Cluster unmanned aerial vehicle path planning method
By employing a swarm drone path planning method, and using a hierarchical framework and clearly defined group planning, the performance degradation and responsiveness of distributed collaborative algorithms in dynamic environments are solved, achieving efficient and reliable communication and task execution.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA ORDNANCE EQUIP GRP AUTOMATION RES INST CO LTD
- Filing Date
- 2026-01-06
- Publication Date
- 2026-04-21
AI Technical Summary
Existing distributed cooperative algorithms suffer from performance degradation under packet loss, latency, and bandwidth constraints, lack rapid response capabilities in dynamic environments, have limited research on differentiated heterogeneous cooperative systems for UAVs with different payloads, and lack sufficient verification of safety and reliability.
A hierarchical framework of centralized allocation and distributed execution is adopted. Through the division of labor among the high-altitude decision-making unit, the low-altitude execution unit, and the relay communication hub, reconnaissance, search, and surround-view path planning are carried out, communication scheduling and path decision-making are optimized, and the needs of large-scale swarm UAV missions are adapted.
It improves real-time communication, reduces the risk of single point of failure, saves communication bandwidth, meets the identification and obstacle avoidance requirements of the pod and vision module, and improves the efficiency of large-scale cluster control.
Smart Images
Figure CN121900446A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of path planning technology, and in particular to a path planning method for swarm drones based on a distributed cooperative algorithm. Background Technology
[0002] Path planning for swarm drones is a core challenge in the field of autonomous collaboration of unmanned systems. As the swarm size increases and the application scenarios become more complex, distributed collaborative planning is widely chosen due to its advantages such as decentralization, high robustness, and strong scalability. Based on local perception and neighbor communication, autonomous decision-making is achieved through collaboration to reach global goals. Representative approaches include consensus algorithms, distributed optimization, and distributed model predictive control.
[0003] Current distributed collaborative algorithms have limitations such as sensitivity to communication constraints that lead to performance degradation under packet loss, latency, and bandwidth limitations; poor adaptability to dynamic environments due to insufficient rapid response capabilities in the event of sudden obstacles or task changes; limited research on differentiated and heterogeneous collaborative algorithms for UAVs with different payloads; and limited verification of safety and reliability. Summary of the Invention
[0004] In view of the above problems, the present invention provides a swarm drone path planning method to overcome or at least partially solve the above problems.
[0005] This invention provides the following solution: A method for path planning of swarm drones includes: The system receives input parameters from the user, including reconnaissance area data, surround view area data, and configuration data for each UAV; the configuration data includes the group to which the UAV belongs, which includes a high-altitude group and a low-altitude group. One of the UAVs from the high-altitude group is selected as the decision-making machine; Based on the reconnaissance area data, reconnaissance paths are planned for all UAVs in the high-altitude group and the low-altitude group respectively, and the results of the reconnaissance path planning for each UAV are sent to the corresponding UAV so that the UAV that receives the results of the reconnaissance path planning can perform the reconnaissance mission. After performing a reconnaissance mission, the UAV uploads the search area data and sends it to the decision-making machine. The decision-making machine is used to obtain the coordinates of all UAVs in the high-altitude group currently performing a reconnaissance mission, perform search path planning for the multiple UAVs closest to the search area, and send the search path planning results for each UAV to the corresponding UAV so that the UAV that receives the search path planning results can perform a search mission on the search target. All available UAVs in the low-altitude group are identified, and surround-view path planning is performed for all available UAVs based on the surround-view area data. The surround-view path results for each UAV are then sent to the corresponding UAV so that the UAV receiving the surround-view path results can perform the surround-view task.
[0006] Preferably, the reconnaissance area data includes an irregular quadrilateral defined on a map by the latitude and longitude of four points, as well as a reconnaissance boundary defined by humans.
[0007] Preferably, the rules for reconnaissance path planning include selecting the longest side of the minimum bounding box of the irregular quadrilateral for scanning, obtaining the number of scan lines based on the coverage width, and the set of intersections between the scan lines and the boundary of the reconnaissance area is a path point list, which is output to the output file in sequence.
[0008] Preferably, the surround view area data includes an irregular quadrilateral formed by the latitude and longitude of four points identical to the reconnaissance area data, as well as an artificially defined surround view boundary.
[0009] Preferably, the surround view area data also includes the aircraft scanning radius, which is the pod coverage width while ensuring recognition accuracy.
