A method and system for cooperative navigation of heterogeneous swarms of unmanned aerial vehicles in complex terrain

Abstract

The present application provides a method and system for cooperative navigation of heterogeneous clusters of unmanned aerial vehicles (UAVs) in complex terrain. Among them, the method collects communication link quality data and position information relative to terrain obstacles between nodes in the cluster; dynamically reconstructs the radiation direction of the selected node's airborne antenna according to the distribution of communication attenuation areas, and establishes a directional communication link around terrain obstacles; associates the connection status of the directional link with the position information, and constructs the joint distribution characteristics of the cluster communication capability and spatial configuration; optimizes the cluster topology using a distributed collaborative mechanism, and dynamically adjusts the relative distance constraints between nodes based on the terrain's line of sight obstruction range and the UAV's heterogeneous maneuverability; generates a cooperative navigation instruction set that adapts to terrain and interference based on the optimized topology, and controls the nodes to perform three-dimensional track correction. The present application improves the navigation continuity of heterogeneous clusters of UAVs in complex terrain environments.

Description

Technical Field

[0001] The present application relates to the field of UAV collaborative navigation technology, and in particular to a method and system for collaborative navigation of heterogeneous clusters of UAVs in complex terrain. Background Art

[0002] In complex combat or disaster relief environments with strong electromagnetic signal interference (such as those around high-voltage power stations and in hostile electronic countermeasures zones), heterogeneous drone swarms must maintain high-precision collaborative navigation in GPS-denied environments. This scenario requires the swarm to overcome the dual effects of terrain obstruction and electromagnetic interference to ensure reliable transmission of navigation commands; adapt to communication link disruptions caused by obstacles and reestablish swarm connectivity; and integrate the hovering capabilities of rotorcraft with the high-speed characteristics of fixed-wing aircraft to generate differentiated maneuvering commands.

[0003] The current mainstream solution adopts fixed-topology collaborative control based on a combination of GPS and inertial navigation: the cluster relies on the fusion of GPS global positioning and inertial navigation data to generate a reference track; a hierarchical star topology structure is preset, and the master node broadcasts navigation instructions; the slave nodes execute formation flight according to preset following rules.

[0004] Existing solutions have the following flaws: when GPS signals fail in interference zones, the accumulated inertial navigation errors cause the formation to drift, and the cluster cannot maintain its relative position; fixed topology cannot dynamically avoid terrain obstructions; once an obstacle appears between the master and slave nodes, command transmission is interrupted, causing the formation to collapse; unified following rules do not take into account the maneuverability differences between rotorcraft and fixed-wing aircraft; during high-speed turns, the rotorcraft lags behind, resulting in a sharp increase in collision risk. Summary of the Invention

[0005] The present application provides a method and system for collaborative navigation of heterogeneous clusters of unmanned aerial vehicles (UAVs) in complex terrain, which is used to solve the problem of insufficient navigation continuity of heterogeneous clusters of UAVs in complex terrain environments in the prior art.

[0006] In a first aspect, the present application provides a method for cooperative navigation of heterogeneous swarms of UAVs in complex terrain, comprising:

[0007] When a heterogeneous UAV cluster flies to an area with both undulating terrain and signal interference, the communication link quality data between UAV nodes in the heterogeneous UAV cluster is collected, and the position information of each UAV relative to the terrain obstacles is simultaneously obtained. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication;

[0008] Dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node according to the distribution of communication attenuation areas indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna;

[0009] Associating the connection state of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment;

[0010] Based on the joint distribution characteristics, a distributed collaborative mechanism is used to optimize the topology of the heterogeneous cluster of drones. During the optimization process, the relative distance constraints of the drone nodes are dynamically adjusted according to the range of line of sight obstruction caused by the undulating terrain and the heterogeneous maneuverability of the drones.

[0011] A collaborative navigation instruction set that adapts to terrain undulations and signal interference is generated based on the optimized topological structure, and the collaborative navigation instruction set is used to control the UAV nodes to perform three-dimensional track correction to maintain the overall navigation continuity of the cluster.

[0012] Optionally, dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node according to the communication attenuation area distribution indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna, includes:

[0013] parsing the boundary coordinates of an area where the signal attenuation value exceeds a preset threshold in the communication link quality data, and identifying a communication node pair within the boundary coordinates of the area that is most severely blocked by the terrain obstacle;

[0014] For the communication node pair, extract the relative azimuth and elevation angle difference between the transmitting and receiving drones, determine the horizontal angle range in which the beam main lobe needs to be deflected based on the relative azimuth, and determine the vertical angle range in which the beam main lobe needs to be raised or lowered based on the elevation angle difference;

[0015] In the radiation direction of the airborne antenna of the transmitting UAV, the main lobe of the beam is switched to the intersection of the horizontal angle range and the vertical angle range, and the beam width is simultaneously narrowed to a preset narrow beam mode to focus the transmission energy on the direction of the receiving UAV;

[0016] The receiving-end drone synchronously performs switching and contraction operations, so that the narrow beams of the transmitting-end and receiving-end antennas establish a directional communication link around the terrain obstacle on the side of the terrain obstacle.

