Multi-unmanned aerial vehicle dynamic path planning system and method

Abstract

The invention provides a multi-unmanned aerial vehicle dynamic path planning system and method, and the system comprises a coordinate conversion module which is used for carrying out the bidirectional conversion of latitude and longitude coordinates and three-dimensional grid coordinates, so as to establish a unified space-time coordinate conversion reference; the map module is used for constructing a three-dimensional airspace representation model; the three-dimensional airspace representation model records position information of an obstacle and boundary range information of the temporary no-fly zone so as to provide real-time airspace information; the route module is used for delimiting an exclusive safety area for each unmanned aerial vehicle in the air; the path planning module is used for performing multi-machine path planning based on the real-time airspace information and the space-time occupancy state; the map module, the route module and the path planning module work cooperatively based on a unified space-time coordinate conversion reference. Through unified three-dimensional grid map representation and a space-time occupation management system, integrated coordination optimization of the whole process from flight plan approval to flight execution is realized, and real-time dynamic management of a large-scale multi-machine system is supported.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicle management, and particularly relates to a multi-unmanned aerial vehicle dynamic path planning system and method. BACKGROUND

[0002] The existing multi-unmanned aerial vehicle management system usually adopts a hierarchical architecture design, and divides airspace management into two relatively independent stages of flight plan approval and flight execution. In the flight plan approval stage, the system mainly relies on a preset fixed route to verify the path, and avoids conflicts through simple time window allocation; in the flight execution stage, the system adopts a priority-based scheduling method to handle real-time conflicts, or uses a CBS (Conflict-Based Search) algorithm for local re-planning. Such architecture usually adopts a two-dimensional grid extension or a simple three-dimensional voxel segmentation in map representation, and lacks an efficient modeling method for complex airspace environments and dynamic no-fly zones.

[0003] It can be seen that the existing multi-unmanned aerial vehicle management system has significant systematic defects, mainly manifested in the lack of a unified space-time coordination framework and full-life-cycle management capability. The functional modules of the system lack organic cooperation, and are independent of each other, which cannot realize real-time sharing and coordinated optimization of information, resulting in a performance far lower than the theoretical level.

[0004] Therefore, how to solve the systematic defects of the independent functional modules and the lack of organic cooperation in the prior art is a problem to be solved at present. SUMMARY

[0005] The present application provides a multi-unmanned aerial vehicle dynamic path planning system and method, which solves the defects of independent functional modules and lack of organic cooperation in the prior art, realizes integrated coordination and optimization from flight plan approval to flight execution through the establishment of a unified three-dimensional grid map representation and space-time occupation management system, and supports real-time dynamic management of large-scale multi-vehicle systems.

[0006] The present application provides a multi-unmanned aerial vehicle dynamic path planning system, which comprises: A coordinate conversion module for bidirectional conversion of latitude-longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion benchmark; A map module for constructing a three-dimensional airspace representation model; the three-dimensional airspace representation model records spatial position information of obstacles and boundary range information of temporary no-fly zones to provide real-time airspace information; A route module for demarcating an exclusive safety area for each unmanned aerial vehicle in space-time, and updating the demarcated safety area information to a space-time occupation map; the space-time occupation map is used to record the space-time occupation state of each unmanned aerial vehicle; a path planning module, configured to perform multi-unmanned aerial vehicle path planning based on the real-time airspace information and the space-time occupation state; The map module, the flight path module, and the path planning module work cooperatively based on the unified space-time coordinate conversion reference.

[0007] According to the multi-unmanned aerial vehicle dynamic path planning system provided in the application, the coordinate conversion module is specifically configured to: The flight space region is divided into a plurality of grids by using a preset minimum coordinate reference point and a preset grid size parameter, and each grid is assigned a grid coordinate, and a latitude-longitude-grid coordinate conversion pair is cached; When receiving a coordinate to be converted, a grid-based direct linear mapping algorithm and a caching optimization technique are used to perform bidirectional conversion between latitude-longitude coordinates and grid coordinates on the coordinate to be converted by using the preset minimum coordinate reference point and the grid size parameter. The caching optimization technique is used to cache frequently accessed latitude-longitude-grid coordinate conversion pairs; the frequently accessed latitude-longitude-grid coordinate conversion pairs refer to latitude-longitude-grid coordinate conversion pairs with a number of accesses greater than a preset access number threshold.

[0008] According to the multi-unmanned aerial vehicle dynamic path planning system provided in the application, the map module is specifically configured to determine a region surrounded by a plurality of boundary points as boundary range information of a temporary no-fly zone, and store space position information of obstacles by using a sparse storage strategy. The map module is further configured to work cooperatively with the flight path module, share unified airspace information, and perform incremental updating on fixed obstacles and dynamic no-fly zones.

[0009] According to the multi-unmanned aerial vehicle dynamic path planning system provided in the application, the flight path module is specifically configured to define an exclusive safety region for each unmanned aerial vehicle in space-time based on flight time interval management and four-dimensional safety region envelope design. The system further includes a conflict detection module integrated with the map module and the flight path module, configured to perform time region overlap detection algorithm, and perform fixed obstacle, dynamic no-fly zone, and multi-unmanned aerial vehicle conflict identification by real-time querying and updating of a space-time occupation graph.

[0010] According to the multi-unmanned aerial vehicle dynamic path planning system provided in the application, the system further includes a risk assessment module; the risk assessment module is configured to: Share information with the map module and the flight path module, and comprehensively evaluate spatial conflict and space-time occupation conflict of flight paths of all unmanned aerial vehicles; The flight plan is analyzed and pre-evaluated in a flight plan approval stage, and the flight states of multiple unmanned aerial vehicles are monitored in real time in a flight execution stage.

[0011] The path planning module is specifically used for: An improved three-dimensional jump point search algorithm is adopted to search for an optimal flight path from a flight starting point to a flight ending point in a three-dimensional airspace; A vertical obstacle avoidance strategy is adopted to consider the height information of obstacles in the path search process, so that the planned path avoids obstacles in the vertical direction; A multi-level cache optimization strategy is adopted to cache frequently used flight paths; the frequently used flight paths refer to flight paths with a number of access times greater than a preset use number threshold.