[0010] Preferably, the search area includes buildings, and the search area data includes a regular polygon formed by the latitude and longitude of multiple points of the buildings.
[0011] Preferably, the group further includes a relay group. When the search distance is too long, the number of aircraft in the cluster is large, or the search area is far from the operator, relay path planning is performed on each UAV in the relay group, and the result of the relay path planning for each UAV is sent to the corresponding UAV so that the UAV that receives the result of the relay path planning can perform the relay task and act as a message forwarding hub.
[0012] Preferably, the planned target points are displayed sequentially according to the tasks performed by different aircraft.
[0013] According to specific embodiments provided by the present invention, the present invention discloses the following technical effects: The path planning method for swarm drones provided in this application embodiment, compared with the previous centralized planning, eliminates the risk of single point of failure by relying on the central controller (ground station) for global optimization. Through algorithm optimization and hardware adaptation, it saves communication bandwidth, improves the real-time performance of communication, and is more advantageous for large-scale swarm control. It also meets the task requirements of the onboard pods and vision modules for subsequent identification and obstacle avoidance tasks.
[0014] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.
[0016] Figure 1 This is a flowchart of reconnaissance area path planning provided in an embodiment of the present invention; Figure 2 This is a flowchart of the search area path planning provided in an embodiment of the present invention; Figure 3 This is a flowchart of the surround view area path planning provided in an embodiment of the present invention. Detailed Implementation
[0017] 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 a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention are within the scope of protection of the present invention.
[0018] See Figure 1 This invention provides a method for swarming unmanned aerial vehicle (UAV) path planning, as exemplified by this invention. Figure 1 As shown, the method may include: The system receives input parameters from the user, including reconnaissance area data, surround view area data, and configuration data for each UAV; the configuration data includes the group to which the UAV belongs, which includes a high-altitude group and a low-altitude group. One of the UAVs from the high-altitude group is selected as the decision-making machine; Based on the reconnaissance area data, reconnaissance paths are planned for all UAVs in the high-altitude group and the low-altitude group respectively, and the results of the reconnaissance path planning for each UAV are sent to the corresponding UAV so that the UAV that receives the results of the reconnaissance path planning can perform the reconnaissance mission. After performing a reconnaissance mission, the UAV uploads the search area data and sends it to the decision-making machine. The decision-making machine is used to obtain the coordinates of all UAVs in the high-altitude group currently performing a reconnaissance mission, perform search path planning for the multiple UAVs closest to the search area, and send the search path planning results for each UAV to the corresponding UAV so that the UAV that receives the search path planning results can perform a search mission on the search target. All available UAVs in the low-altitude group are identified, and surround-view path planning is performed for all available UAVs based on the surround-view area data. The surround-view path results for each UAV are then sent to the corresponding UAV so that the UAV receiving the surround-view path results can perform the surround-view task.
[0019] In a specific implementation, the embodiments of this application may provide the reconnaissance area data, which includes an irregular quadrilateral defined on a map as a combination of the latitude and longitude of four points, as well as a reconnaissance boundary defined by humans.
[0020] Furthermore, the rules for reconnaissance path planning include selecting the longest side of the minimum bounding box of the irregular quadrilateral for scanning, calculating the number of scan lines based on the coverage width, and the set of intersections between the scan lines and the boundary of the reconnaissance area is a path point list, which is output sequentially to the output file.
[0021] The surround view area data includes an irregular quadrilateral formed by the latitude and longitude of four points identical to the reconnaissance area data, as well as artificially defined surround view boundaries.
[0022] The surround view area data also includes the aircraft scanning radius, which is the pod coverage width to ensure recognition accuracy.
[0023] The search area includes buildings, and the search area data includes regular polygons formed by the latitude and longitude of multiple points of the buildings.
[0024] The group also includes a relay group. When the search distance is too long, the number of aircraft in the cluster is large, or the search area is far from the operator, relay path planning is performed on each UAV in the relay group, and the result of the relay path planning for each UAV is sent to the corresponding UAV so that the UAV that receives the result of the relay path planning can perform the relay task and act as a message forwarding hub.
[0025] The planned target points will be displayed sequentially along the routes according to the missions performed by different aircraft.