[0017] Optionally, the topology of the heterogeneous cluster of drones is optimized using a distributed collaborative mechanism based on the joint distribution characteristics. During the optimization process, the relative distance constraints of the drone nodes are dynamically adjusted according to the range of line of sight obstruction caused by terrain undulations and the heterogeneous maneuverability of the drones, including:

[0018] Extracting the communication attenuation gradient distribution within the terrain obstacle occlusion contour from the joint distribution feature, and calculating the communication reachable spatial range of the drone node affected by the terrain occlusion in combination with the current three-dimensional position of the drone node;

[0019] Determining a dynamic distance constraint threshold required to maintain reliable communication between the UAV nodes based on the communication reachable space range and flight mode parameters of adjacent UAV nodes;

[0020] Under the distributed collaborative mechanism, the UAV node broadcasts the communication reachable space range and the dynamic distance constraint threshold to neighboring nodes, receives corresponding data from the neighboring nodes, and negotiates to generate a topological connection priority list, wherein the topological connection priority list arranges the node connection relationships in descending order of communication stability;

[0021] According to the topological connection priority list, low-priority connections within the occlusion contour of the terrain obstacle are removed, and alternative communication paths across the occlusion area are added to complete the topological structure optimization of the heterogeneous cluster of drones.

[0022] Optionally, generating a collaborative navigation instruction set adapted to terrain undulations and signal interference based on the optimized topological structure, and controlling the UAV nodes to perform three-dimensional track correction to maintain overall navigation continuity of the cluster through the collaborative navigation instruction set, includes:

[0023] Extracting the communication connection relationship between the drone nodes and the corresponding reference waypoints from the optimized topology structure, wherein the reference waypoints are located in the gaps between terrain obstacles and meet the dynamic distance constraint threshold;

[0024] Calculate the heading angle, pitch angle, and speed adjustment required for the UAV based on the relative orientation of the reference waypoint and the current three-dimensional position of the UAV, and generate individual track correction parameters;

[0025] After exchanging the individual track correction parameters through adjacent nodes under the communication path defined by the communication connection relationship, a cluster coordinated action sequence is generated to avoid synchronous maneuver conflicts;

[0026] The cluster coordinated action sequence is converted into a flight control steering quantity instruction and a power output instruction, and the flight control steering quantity instruction and the power output instruction are used to drive the UAV to perform three-dimensional track correction with coordinated changes in lift, yaw and speed along the gap between the terrain obstacles.

[0027] Optionally, associating the connection state of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment includes:

[0028] Quantifying the connection status of the directional communication link as a communication quality index including link stability and signal strength values, wherein the link stability is calculated inversely proportional to the number of consecutive communication interruptions, and the signal strength value is obtained from the radio frequency signal sampling value of the receiving end drone;

[0029] Mapping the communication quality index to the three-dimensional spatial coordinates of the corresponding drone position point to generate a communication quality spatial distribution map with the corresponding drone position point as the carrier;

[0030] Marking a vertical projection plane of the terrain obstacle in the communication quality spatial distribution map, identifying a mutation boundary where the difference in the communication quality index on both sides of the vertical projection plane exceeds a preset drop, wherein the mutation boundary represents the range of the terrain blocking effect on the communication link;

[0031] The three-dimensional spatial relationship between the communication quality spatial distribution map and the mutation boundary is superimposed to form a joint distribution feature that simultaneously reflects the communication capability attenuation gradient distribution of the heterogeneous cluster of drones and the occlusion contour of the terrain obstacle.

[0032] Optionally, in the radiation direction of the airborne antenna of the transmitting UAV, switching the main lobe direction of the beam to the intersection of the horizontal angle range and the vertical angle range, and simultaneously shrinking the beam width to a preset narrow beam mode to focus the transmission energy on the direction of the receiving UAV, including:

[0033] Determining the azimuth of a three-dimensional target to which the main lobe of the beam is to be directed based on the center value of the horizontal angle range and the center value of the vertical angle range;

[0034] driving the antenna radiation unit of the transmitting UAV to rotate to a horizontal deflection angle corresponding to the orientation of the three-dimensional space target, and adjusting the antenna radiation unit to a vertical pitch angle corresponding to the orientation of the three-dimensional space target after the rotation is completed;

[0035] Narrowing the excitation phase distribution range of the adjusted antenna radiating element to a preset narrow beam mode, so that the electromagnetic wave energy is concentrated to cover a conical area centered on the three-dimensional target orientation through the narrowing process;

[0036] Detect the radio frequency signal strength at the location of the receiving drone, and dynamically fine-tune the horizontal deflection angle and the vertical pitch angle so that the conical area completely covers the location of the receiving drone.