[0012] The system further comprises a dynamic rerouting module; the dynamic rerouting module is used for: The path planning module and the conflict detection module work cooperatively to re-plan a path when receiving an alarm information; When detecting that a temporary no-fly zone appears in a front target region, the path segment belonging to the target region is re-planned.

[0013] The application further provides a multi-unmanned aerial vehicle dynamic path planning method, and the method comprises: Bidirectional conversion is performed on latitude and longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion benchmark; A three-dimensional airspace representation model is constructed; the three-dimensional airspace representation model records the spatial position information of obstacles and the boundary range information of temporary no-fly zones to provide real-time airspace information; An exclusive safety area is demarcated for each unmanned aerial vehicle in space-time, and the demarcated safety area information is updated to a space-time occupancy graph; the space-time occupancy graph is used to record the space-time occupancy state of each unmanned aerial vehicle; The real-time airspace information and the space-time occupancy state are used for multi-unmanned aerial vehicle path planning; The real-time airspace information, the space-time occupancy state of each unmanned aerial vehicle and the multi-unmanned aerial vehicle path planning are realized based on the unified space-time coordinate conversion benchmark.

[0014] The application further provides an electronic device comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor; when the processor executes the computer program, the method for multi-unmanned aerial vehicle dynamic path planning is realized.

[0015] The application provides a multi-unmanned aerial vehicle dynamic path planning system and method. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the application or the prior art, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the application, and other drawings can be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0017] Figure 1 FIG. 1 is a structural schematic diagram of a multi-unmanned aerial vehicle dynamic path planning system provided by an embodiment of the application.

[0018] Figure 2 FIG. 2 is another structural schematic diagram of a multi-unmanned aerial vehicle dynamic path planning system provided by an embodiment of the application.

[0019] Figure 3 FIG. 3 is a flow schematic diagram of a multi-unmanned aerial vehicle dynamic path planning method provided by an embodiment of the application.

[0020] Figure 4 FIG. 4 is a structural schematic diagram of an electronic device provided by an embodiment of the application. DETAILED DESCRIPTION

[0021] In order to make the objects, technical solutions and advantages of the application clearer, the technical solutions in the application will be described clearly and completely in combination with the drawings in the application. Obviously, the described embodiments are some embodiments of the application, rather than all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without any creative effort belong to the protection scope of the application.

[0022] In the description of the embodiments of the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or equipment. The specific meaning of the above terms in the present application can be understood by the person skilled in the art according to the specific circumstances.

[0023] Figure 1 is one of the structural schematic diagrams of the multi-unmanned aerial vehicle dynamic path planning system provided by the embodiments of the present application. Referring to Figure 1 , the embodiments of the present application provide a multi-unmanned aerial vehicle dynamic path planning system, which can specifically include: A coordinate conversion module 110 is configured to perform bidirectional conversion between latitude-longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion reference. A map module 120 is configured to construct a three-dimensional airspace representation model; the three-dimensional airspace representation model records spatial position information of obstacles and boundary range information of temporary no-fly zones to provide real-time airspace information. A flight path module 130 is configured to demarcate an exclusive safety area for each unmanned aerial vehicle in space-time and update the demarcated safety area information to a space-time occupancy map; the space-time occupancy map is configured to record space-time occupancy states of each unmanned aerial vehicle. A path planning module 140 is configured to perform multi-unmanned aerial vehicle path planning based on the real-time airspace information and the space-time occupancy states. The map module 120, the flight path module 130 and the path planning module 140 are configured to cooperatively work based on the unified space-time coordinate conversion reference.

[0024] The existing multi-unmanned aerial vehicle management system lacks a unified coordinate reference and space-time occupancy management mechanism, different data formats and processing methods are used in different modules, frequent data conversion and format matching become a system performance bottleneck, and seriously affect the real-time response capability. In view of the above problems, the embodiments of the present application perform bidirectional conversion between latitude-longitude coordinates and three-dimensional grid coordinates through the coordinate conversion module, establish a unified space-time coordinate conversion reference, and all modules in the multi-unmanned aerial vehicle dynamic path planning system cooperatively work based on the bidirectional conversion mechanism of the WGS84 (World Geodetic System 1984, 1984 World Geodetic System) coordinate system and three-dimensional grid coordinates (i.e. the unified space-time coordinate conversion reference), thereby avoiding the data conversion bottleneck caused by different coordinate references of modules in the traditional unmanned aerial vehicle management system.

[0025] The WGS84 coordinate system can include parameters of three dimensions of longitude, latitude and height.

[0026] In some embodiments, in terms of space modeling, the embodiments of the present application can adopt a method combining sparse storage strategy and boundary modeling technology to construct an efficient three-dimensional space representation model.

[0027] The three-dimensional space representation model can be used to represent various elements in the flight environment of the unmanned aerial vehicle, including obstacles and temporary no-fly zones. The specific spatial position information of the obstacles can be recorded in the three-dimensional space representation model. The obstacles can be natural terrain (such as mountains, trees, etc.), buildings (such as high-rise buildings, bridges, etc.) or other objects that can affect the safety of the unmanned aerial vehicle flight; by recording the spatial position information of the obstacles, it is beneficial to understand the position and shape of the obstacles in real time, so that the unmanned aerial vehicle avoids these obstacles during flight and ensures flight safety.

[0028] The boundary range information of the temporary no-fly zone can also be recorded in the three-dimensional space representation model. The temporary no-fly zone can refer to a region in which the unmanned aerial vehicle is prohibited from flying at a specific time and in a specific region. The embodiments of the present application can identify the position of the no-fly zone in time by recording the boundary range information of the temporary no-fly zone, and ensure that the unmanned aerial vehicle does not enter the no-fly zone.