[0026] The UAV path planning optimization method provided in this application adopts a hierarchical framework of centralized allocation and distributed execution, balancing optimality and real-time performance, and jointly optimizing communication scheduling and path decision-making to improve resource efficiency. This includes dividing planning into daytime and nighttime based on different payload states, and performing reconnaissance planning, search planning, and surround-view planning according to different tasks. The high-altitude group is divided into three roles: leader, assistant, and relay. Each node performs planning, execution, and communication functions respectively. For the leader responsible for planning, unnecessary calculations are eliminated, and all threads are used to collect perception information and send decision commands. All three planning methods it is responsible for consider maximizing target coverage. Therefore, with a self-organizing network communication bandwidth of 20MHz, this planning method will not affect the real-time performance of identification and strike in important tasks due to the delay of real-time dynamic calculation. Furthermore, combined with the identification capabilities of general identification payloads, it can meet their viewing angle and position requirements, maximizing the authenticity and integrity of identification and the accuracy of strikes.
[0027] The following section uses a four-corner building as an example to describe in detail the method provided in the embodiments of this application.
[0028] This invention provides a day and night path planning method for swarm drones, adaptable to handling the mission paths of multiple drones. The method includes: (1) First, the reconnaissance area data, search area data, surround view area data, and configuration data of the aircraft to be used are input into the planning module as input parameters; the configuration information of the aircraft by the operator based on the actual situation is processed and transmitted to the planning module for planning. Specifically, For reconnaissance area data, it is defined on the map as an irregular quadrilateral formed by the latitude and longitude of four points. The reconnaissance boundary needs to be manually defined. For search area, the daytime task is a regular quadrilateral formed by the latitude and longitude of four points that is automatically uploaded after reconnaissance. It is generally a building. For the surrounding area, which is also the information of the reconnaissance area, the aircraft scanning radius is the pod coverage width under the condition of ensuring recognition accuracy. For aircraft configuration data, it is necessary to distinguish between high-altitude groups and low-altitude groups. If there is a high-altitude group, one lead aircraft in the high-altitude group should be configured as the decision-making aircraft, and the wingman should be configured as the execution aircraft. If the search distance is too long, or there are too many aircraft in the cluster, or the search area is far from the operator, then a relay group drone should be configured as a message forwarding hub.
[0029] (2) Based on the received input data, path planning is performed for three types of areas: reconnaissance area, search area, and surrounding area. The reconnaissance and surrounding areas are planned using a planning module deployed on the ground station based on pre-configured task data. For the search area, it is determined whether planning is required; if no planning is needed beforehand, it enters a waiting state. When valid input data is received, the lead aircraft performs the planning. Valid input data refers to the coordinates of the building to be searched transmitted back by the reconnaissance UAV. Specifically: For the reconnaissance area: All aircraft, excluding relay aircraft (if included), are used for reconnaissance, divided into high-altitude and low-altitude groups. After arriving at the pre-launch departure point, the aircraft are divided into two groups for patrol missions. The longest side of the minimum bounding box of the quadrilateral is selected for scanning to reduce the number of turns. The number of scan lines is calculated based on the coverage width. The set of intersections between the scan lines and the boundary of the reconnaissance area is the path point list, which is output sequentially to the output file. This step has a time complexity of O(N), suitable for real-time computation. The specific task flow is as follows: Figure 1 As shown.
[0030] For the search area: For high-altitude group aircraft performing daytime missions, after the lead aircraft receives the coordinates of the buildings in the search area transmitted back by the reconnaissance UAV, it will make plans and request the position information of all current high-altitude group wingmen. The four nearest aircraft will be dispatched to the output position, that is, the four corners of the smallest bounding box of the building, so as to conduct a circumferential search of the building. The remaining aircraft will continue the search.
[0031] The main parameters for building search include horizontal scan count, vertical step length, overlap rate, and safe height. An obstacle avoidance mechanism and a low battery replacement mechanism are added using the artificial potential field method.
[0032] This step has a time complexity of O(building facade area * number of buildings found) and a total latency of <300ms, making it suitable for real-time computation. The specific task flow is as follows: Figure 2 As shown.
[0033] The communication mechanism is as follows: Ground control station >> Lead aircraft: Assign building search tasks (facade parameters) Lead aircraft >> Wingman: Synchronize path planning data + acquisition commands Loop Each scan line The lead aircraft calculates the next waypoint Lead aircraft >> Wingman: Broadcast waypoint + timestamp Wingman >> Leading aircraft: Arrival confirmed Ground control station >> Lead aircraft: Completion report For the surrounding area, low-altitude aircraft excluding relay aircraft will be used for the surrounding scan. The input search area coordinates will be planned according to the number of available aircraft. The coordinates and range of the surrounding area can be the same as the coordinates and range of the reconnaissance area. Aircraft will be dispatched to designated locations to perform the ring scan mission. This step has a time complexity of O(the square of the number of discretized points * the number of UAVs), with a total latency of <100ms, suitable for real-time computation. The specific task flow is as follows: Figure 3 As shown.