[0037] Optionally, the step of removing low-priority connections within the terrain obstacle occlusion outline based on the topological connection priority list and adding alternative communication paths across the occlusion area to complete the topology optimization of the heterogeneous drone cluster includes:

[0038] Traversing the topology connection priority list, taking the connection relationship that ranks last in the communication stability ranking in the topology connection priority list and is located within the terrain obstacle occlusion outline as a marked connection;

[0039] Selecting drone nodes outside the terrain obstacle occlusion outline that have neighbor relationships with both end nodes connected to the marker as relay candidate nodes;

[0040] Establishing a communication connection relationship between the two end nodes and the candidate relay node, and forming an alternative path that crosses the occlusion contour of the terrain obstacle based on the communication connection relationship;

[0041] The marked connection is deleted and the alternative path is added to complete the topology optimization of the heterogeneous cluster of drones.

[0042] Secondly, this application provides a UAV heterogeneous cluster collaborative navigation system for complex terrain, including:

[0043] A collection module is used to collect communication link quality data between drone nodes within a heterogeneous drone cluster when the cluster flies to an area with both terrain undulations and signal interference, and simultaneously obtain the position information of each drone relative to terrain obstacles. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication;

[0044] a reconstruction module, configured to dynamically reconstruct the radiation direction of the airborne antenna of the selected UAV node based on the distribution of the communication attenuation area indicated by the communication link quality data, and establish a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna;

[0045] an association module, configured to associate the connection status of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment;

[0046] an optimization module for optimizing the topology of the heterogeneous cluster of drones using a distributed collaborative mechanism based on the joint distribution characteristics, and dynamically adjusting the relative distance constraints of the drone nodes according to the range of line of sight obstruction caused by terrain undulations and the heterogeneous maneuverability of the drones during the optimization process;

[0047] The correction module is used to generate a collaborative navigation instruction set that adapts to terrain undulations and signal interference based on the optimized topological structure, and to control the UAV nodes to perform three-dimensional track correction to maintain the overall navigation continuity of the cluster through the collaborative navigation instruction set.

[0048] In a third aspect, the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a heterogeneous cluster collaborative navigation method for complex terrain of unmanned aerial vehicles as described in the first aspect above.

[0049] In a fourth aspect, the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a method for collaborative navigation of heterogeneous clusters of unmanned aerial vehicles for complex terrain as described in the first aspect.

[0050] When a heterogeneous drone cluster flies into an area with both undulating terrain and signal interference, the present application collects communication link quality data between drone nodes within the heterogeneous drone cluster and simultaneously obtains the position information of each drone relative to terrain obstacles. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication. Based on the distribution of communication attenuation areas indicated by the communication link quality data, the application dynamically reconstructs the radiation direction of the airborne antenna of the selected drone node and establishes a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or narrowing the beam width in the airborne antenna radiation direction. The application associates the connection status of the directional communication link with the position information to construct a joint distribution feature that reflects the cluster communication capability and spatial configuration under terrain obstruction. Based on the joint distribution feature, the application optimizes the topology of the heterogeneous drone cluster using a distributed collaborative mechanism. During the optimization process, the relative distance constraints of the drone nodes are dynamically adjusted based on the range of line of sight obstruction caused by the undulating terrain and the heterogeneous maneuverability of the drones. Based on the optimized topology, a collaborative navigation instruction set that adapts to the undulating terrain and signal interference is generated. The collaborative navigation instruction set controls the drone nodes to perform three-dimensional track corrections to maintain overall navigation continuity of the cluster.

[0051] The technical solution of this application has the following beneficial effects:

[0052] Perceive the dual coupling effects of terrain obstruction and electromagnetic interference on cluster communications, and provide a data basis for anti-interference communications; establish directional communication links around terrain obstacles through dynamic antenna reconstruction, breaking through communication interruptions caused by physical obstructions; construct joint distribution characteristics of communication capabilities and spatial configurations, and quantify the dynamic constraints of terrain obstruction on cluster connectivity; dynamically optimize the topological structure based on communication and spatial characteristics, and synchronously adapt to terrain line of sight obstruction and heterogeneous maneuverability differences; generate three-dimensional collaborative navigation instructions to drive the cluster to maintain continuous formation flight in complex electromagnetic terrain environments.

