAV's flight speed, attitude change, turning frequency, and load change, the energy consumption rate is calculated, and an energy consumption deviation index is generated. Based on the parameters of the UAV's flight speed, attitude change, turning frequency, and load change, the energy consumption rate is calculated. By accumulating the consumption of each flight segment, the energy consumption index of each UAV within the planning period is obtained. This index is then compared with the expected energy consumption to form an energy consumption deviation index, which is used to evaluate the energy deviation of the UAV in performing the mission.

[0076] The probability of flight path conflict, coverage redundancy index and energy consumption deviation index are integrated with weights to form a multi-factor prediction dataset; the probability of flight path conflict, coverage redundancy index and energy consumption deviation index are integrated with predetermined weights to form a structured multi-factor prediction dataset, including the conflict risk, redundancy coverage degree and energy consumption deviation of each UAV in each operation unit.

[0077] The process in S3 that dynamically adjusts the UAV's path priority, flight altitude, and heading offset based on the prediction results is as follows:

[0078] Based on a multi-factor prediction dataset, the path priority score of each UAV in the collaborative coverage task is calculated. Using the multi-factor prediction dataset, including the probability of track conflict, coverage redundancy index and energy consumption deviation index, the task execution sequence of each UAV is quantitatively analyzed. Through a weighted calculation method, the priority of each UAV in the collaborative coverage task is scored, and a path priority score matrix is ​​generated to provide a basis for subsequent adjustments.

[0079] An adaptive path adjustment model is introduced to separate flight altitude levels for UAVs with a high risk of track conflict and adjust the heading offset according to the area density. For UAVs with a high probability of track conflict, different flight altitude levels are allocated in three-dimensional space through the adaptive path adjustment model to avoid track intersection. At the same time, combined with the density information of the operation area, the heading offset is adjusted according to the local UAV distribution to form an optimized track trajectory, ensuring that the spatial distribution of each UAV mission is reasonable.

[0080] By combining real-time communication link bandwidth and feedback latency, the path execution order is dynamically corrected. During flight, the real-time communication link bandwidth and feedback latency between UAVs are monitored, and the path execution order of the UAVs is dynamically adjusted according to the link status and mission execution progress to ensure that the mission order can be updated in real time, while avoiding track conflicts or mission duplication caused by communication latency.

[0081] A rolling update mechanism is used to periodically evaluate the effectiveness of path adjustments and correct the optimization direction based on coverage redundancy and energy consumption balance indicators. This mechanism also periodically collects actual UAV flight paths, mission completion status, and coverage data. The effects of the adjusted flight paths are evaluated, including changes in conflict probability, coverage redundancy, and energy consumption, providing data support for the next cycle of path optimization. Based on each rolling evaluation, coverage redundancy indicators and energy consumption deviation information are used to further adjust the UAV path priority, flight altitude, and heading offset. Through iterative correction, a balanced task distribution and energy consumption control for collaborative coverage tasks across multiple UAVs are achieved.

[0082] S4: During the inspection process, collect drone coverage information and abnormal status in real time. When missing or duplicate coverage data is found, perform incremental replanning on the affected local operation units, and update the drone's task sequence, trajectory path and flight parameters in combination with the adaptive collaborative optimization model.

[0083] The process of performing incremental replanning for affected local work units in S4 is as follows:

[0084] When coverage omissions, duplicate coverage, or communication interruptions are detected in real time, the set of work units corresponding to the abnormal area is locked. During the inspection, coverage information is collected in real time by drone sensor data and combined with communication link status monitoring to determine whether there are coverage omissions, duplicate coverage, or communication interruptions in the work units. For areas where abnormalities are detected, spatial coordinates and work unit identifiers are recorded, and these abnormal units are set into a list to be replanned.

[0085] Based on the adaptive collaborative optimization model, the spatial dependency and trajectory interaction relationship between the affected unit and its neighboring units are analyzed to determine the scope of impact of replanning. Based on the adaptive collaborative optimization model, the spatial dependency relationship between the unit to be replanned and its neighboring units is analyzed, including the overlap of adjacent units, the intersection of flight paths, and the constraints of the task execution order. At the same time, combined with UAV trajectory interaction data, it is assessed which neighboring units may be affected, thereby defining the spatial range of incremental replanning.

[0086] An incremental planning algorithm is used to fine-tune or redistribute local tracks without affecting the overall task allocation structure. The algorithm also fine-tunes or redistributes locked local task unit tracks without changing the overall task allocation structure. This involves adjusting track node coordinates, modifying flight order, recalculating path connection points, and updating the cooperation interval between UAVs, ensuring that local tasks can be smoothly connected to the overall track.

[0087] Simultaneously, the scheduling table is updated and the local path cost function is recalculated based on the remaining battery power of the drones and the task priority. After the local trajectory adjustment is completed, the scheduling table is updated based on the remaining battery power, flight speed and task priority of each drone, and the task order is reallocated. At the same time, the local path cost function, including flight distance, energy consumption and coverage efficiency parameters, is calculated based on the adjusted trajectory data.