[0029] In some embodiments, in the case that the position information of the obstacles changes (for example, a moving vehicle or a temporarily built building) or the range or state of the temporary no-fly zone changes (for example, the establishment or removal of the no-fly zone), the three-dimensional space representation model can be updated in real time based on these latest information, so as to provide real-time space information, so that the unmanned aerial vehicle can adjust the flight path according to the real-time latest space situation, and ensure flight safety.

[0030] In the embodiments of the present application, the flight path of each unmanned aerial vehicle can be planned by the route module, and a dedicated safety region can be delimited in space-time, so as to ensure that the unmanned aerial vehicle does not collide with other unmanned aerial vehicles or obstacles during flight.

[0031] In some embodiments, the space-time occupation state of the unmanned aerial vehicle can refer to the spatial position occupied by the unmanned aerial vehicle at different time points. By querying the space-time occupation graph in real time, the flight state of each unmanned aerial vehicle can be monitored in real time and collision detection can be performed.

[0032] In the embodiments of the present application, the path planning module can work cooperatively with the map module, the route module and other modules, perform path calculation based on the real-time space information provided by the map module and the space-time occupation state of each unmanned aerial vehicle obtained by querying the space-time occupation graph, ensure that the planned path avoids obstacles and temporary no-fly zones, and does not collide with the paths of other unmanned aerial vehicles, so as to realize the cooperative safe flight of multiple unmanned aerial vehicles.

[0033] In some embodiments, the path planning module can respond to changes in airspace information and spatio-temporal occupancy status in real time, so as to timely adjust the path planning. For example, if a new obstacle or no-fly zone is detected, or a potential conflict is found, the path planning module can re-plan the path, so as to ensure flight safety.

[0034] The embodiment of the present application constructs a coordinate conversion module for bidirectional conversion of latitude-longitude coordinates and three-dimensional grid coordinates, establishes a unified spatio-temporal coordinate conversion reference, so that each module such as a map module, a flight route module and a path planning module in the multi-unmanned aerial vehicle dynamic path planning system cooperates based on the unified spatio-temporal coordinate conversion reference, thereby constructing a multi-unmanned aerial vehicle dynamic path planning system based on a unified spatio-temporal coordination framework, and solving the systematic defects of independent and lack of organic cooperation of each functional module in the prior art. The present application can realize integrated coordination and optimization from flight plan approval to flight execution through establishment of a unified three-dimensional grid map representation and a spatio-temporal occupancy management system, and supports real-time dynamic management of a large-scale multi-unmanned aerial vehicle system.

[0035] In an alternative embodiment, the coordinate conversion module can be used to: divide a flight space region into a plurality of grids by using a preset minimum coordinate reference point and a preset grid size parameter, and allocate a grid coordinate to each grid; cache latitude-longitude-grid coordinate conversion pairs; when receiving a coordinate to be converted, perform bidirectional conversion of latitude-longitude coordinates and grid coordinates on the coordinate to be converted by using a grid-based direct linear mapping algorithm and a cache optimization technique, and by using the preset minimum coordinate reference point and the grid size parameter; and cache frequently accessed latitude-longitude-grid coordinate conversion pairs by using the cache optimization technique. The frequently accessed latitude-longitude-grid coordinate conversion pairs refer to latitude-longitude-grid coordinate conversion pairs with a number of accesses greater than a preset access number threshold.

[0036] In the embodiment of the present application, the coordinate conversion module can use a grid-based direct linear mapping algorithm and a cache optimization technique to realize efficient bidirectional conversion of latitude-longitude coordinates and three-dimensional grid coordinates by using a preset minimum coordinate reference point and a grid size parameter, provide a consistent spatial positioning basis for the entire multi-unmanned aerial vehicle dynamic path planning system, and ensure seamless flow and real-time sharing of information among functional modules.

[0037] In some embodiments, the minimum coordinate reference point and the grid size parameter in the geographic area can be preset; the geographic area is divided into a plurality of grids according to the preset minimum coordinate reference point and the grid size parameter, each grid is assigned a grid coordinate, and a latitude-longitude-grid coordinate conversion pair is cached. If the received coordinate to be converted is a latitude-longitude coordinate and the latitude-longitude coordinate is not cached, a direct linear mapping algorithm based on gridding can be used to calculate the offset of the latitude-longitude point (lat, lon) relative to the minimum coordinate reference point, and the offset is mapped to the grid coordinate; if the received coordinate to be converted is a grid coordinate and the latitude-longitude coordinate is not cached, the center point distance of the grid coordinate relative to the reference point can be calculated, the distance is converted into a latitude-longitude offset, and the offset is added to the latitude-longitude of the reference point to obtain the latitude-longitude coordinate of the target point corresponding to the grid coordinate.

[0038] In some embodiments, frequently used latitude-longitude-grid coordinate pairs can be cached to improve coordinate conversion efficiency.

[0039] In some embodiments, when the reference point or the grid size parameter changes, the related data in the cache can be cleared to ensure that the data in the cache is not outdated, In an optional embodiment, the map module can be specifically configured to determine the area surrounded by the plurality of boundary points as the boundary range information of the temporary flight restricted area, and store the spatial position information of the obstacle by using a sparse storage strategy; the map module is further configured to work cooperatively with the route module to share the unified airspace information and perform incremental updating on the fixed obstacle and the dynamic flight restricted area.

[0040] In the airspace modeling aspect, the traditional volume filling strategy causes great waste of memory resources, and the system scalability is severely restricted, which cannot support the real-time management requirements of large-scale urban airspace. Compared with the traditional scheme, in the embodiment of the present application, the map module can store only the obstacle coordinates by using a set data structure instead of full-space information, and generate the boundary points of the temporary flight restricted area instead of filling the volume, so as to greatly reduce the memory occupation under the premise of ensuring safety; the map module can work cooperatively with the route module to share the unified airspace information, support incremental updating of the fixed obstacle and the dynamic flight restricted area, and realize real-time response to changes in the airspace environment.