[0053] Furthermore, the communication link quality data is analyzed to identify the boundaries of areas where signal attenuation exceeds the threshold, locating the communication node pairs most severely obstructed by terrain obstacles. The relative azimuth and elevation angle differences of these node pairs are extracted to determine the horizontal angle range and vertical angle range within which the main lobe beam needs to be deflected. At the transmitting end, the main lobe direction is switched to the intersection of the horizontal and vertical angles, and the beam is contracted to a narrow beam mode to focus energy toward the receiving end. The receiving end simultaneously performs the same operation, allowing the bidirectional narrow beam to form a signal reflection channel off the side of the terrain obstacle, establishing a directional link around the obstacle. By identifying attenuation boundaries and calculating angle differences, the communication node pairs requiring reconstruction are precisely located. Three-dimensional beam direction switching and narrow beam contraction focus the transmitted energy toward the receiving end. The bidirectional narrow beams collaboratively construct a reflection channel around terrain obstacles, significantly enhancing link stability in interference environments.

[0054] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0056] Figure 1 A flowchart of a heterogeneous cluster collaborative navigation method for complex terrain provided by the present application is shown;

[0057] Figure 2 A scenario diagram showing a method for cooperative navigation of heterogeneous clusters of UAVs in complex terrain provided by this application is shown;

[0058] Figure 3 The present invention provides a schematic diagram of the structure of a UAV heterogeneous cluster collaborative navigation system for complex terrain;

[0059] Figure 4 A schematic structural diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0060] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0061] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0062] The existing fixed-topology collaborative control scheme based on the combination of GPS and inertial navigation has three contradictions: when GPS fails in an electromagnetic interference area, the accumulated inertial navigation error causes the formation to drift, resulting in cluster instability; the fixed star topology cannot avoid terrain obstruction, and once an obstacle appears between the master and slave nodes, the command is interrupted, causing the formation to collapse; the unified following rule ignores the maneuverability differences between rotorcraft and fixed-wing aircraft, and the rotorcraft's response lag during high-speed maneuvers increases the risk of collision.

[0063] To address these shortcomings, the present invention proposes an autonomous navigation method that coordinates communication, antenna, and topology. Its core innovations include: sensing the coupled effects of terrain obstruction and electromagnetic interference through communication link quality data; dynamically reconfiguring antenna beams to establish directional communication links around terrain obstacles; fusing communication status and location information to construct a joint distribution feature that quantifies the dynamic constraints imposed by terrain obstruction on the cluster; and using this feature to drive distributed topology optimization, synchronously adapting to terrain line-of-sight obstruction and heterogeneous maneuverability. Based on the optimized topology, differentiated navigation commands are generated to control the cluster's execution of three-dimensional track corrections.

[0064] The technical solution of the present application can be applied to scenarios where communications are unstable in signal interference areas and drone clusters need to maintain coordination without relying on external networks.

[0065] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0066] Figure 1 The present invention provides a flowchart of a method for cooperative navigation of heterogeneous clusters of UAVs in complex terrain. Figure 1As shown, the method includes: 101. When a heterogeneous cluster of drones flies to an area where both terrain undulations and signal interference exist, collecting communication link quality data between drone nodes in the heterogeneous cluster of drones, and synchronously obtaining position information of each drone relative to terrain obstacles, wherein the communication link quality data reflects the dual effects of terrain obstruction and signal interference on cluster communication;

[0067] When a heterogeneous swarm of drones flies into an area with both rugged terrain and signal interference, the system first uses its onboard RF module to collect communication link quality data between each drone node in the swarm. This data includes RF signal strength sampling values, the number of consecutive communication interruptions, and the bit error rate. These data collectively reflect the dual impact of terrain obstruction and signal interference on communication. Simultaneously, each drone uses a fusion of LiDAR and GPS / IMU positioning systems to obtain its precise position relative to terrain obstacles, including horizontal azimuth, pitch angle, and relative distance, ensuring accurate matching of communication data and spatial position.

[0068] To achieve efficient data utilization, the system aligns collected communication quality data with location information using a unified timestamp and spatial grid. For example, if a drone detects a sudden drop in signal strength from -70dBm to -95dBm and an 8% bit error rate, combined with its location data indicating a horizontal azimuth of 90 degrees, a pitch angle of 5 degrees, and a distance of 30 meters from a high-voltage tower, it can be determined that the signal degradation is primarily due to electromagnetic interference, not terrain obstruction. Conversely, if a drone is located in a valley with a pitch angle of 35 degrees and is only 20 meters from a mountain peak, and the signal strength drops to -90dBm and communication is interrupted, it is clearly due to physical obstruction.