[0088] S5: After each round of inspection is completed, based on the coverage data recorded by the UAV during the inspection and the updated flight parameters, calculate the coverage rate and missing area indicators, and generate mission planning and trajectory adjustment data.

[0089] The process of generating mission planning and trajectory adjustment data in S5 is as follows:

[0090] The system summarizes coverage information and flight status data collected by UAV sensors during the inspection process, including heading, speed, flight altitude, turning frequency, and load parameters. During the inspection, the UAV's onboard sensors collect coverage information in real time, including the coverage status of the work unit and sensor detection results. Simultaneously, it records flight status parameters, including heading angle, flight speed, altitude, turning frequency, and load changes. The collected data is stored according to timestamps and UAV identifiers, forming a preliminary raw data set.

[0091] The coverage information is spatially mapped and statistically analyzed to calculate the actual coverage rate and the index of uncovered or over-covered areas for each work unit. The collected coverage information is mapped to the spatial coordinates of the work units in the inspection area. By comparing the coordinates and calculating the coverage radius, the actual coverage of each work unit is determined. The uncovered or over-covered areas are marked and their areas are calculated. The coverage rate index and the index of the missing areas for each work unit are generated, and a spatial statistical table that can be used for task adjustment is formed.

[0092] The system processes UAV flight data to calculate energy consumption indicators and trajectory status parameters for each UAV during the inspection cycle. It also processes flight data during the inspection process, including trajectory path analysis, speed and altitude change curve calculation, and turning frequency statistics. Combining the UAV's load parameters and flight time, the system calculates the energy consumption indicators for each UAV during the inspection cycle using a predefined energy consumption model and records trajectory status parameters such as path node sequence, flight interval, and speed changes.

[0093] The coverage index, the missing area index, and the UAV flight parameters are integrated to form a structured dataset. The obtained coverage index, missing area index, and UAV flight parameters are then integrated and associated with each work unit and UAV identifier. The final output is a structured dataset, including coverage information, UAV trajectory parameters, and energy consumption data for each work unit.

[0094] Example 2: Figure 2 As shown, the multi-UAV collaborative full-coverage path planning and optimization system includes:

[0095] Area scanning module: Performs continuous grid scanning of the inspection area and generates variable-size work units by combining terrain and obstacle information;

[0096] Capability matching module: Evaluates the battery power, load, speed and communication status of each UAV, constructs a collaborative capability matrix and generates a matching map between UAVs and operational units;

[0097] Collaborative optimization module: Based on the matching map, an adaptive collaborative optimization model is introduced to predict track conflicts and energy consumption deviations, and dynamically adjust path priorities and flight parameters;

[0098] Dynamic replanning of blocks: During the inspection process, missed or duplicate covered areas are detected, and the task sequence and path are incrementally replanned.

[0099] Results evaluation module: After the inspection is completed, the coverage rate and omission indicators are calculated, and mission planning and track adjustment data are generated.

[0100] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0101] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-UAV collaborative full-coverage path planning optimization method, characterized in that, Includes the following steps: The inspection area is continuously scanned in a grid pattern, and variable-size operation units are automatically generated based on local terrain complexity, obstacle density, and drone sensor coverage radius. Based on the variable-size work unit, the real-time power, load capacity, speed limit and communication link status of each UAV are evaluated, a collaborative capability matrix is ​​constructed, and a matching map of multiple UAVs and work units is generated by combining the coverage priority and reachability of the work unit. Based on variable-size work units and matching maps, an adaptive collaborative optimization model is introduced to predict track conflicts, coverage redundancy and energy consumption deviations through multiple factors, and dynamically adjust the UAV path priority order, flight altitude and heading offset according to the prediction results. During the inspection, the drone coverage information and abnormal status are collected in real time. When missing or duplicate coverage data is found, incremental replanning is performed on the affected local operation units. Combined with the adaptive collaborative optimization model, the task sequence, trajectory path and flight parameters of the drone are updated. After each round of inspection, based on the coverage data recorded by the UAV during the inspection and the updated flight parameters, the coverage rate and missing area indicators are calculated, and mission planning and trajectory adjustment data are generated.

2. The multi-UAV collaborative full-coverage path planning optimization method according to claim 1, characterized in that, The process of automatically generating variable-size work cells is as follows: Based on the geographic information data and three-dimensional terrain information of the inspection area, the inspection area is divided into layers and blocks in a grid according to the effective detection radius and resolution threshold of the UAV sensor. The terrain complexity index and obstacle density function are used to dynamically adjust the size of each grid cell; During the generation process, through grid adjacency analysis and boundary fusion, variable-size job cells with spatial coordinate attributes, terrain labels and coverage priority parameters are output.

3. The multi-UAV collaborative full-coverage path planning optimization method according to claim 2, characterized in that, The process of constructing the collaborative capability matrix is ​​as follows: Based on variable-size operating units and real-time status information of each UAV, including remaining power, payload margin, maximum flight time, flight speed and communication signal quality, the mission execution capability of the UAV is quantitatively evaluated. By combining the communication link topology between UAVs and the mission synchronization delay parameters, a collaborative capability matrix reflecting the degree of multi-UAV cooperation coupling is constructed. During the matrix construction process, the complementary performance of different UAVs in collaborative execution is normalized by introducing energy consumption balance factors and task coupling coefficients. Ultimately, a collaborative capability matrix is ​​formed to match the suitability of operational units with UAV missions.