[0041] In some embodiments, the set data structure can include point, line, surface and other geometric elements for accurately representing the position and shape of the obstacle. By storing only the spatial position information of the obstacle, the map module can quickly query and update the obstacle information, while reducing unnecessary data storage and improving the performance and response speed of the system; by using the boundary point generation method to determine the boundary range information of the temporary flight restricted area, the amount of data can be reduced, which is conducive to improving the efficiency of conflict detection.

[0042] In the embodiment of the present application, the map module supports incremental updating of fixed obstacles and dynamic no-fly zones, and through the incremental updating mechanism, the map module can quickly respond to changes in airspace environment, so as to ensure that the route module can perform path planning and conflict detection according to the latest airspace information.

[0043] In the embodiment of the present application, the boundary no-fly zone modeling technology generates and stores only the boundary points of the no-fly zone instead of filling the entire volume, and the cylindrical no-fly zone is reduced from to boundary points, with a memory reduction ratio of 99.8%, and the polygon no-fly zone minimizes storage requirements while ensuring safety through an improved boundary tracking algorithm combined with a safety buffer expansion technology, laying a foundation for large-scale system deployment.

[0044] The embodiment of the present application can greatly improve resource utilization efficiency and system scalability while ensuring high-precision multi-vehicle coordination through the boundary modeling technology and the incremental updating mechanism, and provides a complete technical solution for intelligent management of future urban airspace.

[0045] In an optional embodiment, the route module can be specifically configured to: based on flight time interval management and four-dimensional safety region envelope design, demarcate an exclusive safety region in space-time for each unmanned aerial vehicle; and the system further comprises a conflict detection module integrated with the map module and the route module, configured to adopt a time region overlap detection algorithm to perform fixed obstacle, dynamic no-fly zone and multi-unmanned aerial vehicle conflict identification through real-time query and update of a space-time occupancy graph.

[0046] The existing system mainly relies on centralized or distributed conflict detection algorithms for multi-vehicle coordination, and achieves collision avoidance through simple division of time or space, lacking accurate space-time occupancy modeling and global optimization mechanism. In view of the above problems, in the embodiment of the present application, the space-time occupancy management system is the core mechanism for coordinating multi-vehicle flight, and the route module can demarcate an exclusive safety region in space-time for each unmanned aerial vehicle through time interval management and four-dimensional safety envelope design. The space-time occupancy management system can be deeply integrated with the conflict detection module to achieve millisecond-precision multi-vehicle conflict identification through real-time query and update of a space-time occupancy graph. The conflict resolution mechanism can adopt a hierarchical strategy, and a multi-level response scheme from time adjustment to path re-planning, to ensure the stability and efficiency of the multi-unmanned aerial vehicle dynamic path planning system in complex conflict scenarios.

[0047] In some embodiments, the time interval management can refer to planning the flight task for each UAV, and explicitly defining the flight area and task content of each UAV in different time periods. The four-dimensional safety envelope design can refer to designing a dedicated safety area for each UAV by combining the time dimension with the three-dimensional space (length, width, and height). The four-dimensional safety envelope design can be used to define the dedicated safety area of the UAV. Even if two UAVs overlap in time, as long as their safety envelope areas do not overlap in space, flight conflicts can be avoided.

[0048] In some embodiments, the space-time occupation management system can construct a space-time occupation map, which can record the spatial position occupied by each UAV at different time points, with time as the axis and space as the dimension. During the flight of the UAV, the position information of the UAV will change, and the space-time occupation map will be updated in real time accordingly.

[0049] In some embodiments, the conflict detection module can be integrated with the map module and the route module. The conflict detection module can query the space-time occupation map in real time, and use a time region overlap detection algorithm to check whether two or more UAVs occupy the same or similar spatial position at the same time point or within a short period of time, thereby achieving fixed obstacle, dynamic no-fly zone, and multi-UAV conflict identification.

[0050] The embodiments of the present application design a four-dimensional space-time management system through a space-time occupation accurate modeling mechanism, set a spatial safety envelope with a custom grid size for each UAV, and combine a custom time buffer to form a millisecond-level precision multi-UAV coordination capability. The space-time occupation management mechanism can achieve accurate conflict identification through a time interval overlap detection algorithm, support intelligent merging optimization of time intervals, and work cooperatively with boundary modeling technology to construct an efficient airspace resource management system.

[0051] In an alternative embodiment, the system can further include a risk assessment module. The risk assessment module can be used to share information with the map module and the route module, to comprehensively evaluate the spatial conflict and space-time occupation conflict of the flight path of all UAVs, to perform risk analysis and pre-evaluation on the flight plan during the flight plan approval stage, and to monitor the flight state of multiple UAVs in real time during the flight execution stage.

[0052] The existing system lacks a unified optimization strategy from flight plan approval to flight execution, different evaluation criteria and optimization objectives are used in each stage, which leads to low overall efficiency of the system, cannot fully utilize airspace resources, and is difficult to meet the safety and efficiency requirements of future large-scale unmanned aircraft operations. In order to solve the above problems, in the embodiment of the present application, the risk assessment module realizes the unified safety management of fixed routes and autonomously planned paths, and through information sharing with the map module and the route module, it can comprehensively evaluate the spatial conflict and time-space occupation conflict of all flight paths. The risk assessment module supports pre-evaluation in the flight plan approval stage and real-time monitoring in the flight execution stage, and establishes a safety guarantee system covering the entire flight life cycle.

[0053] In some embodiments, the risk assessment module can conduct a comprehensive risk analysis and pre-evaluation of the flight plan in the flight plan approval stage, quantitatively evaluate the potential risks in the flight plan through the risk assessment model, and generate a risk assessment report, thereby providing a scientific basis for the approval of the flight plan and ensuring the safety and feasibility of the flight plan. The risk assessment module can also monitor the flight status of the unmanned aerial vehicle in real time in the flight execution stage, receive real-time flight data such as position, speed, and height of the unmanned aerial vehicle, combine pre-set safety rules and risk thresholds, and dynamically evaluate the risks in the flight process, so as to timely issue an alarm when a potential risk is found and provide corresponding risk response suggestions to ensure the safety of the unmanned aerial vehicle during flight.