[0069] Through this synchronized data collection and joint analysis mechanism, the system can dynamically distinguish the degree of impact of terrain obstruction and external interference on the communication link. For example, when a drone simultaneously detects building obstruction and abnormal signal fluctuations, it can accurately identify the combined impact of these two factors by combining location characteristics (a 25-degree pitch angle, a 15-meter distance), signal strength of -88dBm, and a bit error rate of 7%. This process provides a reliable basis for subsequent cluster path planning and anti-interference strategies, ensuring stable cluster collaboration in complex environments.

[0070] 102. Dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node based on the distribution of communication attenuation areas indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna;

[0071] Optionally, step 102 may specifically include the following steps: 1021. parsing the coordinates of the region boundary where the signal attenuation value exceeds a preset threshold in the communication link quality data, and identifying the communication node pair most severely blocked by the terrain obstacle within the region boundary coordinates;

[0072] 1022. For the communication node pair, extract the relative azimuth and elevation angle difference between the transmitting UAV and the receiving UAV, determine the horizontal angle range by which the beam main lobe needs to be deflected based on the relative azimuth, and determine the vertical angle range by which the beam main lobe needs to be raised or lowered based on the elevation angle difference;

[0073] 1023. Switch the main lobe direction of the beam to the intersection of the horizontal angle range and the vertical angle range in the radiation direction of the airborne antenna of the transmitting UAV, and simultaneously shrink the beam width to a preset narrow beam mode to focus the transmission energy on the direction of the receiving UAV;

[0074] Among them, step 1023 may specifically include the following processes: determining the three-dimensional space target orientation to which the main lobe of the beam needs to point based on the center value of the horizontal angle range and the center value of the vertical angle range; driving the antenna radiating unit of the transmitting drone to rotate to the horizontal deflection angle corresponding to the three-dimensional space target orientation, and adjusting the antenna radiating unit to the vertical pitch angle corresponding to the three-dimensional space target orientation after the rotation is completed; reducing the excitation phase distribution range of the adjusted antenna radiating unit to a preset narrow beam mode, and through the reduction process, the electromagnetic wave energy is concentrated to cover a conical area centered on the three-dimensional space target orientation; detecting the RF signal strength at the orientation of the receiving drone, and dynamically fine-tuning the horizontal deflection angle and the vertical pitch angle so that the conical area completely covers the position point of the receiving drone.

[0075] 1024. The receiving-end drone synchronously performs switching and contraction operations, so that the narrow beams of the transmitting-end and receiving-end antennas establish a directional communication link on the side of the terrain obstacle that bypasses the terrain obstacle.

[0076] In the above scheme, the communication attenuation area distribution refers to the continuous spatial range where the signal attenuation value exceeds the preset threshold, and its boundary is composed of a sequence of three-dimensional coordinate points. A communication node pair is a transmitter and receiver drone combination where communication quality degrades due to obstruction by terrain obstacles. The relative azimuth angle is the angle between the transmitter and receiver drones in the horizontal plane. The elevation difference is the angle formed by the difference in vertical height between the transmitter and receiver drones. A narrow beam pattern is an energy-concentrated radiation pattern formed by narrowing the phase distribution range of the antenna radiating elements. A signal reflection channel is a non-line-of-sight communication path formed by the side reflection of a bidirectional narrow beam from a terrain obstacle.

[0077] In this embodiment of the application, when a drone cluster detects an area in the communication link quality data where the signal attenuation value exceeds a preset threshold, the system dynamically reconfigures the radiation direction of the selected drone node's onboard antenna, adjusting the main lobe direction or shrinking the beam width to establish a directional communication link that bypasses terrain obstacles. The specific implementation process is as follows:

[0078] First, the system analyzes the communication link quality data, screening out the boundary coordinates of continuous areas where the signal attenuation value is below a preset threshold (such as -10dB). It then uses a density clustering algorithm to merge adjacent attenuation points to form the outer contour of the obstructed area. Within this area, the system further analyzes the packet loss rate and signal strength of each drone node pair to identify the communication node pairs most severely affected by terrain obstruction. For example, in a certain mountain-obstructed area, the signal attenuation within the coordinate range of (100,50,80) to (120,70,90) was analyzed to be -15dB. After clustering, this area was confirmed to be a strong obstruction area, and the communication packet loss rate from drone A to drone B was identified as as high as 60%, making it the node pair most in need of optimization.