4. The multi-UAV collaborative full-coverage path planning optimization method according to claim 3, characterized in that, The process of generating a matching map of multiple UAVs and operational units is as follows: Based on the comprehensive execution index of each UAV in the collaborative capability matrix and the coverage priority of the operation unit, a two-layer matching mechanism is adopted to calculate the matching weight between UAVs and operation units, and form an initial matching relationship matrix. Based on accessibility constraints, and considering the drone's energy reserves, flight radius, and communication signal coverage, a set of operational units that each drone can perform is selected. Based on the screening results, the optimal matching algorithm is used to comprehensively solve the matching weights between the UAV and the operation unit to determine the matching relationship; The matching results are transformed into a graph structure, recording the allocation relationship, spatial location association, and communication relay link information between UAVs and work units, forming a matching graph of multiple UAVs and work units.

5. The multi-UAV collaborative full-coverage path planning optimization method according to claim 4, characterized in that, The process of predicting track conflicts, coverage redundancy, and energy consumption deviations through multiple factors is as follows: Based on the allocation relationship between UAVs and work units in the matching map, the initial trajectory data of each UAV is extracted, and trajectory simulation analysis is performed according to the task execution order. Based on the spatiotemporal coordinate information of the flight tracks, the intersection areas, adjacent flight areas and communication interference areas between flight tracks are identified, and a flight track interaction dataset is established. Using spatiotemporal prediction calculation methods, the probability of conflict is estimated for each track interaction dataset, and the coverage redundancy index is calculated by combining the spatial overlap area between operation units and the sensor coverage radius. Based on the drone's flight speed, attitude changes, turning frequency, and load change parameters, the energy consumption rate is calculated, and an energy consumption deviation index is generated. By integrating the probability of flight path conflict, coverage redundancy index, and energy consumption deviation index with weights, a multi-factor prediction dataset is formed.

6. The multi-UAV collaborative full-coverage path planning optimization method according to claim 5, characterized in that, The process of dynamically adjusting the UAV's path priority, flight altitude, and heading offset based on the prediction results is as follows: Based on a multi-factor prediction dataset, the path priority score of each UAV in the collaborative coverage task is calculated. An adaptive path adjustment model is introduced to perform flight altitude level separation for UAVs with high risk of track conflict, and adjust the heading offset according to the area density. By combining real-time communication link bandwidth and feedback latency, the path execution order is dynamically corrected; The effectiveness of path adjustments is periodically evaluated through a rolling update mechanism, and the optimization direction is corrected based on coverage redundancy and energy consumption balance indicators.

7. The multi-UAV cooperative full-coverage path planning optimization method according to claim 6, characterized in that, The process of performing incremental replanning for the affected local work units is as follows: When coverage omissions, duplicate coverages, or communication interruptions are detected in real time, the set of job units corresponding to the abnormal area is locked. Based on the adaptive collaborative optimization model, the spatial dependence and trajectory interaction relationship between the affected unit and its neighboring units are analyzed to determine the scope of the replanning impact. An incremental planning algorithm is used to reallocate local tracks without affecting the overall task allocation structure; Simultaneously, the scheduling table is updated and the local path cost function is recalculated by combining the remaining battery power of the drone with the task priority.

8. The multi-UAV collaborative full-coverage path planning optimization method according to claim 7, characterized in that, The process of generating mission planning and trajectory adjustment data is as follows: Summarize the coverage information and flight status data collected by the drone sensors during the inspection process, including heading, speed, flight altitude, turning frequency and load parameters; Spatial mapping and statistics are performed on the coverage information to calculate the actual coverage rate and the indicators of uncovered or overlapping areas for each work unit. Process the UAV flight data and calculate the energy consumption index and flight path status parameters of each UAV during the inspection cycle; By integrating coverage metrics, omission area metrics, and UAV flight parameters, mission planning and trajectory adjustment data are generated.

9. A multi-UAV collaborative full-coverage path planning and optimization system, applied to the method described in any one of claims 1-8, characterized in that, include: Area scanning module: Performs continuous grid scanning of the inspection area and generates variable-size work units by combining terrain and obstacle information; Capability matching module: Evaluates the battery power, load, speed and communication status of each UAV, constructs a collaborative capability matrix and generates a matching map between UAVs and operational units; Collaborative optimization module: Based on the matching map, an adaptive collaborative optimization model is introduced to predict track conflicts and energy consumption deviations, and dynamically adjust path priorities and flight parameters; Dynamic replanning of blocks: During the inspection process, missed or duplicate covered areas are detected, and the task sequence and path are incrementally replanned. Results evaluation module: After the inspection is completed, the coverage rate and omission indicators are calculated, and mission planning and track adjustment data are generated.

Images