[0054] In the embodiment of the present application, the risk assessment module can effectively identify and manage various risks in the flight process, ensuring that the unmanned aerial vehicle is in a safe and controllable state from takeoff to landing, and providing comprehensive protection for the safe flight of the unmanned aerial vehicle.

[0055] In some embodiments, a dynamic time adjustment algorithm can be used to automatically find the optimal takeoff time when a time-space conflict is detected, supporting a bidirectional time search strategy and a time adjustment range of up to 3600 seconds. The dynamic time adjustment algorithm can be deeply integrated with the time-space occupation management mechanism, reducing the need for path re-planning through intelligent time window optimization, and improving the overall coordination efficiency of the system.

[0056] In an alternative embodiment, the path planning module can be used to: use an improved three-dimensional jump point search algorithm to search for an optimal flight path from the flight starting point to the flight ending point in a three-dimensional airspace; use a vertical obstacle avoidance strategy to consider the height information of obstacles during path search, so that the planned path avoids obstacles in the vertical direction; and use a multi-level cache optimization strategy to cache frequently used flight paths; the frequently used flight paths refer to flight paths with a number of accesses greater than a pre-set usage threshold.

[0057] In the embodiments of the present application, the path planning module can adopt an improved three-dimensional jump point search algorithm, combine a vertical obstacle avoidance strategy and a multi-level cache optimization, and realize efficient path search in a unified three-dimensional grid space. The path planning module can work cooperatively with a map module, a flight route module, a conflict detection module, a risk assessment module and other modules in a multi-unmanned aerial vehicle dynamic path planning system, perform path calculation based on real-time airspace information and space-time occupation state, integrate take-off, cruising and landing into overall optimization through a three-stage unified planning framework, and realize balance between safety and efficiency.

[0058] In some embodiments, the position information of known fixed obstacles and the boundary point information of temporary flight restricted areas can be converted into grid coordinates, and the positions of these obstacles and flight restricted areas are marked in a three-dimensional grid space to form a three-dimensional grid map. Then, an improved three-dimensional jump point search algorithm can be adopted for heuristic search, obstacles in horizontal and vertical directions are considered in the search process to ensure that the path can avoid obstacles in all directions, and the priority of each node is evaluated through a heuristic function to preferentially expand nodes closer to the end point. After the search is completed, the optimal flight path from the flight start point to the flight end point can be obtained.

[0059] In some embodiments, when searching for a path, not only obstacles in the horizontal direction are considered, but also obstacles in the vertical direction. According to the vertical obstacle avoidance strategy, the height of the flight path can be adjusted, so that when an obstacle in the vertical direction is detected, the path planning module can automatically adjust the height of the path to avoid the obstacle.

[0060] In some embodiments, frequently used flight paths and intermediate results can be cached; paths and intermediate results that are likely to be frequently used can be preloaded to reduce the number of real-time calculations; when the position or height of an obstacle changes, the related data in the cache can be cleared and path search is performed again, so that the data in the cache will not become outdated.

[0061] The existing overall multi-unmanned aerial vehicle management system architecture often adopts a static allocation strategy when processing large-scale multi-airport scenarios, and it is difficult to adapt to dynamically changing airspace environments and real-time task demand adjustments. The embodiments of the present application improve the traditional JPS (Jump Point Search) algorithm through a three-dimensional jump point search optimization algorithm to adapt to three-dimensional space and dynamic environment, add a vertical obstacle avoidance strategy and a multi-level LRU (Least Recently Used) cache mechanism, and when combined with a unified coordinate conversion benchmark and a space-time occupation management system, realize efficient path search in a complex three-dimensional airspace, and significantly improve the real-time response capability of the system.

[0062] The embodiment of the present application integrates take-off, cruising and landing into a unified optimization target through a three-stage unified planning framework, breaks the limitation of mutual independence of each stage in the traditional system, realizes overall optimization of the whole flight cycle through the cooperation of boundary modeling, space-time occupation management and three-dimensional path search, and ensures the balance between safety and efficiency.

[0063] In an optional embodiment, the system can further comprise a dynamic rerouting module; the dynamic rerouting module can be used for: cooperating with the path planning module and the conflict detection module to perform path re-planning when receiving the alarm information; and re-planning the path segment belonging to the target region when detecting that a temporary no-fly zone appears in the front target region.

[0064] In the embodiment of the present application, the dynamic rerouting module as a key component of the system responding to sudden situations can realize rapid response of alarm information and path re-planning through close cooperation with the path planning module and the conflict detection module. The dynamic rerouting module can support local re-planning and incremental updating, maximize the use of existing path information, and reduce the calculation amount and time consumption of re-computation.

[0065] In some embodiments, the dynamic rerouting module supports local re-planning and incremental updating, can make local adjustment to the existing flight path according to the specific location and influence range of the sudden situation, maximizes the use of existing path information, avoids re-computation of the whole path, and significantly reduces the calculation amount and time consumption. Exemplarily, when detecting that a temporary no-fly zone appears in a certain region in front, the dynamic rerouting module only re-plans the path segment near the region, while keeping other parts of the path unchanged.

[0066] In terms of dynamic environment adaptability, the prior art cannot realize incremental updating and local re-planning, and when the airspace environment changes, the whole system often needs to be restarted or all paths need to be re-computed, which has long response time and large resource consumption. In view of the above problems, the embodiment of the present application supports efficient processing of real-time environmental changes through an incremental dynamic updating mechanism, realizes the local optimization capability that a no-fly zone change only affects the related regional path and an environmental change only re-plans the affected path segment when combined with a unified space-time coordination framework, thereby greatly improving the dynamic adaptability and resource utilization efficiency of the system, and forming a complete intelligent management system.