[0079] Next, the system calculates the relative azimuth and elevation difference between the transmitting and receiving drones to determine the adjustment range of the main lobe beam. The relative azimuth angle θ is calculated using the 3D coordinates of the GPS / INS fusion system using the formula θ=atan²(Ry-Sy,Rx-Sx), with the horizontal angle range set to [θ-5°,θ+5°]. The elevation difference φ is calculated using the formula φ=atan²(Rz-Sz,sqrt((Rx-Sx)²+(Ry-Sy)²), with the vertical angle range set to [φ-3°,φ+3°]. For example, if transmitter A is located at (0,0,0) and receiver B is located at (100,100,50), the relative azimuth angle θ = 45°, with a horizontal adjustment range of 40°-50°. The elevation difference φ is ≈19.5°, with a vertical adjustment range of 16.5°-22.5°.

[0080] The transmitting drone's onboard antenna then adjusts its main beam lobe toward the target based on the calculated horizontal and vertical angle ranges and narrows the beamwidth to a preset narrow beam pattern (for example, from 60° to 10°) to concentrate the transmitted energy. The system then monitors the signal strength at the receiving end and determines the beam adjustment is successful if it improves by more than 10dB. For example, if drone A's antenna main lobe is switched to a 45° horizontal angle and a 19.5° elevation angle, and the beamwidth is narrowed to 10°, the signal strength on drone B will increase from -85dB to -75dB, confirming that the beam focusing is effective.

[0081] Finally, the receiving drone synchronously performs reverse beam adjustment, calculates the transmitting drone's reverse angle (θ + 180°, -φ), and similarly narrows the beamwidth, enabling the bidirectional narrow beam to bypass terrain obstacles (such as the side of a mountain) and establish a stable non-line-of-sight communication link. The system continuously monitors link quality and determines that the obstacle-avoidance communication link is successfully established if the bit error rate (BER) remains below 1e-6 for at least 10 seconds. For example, if drone B's antenna is pointed at a horizontal angle of 225° and an elevation angle of -19.5°, the bidirectional narrow beam, after reflecting off the mountain, establishes a stable connection, reducing the BER to 5e-7 and achieving reliable communication.

[0082] In a practical application, assuming a swarm of drones flies into an area obscured by a mountain in a strong electromagnetic interference environment near a high-voltage power station, the processor analyzes communication link quality data. In step 1021, all inter-node communication packets are traversed, and the coordinates of points with signal attenuation values ​​less than -10 dB are selected (102.3, 205.7, 50.1) to (108.9, 210.5, 52.8). The DBSCAN spatial clustering algorithm is used to cluster the discrete attenuation points into three continuous regions. A comprehensive degradation index is calculated for each node pair within each region. The comprehensive degradation index is equal to the packet loss rate multiplied by 0.6 plus the signal attenuation value multiplied by 0.4. The highest comprehensive degradation index of 0.82 is identified for the fixed-wing and rotary-wing drones in region 2 (packet loss rate 0.7), making them the most severely obscured node pair. Through step 1022, for the fixed-wing (100.0, 200.0, 30.0) and rotor-wing (250.0, 400.0, 120.0) node pairs, the horizontal relative azimuth angle is calculated: ΔX = 150.0, ΔY = 200.0, azimuth angle θ = atan2(200, 150) = 53.13°, and the horizontal angle range is set to [48.13°, 58.13°] (±5° tolerance). The vertical elevation angle difference is calculated: horizontal distance = sqrt(150² + 200²) = 250.0 meters, altitude difference ΔZ = 90.0 meters, pitch angle φ = atan(90 / 250) ≈ 19.80°, and the vertical angle range is set to [16.80°, 22.80°] (±3° tolerance). In step 1023, antenna reconfiguration is performed on the transmitting fixed-wing UAV: ​​the 2.4 GHz phased array antenna's horizontal rotation mechanism is driven to 53.13°, and the pitch adjustment mechanism is driven to 19.80°. The phase distribution of the 64 antenna elements is switched from an omnidirectional mode to a narrow beam mode with a beamwidth of 10°. The phase offset is calculated as ΔΦ = 2πdsin(α) / λ, where d is the element spacing of 0.6λ. The rotor-end signal strength is monitored and increases from -92 dBm to -77 dBm (increment of 15 dB), verifying that energy focusing is successful. The Bravo drone on the receiving end responds synchronously through step 1024: the angle pointing to Alpha is calculated according to the reverse geometric relationship: the horizontal angle is 53.13° + 180° = 233.13°, and the pitch angle is -19.80°; the same antenna rotation and beam contraction operations are performed; the bidirectional narrow beam passes through the granite side of the mountain, coordinates (180.0, 300.0, 80.0), and after reflection, a signal superposition area is formed, and the link bit error rate is reduced from 1.2×10⁻ 4 Reduced to 3.5×10⁻ 7 , establish a stable obstacle-avoiding communication channel with a bandwidth of 5Mbps.