[0034] (3) The planning module outputs the coordinates and altitude of the planned points in sequence and sends them to the flight control software in a pre-agreed serial port format. Wingmen that do not receive the next waypoint will remain in their original positions, as shown in the table below.
[0035]
[0036] (4) Finally, the planned target points are displayed in sequence according to the tasks performed by different aircraft. Specifically, the point data after the path planning is obtained. Each UAV has a path consisting of multiple points (latitude and longitude), which are stored in the output parameter file in the order of coordinates. The polyline is drawn using the Google Maps JavaScript API, and different UAVs are distinguished by different colors. The path can also be updated dynamically.
[0037] In summary, the swarm UAV path planning method provided in this application, compared with the previous centralized planning, eliminates the risk of single point of failure by relying on the central controller (ground station) for global optimization. It saves communication bandwidth and improves the real-time performance of communication through algorithm optimization and hardware adaptation. It is more advantageous for large-scale swarm control and meets the task requirements of the onboard pods and vision modules for subsequent identification and obstacle avoidance tasks.
[0038] 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 a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0039] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.
[0040] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for system or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the descriptions in the method embodiments. The systems and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.
[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.
Claims
1. A method for path planning of swarmed unmanned aerial vehicles (UAVs), characterized in that, include: The system receives input parameters from the user, including reconnaissance area data, surround view area data, and configuration data for each UAV; the configuration data includes the group to which the UAV belongs, which includes a high-altitude group and a low-altitude group. One of the UAVs from the high-altitude group is selected as the decision-making machine; Based on the reconnaissance area data, reconnaissance paths are planned for all UAVs in the high-altitude group and the low-altitude group respectively, and the results of the reconnaissance path planning for each UAV are sent to the corresponding UAV so that the UAV that receives the results of the reconnaissance path planning can perform the reconnaissance mission. After performing a reconnaissance mission, the UAV uploads the search area data and sends it to the decision-making machine. The decision-making machine is used to obtain the coordinates of all UAVs in the high-altitude group currently performing a reconnaissance mission, perform search path planning for the multiple UAVs closest to the search area, and send the search path planning results for each UAV to the corresponding UAV so that the UAV that receives the search path planning results can perform a search mission on the search target. All available UAVs in the low-altitude group are identified, and surround-view path planning is performed for all available UAVs based on the surround-view area data. The surround-view path results for each UAV are then sent to the corresponding UAV so that the UAV receiving the surround-view path results can perform the surround-view task.
2. The swarm drone path planning method according to claim 1, characterized in that, The reconnaissance area data includes an irregular quadrilateral defined on the map by the latitude and longitude of four points, as well as the reconnaissance boundary defined by humans.
3. The swarm drone path planning method according to claim 2, characterized in that, The rules for reconnaissance path planning include selecting the longest side of the minimum bounding box of the irregular quadrilateral for scanning, determining the number of scan lines based on the coverage width, and using the set of intersections between the scan lines and the boundary of the reconnaissance area as a path point list, which is output sequentially to the output file.
4. The swarm drone path planning method according to claim 2, characterized in that, The surround view area data includes an irregular quadrilateral formed by the latitude and longitude of four points identical to the reconnaissance area data, as well as artificially defined surround view boundaries.
5. The swarm drone path planning method according to claim 4, characterized in that, The surround view area data also includes the aircraft scanning radius, which is the pod coverage width to ensure recognition accuracy.
6. The swarm drone path planning method according to claim 1, characterized in that, The search area includes buildings, and the search area data includes regular polygons formed by the latitude and longitude of multiple points of the buildings.
7. The swarm drone path planning method according to claim 1, characterized in that, The group also includes a relay group. When the search distance is too long, the number of aircraft in the cluster is large, or the search area is far from the operator, relay path planning is performed on each UAV in the relay group, and the result of the relay path planning for each UAV is sent to the corresponding UAV so that the UAV that receives the result of the relay path planning can perform the relay task and act as a message forwarding hub.
8. The swarm drone path planning method according to claim 1, characterized in that, The planned target points will be displayed sequentially along the routes according to the missions performed by different aircraft.