[0067] Figure 2 FIG. 2 is a structural schematic diagram of a multi-unmanned aerial vehicle dynamic path planning system according to an embodiment of the present application; and Figure 2In one specific embodiment, a multi-UAV dynamic path planning system based on a unified space-time coordination framework is constructed, and coordination and optimization are achieved from flight plan approval to flight execution through an integrated architecture design. The multi-UAV dynamic path planning system takes a unified three-dimensional grid coordinate system as the basis, organically integrates map representation, space-time occupation management, path planning, conflict detection, risk assessment, and dynamic rerouting, and constructs a complete multi-UAV cooperative management solution.

[0068] The core innovation of the system is to establish a unified space-time coordinate conversion benchmark, and all modules work cooperatively based on the bidirectional conversion mechanism of the WGS84 coordinate system and the three-dimensional grid coordinate, avoiding the data conversion bottleneck caused by the use of different coordinate benchmarks by various modules in traditional systems. The coordinate conversion module adopts a direct linear mapping algorithm based on gridding and a cache optimization technique, and realizes efficient bidirectional conversion of latitude and longitude and grid coordinates through preset minimum coordinate benchmark points and grid size parameters, providing a consistent spatial positioning basis for the entire system and ensuring seamless transfer and real-time sharing of information between various functional modules.

[0069] In terms of airspace modeling, the system adopts an innovative method combining sparse storage strategy and boundary modeling technology, and constructs an efficient three-dimensional airspace representation model. The map module only stores obstacle coordinates rather than full-space information through a set data structure, and generates boundary points for temporary no-fly zones rather than filling volumes, greatly reducing memory occupation while ensuring safety. The module works cooperatively with the route module, sharing unified airspace information, supporting incremental update of fixed obstacles and dynamic no-fly zones, and realizing real-time response to changes in airspace environment.

[0070] The space-time occupation management system is the core mechanism for coordinating multi-UAV flight, and the route module defines an exclusive safety area for each UAV in space-time through time interval management and four-dimensional safety envelope design. The system is deeply integrated with the conflict detection module, and realizes millisecond-level precision multi-UAV conflict identification through real-time query and update of space-time occupation graph. The conflict resolution mechanism adopts a hierarchical strategy, and the multi-level response scheme from time adjustment to path re-planning ensures the stability and efficiency of the system in complex conflict scenarios.

[0071] The risk assessment module realizes unified safety management of fixed routes and autonomously planned paths, and through information sharing with the map module and route module, it conducts comprehensive assessment of spatial conflict and space-time occupation conflict for all flight paths. The module supports pre-evaluation in the flight plan approval stage and real-time monitoring in the flight execution stage, and establishes a safety guarantee system covering the entire flight life cycle.

[0072] The path planning module adopts an improved three-dimensional jump point search algorithm, combines a vertical obstacle avoidance strategy and a multi-level cache optimization, and realizes efficient path search in a unified three-dimensional grid space. The module cooperates with the aforementioned modules, performs path calculation based on real-time airspace information and space-time occupation state, integrates take-off, cruising and landing into overall optimization through a three-stage unified planning framework, and realizes balance between safety and efficiency.

[0073] The dynamic rerouting module, as a key component of the system for responding to emergencies, realizes rapid response of alarm information and path re-planning through close cooperation with the path planning module and the conflict detection module. The module supports local re-planning and incremental updating, maximizes the use of existing path information, and reduces the calculation amount and time consumption of re-calculation.

[0074] The entire multi-unmanned aerial vehicle dynamic path planning system realizes coordinated operation of the modules through unified configuration management and real-time monitoring mechanism, establishes a closed-loop management system from data input, processing calculation to result output, and realizes intelligent management and dynamic optimization of a large-scale multi-unmanned aerial vehicle system.

[0075] In summary, the embodiment of the application has the following technical effects: Compared with the existing separate management system, the embodiment of the application realizes significant overall performance improvement through systematic innovation of the unified space-time coordination framework. In the system architecture level, the unified coordinate reference and inter-module cooperation mechanism eliminate the data conversion bottleneck in the traditional system, and the information sharing and coordinated optimization of each functional module make the overall performance far exceed the simple additive effect of independent operation of each module. The organic combination of the boundary modeling technology and the space-time occupation management reduces the memory occupation in the typical urban environment by more than 95%, reduces the storage space of the cylindrical no-fly zone by 99.8%, and reduces the storage demand of the complex polygon no-fly zone by an average of 98%, laying a solid foundation for large-scale system deployment and expansion.

[0076] The improvement of path planning efficiency not only reflects in algorithm optimization, but also reflects in the efficiency multiplication effect brought by systematic cooperation. The optimized three-dimensional JPS algorithm is deeply integrated with the unified coordinate system and the space-time occupation management, and can realize more than 5 times of comprehensive acceleration effect compared with the traditional separate system, and the average search time is reduced from 1.5 seconds to 0.05 seconds. The 78%-92% hit rate of the multi-level cache mechanism further magnifies the overall performance of the system. More importantly, the system supports the scalability of 1000 unmanned aerial vehicles and the path planning success rate of 99.7%, which proves that the cooperation effect under the unified framework far exceeds the linear expansion capability of the traditional system.

[0077] The fundamental improvement of multi-machine coordination accuracy is derived from the accurate modeling of space-time occupation and the systematic advantage of collaborative work of each module. The realization of millisecond-level time accuracy and grid-level space accuracy, the reduction of conflict false alarm rate to below 0.1%, and the performance of conflict missed alarm rate close to 0%, reflect the essential advantage of the unified space-time coordination framework compared with the traditional simple space-time division method. The efficiency of controlling the average coordination time within 2.3 seconds reflects the overall optimization effect brought by the collaborative work of the system modules.