[0083] The overall solution of step 102 above, by dynamically adjusting the beam direction and width of the drone antenna, allows the communication signal to bypass terrain obstacles, effectively solving the communication obstruction problem in complex environments and significantly improving the collaborative communication capability of the drone cluster in obstructed areas.

[0084] 103. Associating the connection status of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment;

[0085] Optionally, step 103 may specifically include the following steps: 1031. Quantifying the connection status of the directional communication link as a communication quality index including link stability and a signal strength value, wherein the link stability is calculated inversely proportional to the number of consecutive communication interruptions, and the signal strength value is obtained from a radio frequency signal sampling value of the receiving end drone;

[0086] 1032. Map the communication quality index to the three-dimensional spatial coordinates of the corresponding drone position point to generate a communication quality spatial distribution map with the corresponding drone position point as a carrier;

[0087] 1033. Mark the vertical projection plane of the terrain obstacle in the communication quality spatial distribution map, and identify a sudden change boundary where the difference in the communication quality index on both sides of the vertical projection plane exceeds a preset drop, wherein the sudden change boundary represents the range of the terrain blocking effect on the communication link;

[0088] 1034. Superimpose the three-dimensional spatial relationship between the communication quality spatial distribution map and the mutation boundary to form a joint distribution feature that simultaneously reflects the communication capability attenuation gradient distribution of the heterogeneous UAV cluster and the occlusion contour of the terrain obstacle.

[0089] In the above scheme, the communication quality index is a comprehensive evaluation indicator composed of a weighted combination of the link stability coefficient (calculated inversely proportional to the number of consecutive communication interruptions) and the RF signal strength value. The communication quality spatial distribution map is a visualized 3D heat map generated by binding the communication quality index values ​​at corresponding locations to the drone's 3D geographic coordinates. The vertical projection surface is a 2D polygonal area formed by orthogonally projecting the point cloud data of the terrain obstacle surface onto a horizontal reference plane along the direction of gravity. The abrupt boundary is a continuous spatial boundary at the edge of the vertical projection surface where the communication quality index undergoes a step-like change, with the index difference on both sides exceeding 20 decibels. The communication capability attenuation gradient distribution is a characteristic of the rate at which the communication quality index value decreases with spatial position. The terrain obstacle obstruction contour is a 3D model of the terrain obstacle's impact range, enclosed by the abrupt boundary.

[0090] In an embodiment of the present application, the connection status of the directional communication link is first quantitatively evaluated, and the communication quality index is generated by calculating the link stability and signal strength value. The link stability is derived from the inverse ratio of the number of consecutive communication interruptions. For example, if a link is interrupted twice, the stability coefficient is 1 / (2+1)=0.333; at the same time, the RF signal strength value of the receiving drone is collected, such as -80dBm which is normalized to 0.6. These two indicators are weighted and summed according to a weight of 6:4 to obtain a communication quality index in the range of 0-1, for example, 0.333×0.6+0.6×0.4=0.48. These index values ​​are then associated with the three-dimensional spatial coordinates of the drone, and an inverse distance weighted interpolation algorithm is used to generate a continuous communication quality spatial distribution map within a radius of 5 meters, and a red, orange and green heat map is used to intuitively display the communication quality of different areas. For example, the index 0.48 at the coordinates (102.3, 205.7, 50.1) is displayed as an orange area.

[0091] After establishing a spatial distribution map of communication quality, the system further analyzes the impact of terrain obstacles. By projecting the terrain point cloud data collected by the LiDAR vertically, the system constructs two-dimensional polygonal projection surfaces of the terrain obstacles. Along the edges of these projection surfaces, the system detects sudden changes in the communication quality index at 1-meter intervals. When the difference between the inner and outer indices exceeds 20 decibels, it marks them as sudden changes. For example, at the projection boundary point (150.0, 280.0) of the high-voltage tower, the index difference between the inner index of 0.35 (red) and the outer index of 0.85 (green) is 50 decibels, significantly exceeding the threshold and thus identified as a valid sudden change boundary. These sudden changes clearly indicate the extent to which terrain obstacles disrupt the communication link.