[0078] The significant enhancement of dynamic adaptability is an important embodiment of the systematic innovation of the present application. The combination of incremental update mechanism and unified framework realizes the rapid response ability of less than 1 second for no-fly zone change response time and less than 1 second for alarm rerouting processing control. Compared with the cumbersome way of restarting or full recalculation required by the traditional system, the present application realizes real-time dynamic management, and the overall improvement of fault recovery ability and system stability proves the significant advantage of integrated architecture design.

[0079] In order to better understand the embodiments of the present application, the embodiments of the present application are described below based on the large-scale unmanned aerial vehicle distribution management system in the main urban area as an example.

[0080] Taking the large-scale unmanned aerial vehicle distribution management in the main urban area as an example, the actual application effect of the technical scheme of the present application is verified. This scene can cover a city airspace of 15x12x1 kilometers, and needs to manage 200 unmanned aerial vehicles to perform express delivery tasks at the same time, and the airspace can contain about 500 fixed obstacles and 20 dynamic no-fly zones.

[0081] The system can adopt a three-dimensional grid division of 3000x2400x200, with a grid accuracy of 5 meters, a minimum cruising altitude of 100 meters, and a maximum cruising altitude of 500 meters. In the initialization stage, 500 building models are imported from the GIS database. The traditional method needs to store about 250 million grid points, occupying 250MB of memory, and the present application method actually stores about 150,000 obstacle boundary points, occupying only 12MB of memory, saving 95.2% of memory. Add a cylindrical no-fly zone with a radius of 5 kilometers around the airport. The traditional method needs about 314 million grid points, and the present application method only needs about 3140 boundary points, saving 99.9% of memory.

[0082] In the path planning stage, the delivery paths are planned for 200 unmanned aerial vehicles at the same time, with a total planning time of 5 seconds, an average of 0.025 seconds per unmanned aerial vehicle, 198 unmanned aerial vehicles successfully planned, and a success rate of 99%. The two failed unmanned aerial vehicles are because the target address is located in the no-fly zone. The average length of the planned path is 3.2 kilometers, and the average flight time is 18 minutes, which is an average of 15% more than the straight-line distance. All paths pass the safety check.

[0083] The dynamic running stage simulates a temporary flight restricted area event, and a new temporary flight restricted area of a cultural activity is added at the 30th minute, with an area of 2 square kilometers and an influence on 12 unmanned aerial vehicles. The system response process takes 0.08 seconds to add the flight restricted area, 0.12 seconds for conflict detection, 0.05 seconds per unmanned aerial vehicle for path re-planning on average, 0.05 seconds per unmanned aerial vehicle for new path issuing, and the overall response is efficient. In the emergency rerouting event, a certain unmanned aerial vehicle detects a low-altitude aircraft in front that needs to be avoided, and the system achieves a fast dynamic response with a total response time of 0.5 seconds from alarm reception to new path issuing.

[0084] The system performs stably in 24-hour continuous operation, with memory usage increasing from the initial 280MB to the peak 350MB, with a growth rate of 28.9% without memory leakage. The average CPU usage is 25%, and the peak value of 85% appears during batch planning. The network communication processes 2850 Kafka messages, with an average processing time of 8 milliseconds and a success rate of 99.96%. The database performs 456 route queries and 234 path storage, with an average response time of 18 milliseconds and a connection pool utilization rate of 65%.

[0085] Figure 3 is a flowchart of the multi-unmanned aerial vehicle dynamic path planning method provided by the embodiment of the present application. Referring to Figure 3 The embodiment of the present application provides a multi-unmanned aerial vehicle dynamic path planning method, which can specifically include the following steps: Step 301, bidirectional conversion is performed on latitude and longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion benchmark; Step 302, a three-dimensional airspace representation model is constructed; the three-dimensional airspace representation model records the spatial position information of obstacles and the boundary range information of temporary flight restricted areas to provide real-time airspace information; Step 303, a dedicated safety area is demarcated for each unmanned aerial vehicle in space-time, and the demarcated safety area information is updated to a space-time occupancy map; the space-time occupancy map is used to record the space-time occupancy state of each unmanned aerial vehicle; Step 304, multi-unmanned aerial vehicle path planning is performed based on the real-time airspace information and the space-time occupancy state; Among them, the real-time airspace information, the space-time occupancy state of each unmanned aerial vehicle and the multi-unmanned aerial vehicle path planning are realized based on the unified space-time coordinate conversion benchmark.

[0086] It should be noted that the execution subject of the multi-unmanned aerial vehicle dynamic path planning method provided by the embodiments of the present application can be an electronic device, a component in the electronic device, an integrated circuit or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a notebook computer, a palm computer, a wearable device, an Ultra-mobile Personal Computer (UMPC), a netbook or a Personal Digital Assistant (PDA), etc., and the non-mobile electronic device can be a server, a Network Attached Storage (NAS), a Personal Computer (PC), a Television (TV), a teller machine or a self-service machine, etc., and the embodiments of the present application do not make a specific limitation thereto.

[0087] The embodiments of the present application establish a unified space-time coordinate conversion benchmark by bidirectional conversion between latitude-longitude coordinates and three-dimensional grid coordinates, so that the real-time airspace information, the space-time occupation state of each unmanned aerial vehicle and the multi-unmanned aerial vehicle path planning in the multi-unmanned aerial vehicle dynamic path planning process are realized based on the unified space-time coordinate conversion benchmark, thereby constructing a multi-unmanned aerial vehicle dynamic path planning method based on a unified space-time coordination framework, which can solve the systematic defects of independent steps and lack of organic cooperation in the prior art. The embodiments of the present application can realize integrated coordination and optimization from flight plan approval to flight execution by establishing a unified three-dimensional grid map representation and a space-time occupation management system, which is conducive to supporting real-time dynamic management of a large-scale multi-unmanned aerial vehicle system.

[0088] Figure 4 An example of an entity structure diagram of an electronic device is shown in FIG. 1. Figure 4As shown, the electronic device can include a processor 410, a communications interface 420, a memory 430, and a communications bus 440, wherein the processor 410, the communications interface 420, and the memory 430 complete mutual communication through the communications bus 440. The processor 410 can invoke a logical instruction in the memory 430 to execute a multi-unmanned aerial vehicle dynamic path planning method, which includes: performing bidirectional conversion between latitude and longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion benchmark; constructing a three-dimensional airspace representation model; the three-dimensional airspace representation model records spatial position information of obstacles and boundary range information of temporary flight restricted areas to provide real-time airspace information; demarcating an exclusive safety area for each unmanned aerial vehicle in space-time, and updating the demarcated safety area information to a space-time occupancy graph; the space-time occupancy graph is used to record space-time occupancy states of each unmanned aerial vehicle; performing multi-unmanned aerial vehicle path planning based on the real-time airspace information and the space-time occupancy states; wherein the real-time airspace information, the space-time occupancy states of each unmanned aerial vehicle, and the multi-unmanned aerial vehicle path planning are realized based on the unified space-time coordinate conversion benchmark.

[0089] In addition, the logical instruction in the memory 430 described above can be realized in the form of a software functional unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0090] The system embodiments described above are only schematic, wherein the units shown as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the present embodiment. Those skilled in the art can understand and implement without creative labor.

[0091] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0092] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-UAV dynamic path planning system, characterized in that: The system comprises: Coordinate conversion module, used to perform bidirectional conversion between latitude and longitude coordinates and three-dimensional grid coordinates to establish a unified spatiotemporal coordinate conversion benchmark; A map module is used to construct a three-dimensional airspace representation model; the three-dimensional airspace representation model records the spatial location information of obstacles and the boundary range information of temporary no-fly zones to provide real-time airspace information; The route module is used to define a dedicated safety zone in space and time for each drone and update the designated safety zone information to the space-time occupancy map; the space-time occupancy map is used to record the space-time occupancy status of each drone; A path planning module, configured to perform multi-UAV path planning based on the real-time airspace information and the spatiotemporal occupancy status; The map module, the route module and the path planning module work together based on the unified space-time coordinate conversion benchmark.

2. The multi-UAV dynamic path planning system according to claim 1, characterized in that: The coordinate conversion module is specifically used for: Using the preset minimum coordinate reference point and preset grid size parameters, the flight space area is divided into multiple grids, and grid coordinates are assigned to each grid, and the latitude and longitude-grid coordinate conversion pairs are cached; Upon receiving the coordinates to be converted, a grid-based direct linear mapping algorithm and cache optimization technology are used to perform a bidirectional conversion between latitude and longitude coordinates and grid coordinates on the coordinates to be converted using the preset minimum coordinate reference point and the grid size parameters; Cache optimization technology is used to cache frequently accessed longitude and latitude-grid coordinate conversion pairs; the frequently accessed longitude and latitude-grid coordinate conversion pairs refer to longitude and latitude-grid coordinate conversion pairs whose access times are greater than a preset access times threshold.

3. The multi-UAV dynamic path planning system according to claim 1, characterized in that: The map module is specifically configured to: determine an area enclosed by a plurality of boundary points as boundary range information of a temporary no-fly zone, and store spatial location information of obstacles using a sparse storage strategy; The map module is also used to work in conjunction with the route module to share unified airspace information and perform incremental updates on fixed obstacles and dynamic no-fly zones.

4. The multi-UAV dynamic path planning system according to claim 1, characterized in that: The route module is specifically used to: define a dedicated safety zone in time and space for each drone based on flight time interval management and four-dimensional safety zone envelope design; The system also includes a conflict detection module, which is integrated with the map module and the route module and is used to use a time area overlap detection algorithm to identify fixed obstacles, dynamic no-fly zones and multi-UAV conflicts through real-time query and update of the spatiotemporal occupancy map.

5. The multi-UAV dynamic path planning system according to claim 1, characterized in that: The system further includes a risk assessment module; the risk assessment module is configured to: Sharing information with the map module and the route module to conduct a comprehensive assessment of the spatial conflict and space-time occupancy conflict of the flight paths of all drones; Conduct risk analysis and pre-assessment of flight plans during the flight plan approval phase, and conduct real-time monitoring of the flight status of multiple drones during the flight execution phase.

6. The multi-UAV dynamic path planning system according to claim 1, characterized in that: The path planning module is specifically used to: An improved 3D jumping point search algorithm is used to search for the optimal flight path from the flight start point to the flight end point in the 3D airspace. Adopting a vertical obstacle avoidance strategy, the height information of obstacles is considered during the path search process, so that the planned path avoids obstacles in the vertical direction; A multi-level cache optimization strategy is adopted to cache frequently used flight paths; the frequently used flight paths are flight paths whose access times are greater than a preset usage count threshold.

7. The multi-UAV dynamic path planning system according to claim 4, characterized in that: The system further includes a dynamic rerouting module; the dynamic rerouting module is configured to: Working in conjunction with the path planning module and the conflict detection module, replanning the path upon receiving an alarm message; When a temporary no-fly zone is detected in a target area ahead, the path segment belonging to the target area is replanned.

8. A multi-UAV dynamic path planning method, characterized in that: The method comprises: Perform bidirectional conversion between latitude and longitude coordinates and three-dimensional grid coordinates to establish a unified space-time coordinate conversion benchmark; Constructing a three-dimensional airspace representation model; the three-dimensional airspace representation model records the spatial location information of obstacles and the boundary range information of temporary no-fly zones to provide real-time airspace information; Demarcate a dedicated safety zone for each drone in space and time, and update the designated safety zone information to a space-time occupancy map; the space-time occupancy map is used to record the space-time occupancy status of each drone; Performing multi-UAV path planning based on the real-time airspace information and the spatiotemporal occupancy status; The real-time airspace information, the spatiotemporal occupancy status of each UAV and the multi-UAV path planning are implemented based on the unified spatiotemporal coordinate conversion benchmark.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the multi-UAV dynamic path planning method according to claim 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the multi-UAV dynamic path planning method as claimed in claim 8 is implemented.

Images