[0092] Finally, the system overlays the spatial distribution map of communication quality with the mutation boundary in three dimensions to form a complete joint distribution feature. Within the area enclosed by the mutation boundary, the system calculates the gradient of the communication quality index and uses arrows to indicate the direction of attenuation. For example, a red gradient arrow pointing towards 0.2 is displayed within the tower area. The mutation boundary is then stretched vertically to the actual terrain height to construct a gray, semi-transparent three-dimensional shell model. This visualization simultaneously reflects the gradient distribution of the drone swarm's communication capability attenuation and the outlines of terrain obstacles, providing an important basis for subsequent path planning and communication optimization. For example, in a mountainous mission, the system used this joint distribution feature to accurately identify communication blind spots on the back of a mountain peak and guide the drones to adjust their formation positions to avoid these areas.

[0093] In practical applications, assuming a high-voltage power station with strong electromagnetic interference, the F07 rotorcraft drone performs communication state quantization as the receiver of a directional link. In step 1031, the RF front end captures the communication signal at a 20MHz sampling rate, measuring an instantaneous peak-to-peak voltage of 350 millivolts. The RMS value is calculated: the RMS voltage is equal to 350 divided by the square root of 2, which is approximately 247.49 millivolts. This is converted to power: power in milliwatts is equal to the RMS voltage squared divided by a 50-ohm impedance, or (247.49)^2 / 50≈1224.99 milliwatts. The decibel milliwatt value is calculated: PdBm=10×log10(1224.99)≈-7 8.2dBm; normalized: strength coefficient = (-78.2 + 100) / 50 = 0.436, normalized range -100dBm to -50dBm corresponds to 0 to 1); the communication protocol stack counts three lost ACK response packets within a 500ms time window (packet interval 50ms), and determines three valid interruptions; calculates the stability coefficient: 1 / (3 + 1) = 0.25; composite communication quality index: 0.25 × 0.6 + 0.436 × 0.4 = 0.324.

[0094] Through step 1032, a spatial distribution map is constructed based on the F07 UAV's position: the combined navigation output WGS84 coordinates are 118.7523 degrees east longitude, 32.0415 degrees north latitude, and 150.6 meters above sea level; converted to a local rectangular coordinate system: X = 305700.3 meters, Y = 408205.7 meters, and Z = 150.6 meters; a 5-meter radius spherical grid (resolution 0.5 meters) is established with this point as the center. The grid point index value is calculated by inverse distance weighted interpolation: the numerator = the index value of each neighboring UAV divided by The denominator is the sum of the squares of the distances, and the denominator is the sum of the reciprocals of the squares of the distances. The grid point index value = numerator / denominator, where the index of the five neighboring drones is taken from 0.28 to 0.41, and the distance is the Euclidean distance. The index of the center point (305700.3, 408205.7, 150.6) is 0.324, and the index of the boundary point (305703.0, 408210.0, 152.0) is 0.38. Rendering is based on the value range: less than 0.3 is dark red, 0.3 to 0.35 is orange-red, and greater than 0.35 is bright orange.

[0095] Step 1033 integrates terrain data to calibrate the sudden change boundary: LiDAR scans the high-voltage tower to obtain 12,850 frames of point cloud with an accuracy of ±3 cm; projection generates a minimum bounding rectangle with vertex coordinates (305695.2, 408195.8), (305710.5, 408195.8), (305710.5, 408215.3), and (305695.2, 408215.3); sampling is performed every 0.2 meters along the boundary; the index of the northern boundary point (305702.8, 408215.3) is 0.31 on the inside and 0.84 on the outside, with a difference of 53 decibels; the index of the eastern boundary point (305710.5, 408205.0) is 0.33 on the inside and 0.79 on the outside, with a difference of 46 decibels; the entire 15.3-meter northern boundary and the 7.2-meter eastern boundary with a difference greater than 40 decibels are marked.

[0096] A joint distribution feature model is generated through step 1034. The three-dimensional spatial overlay elements are: communication heat map: an orange-red surface in the tower area with an index of 0.31-0.38; mutation boundary: a red wireframe marks the north / east boundary; a gradient distribution field is constructed, and the boundary normal gradient is calculated: the X-direction change rate = adjacent point index difference / X-coordinate difference, and the Y-direction change rate = adjacent point index difference / Y-coordinate difference. At the north boundary, the X change rate is 0.021, and the Y change rate is -0.033. A blue arrow is rendered, with a length proportional to the change rate and pointing to the center of the tower; an occlusion contour is generated: the mutation boundary is stretched along the Z axis to a tower height of 82.5 meters; a translucent gray shell is constructed with a transparency of 30%; and a three-dimensional joint distribution feature is output that integrates the heat map, gradient arrows, and contour model.

[0097] The overall solution of step 103 above, by accurately quantifying the communication state parameters and dynamically binding them to spatial coordinates, constructs a three-dimensional joint cognitive model that integrates the distribution of communication capabilities and the physical obstruction characteristics of the terrain. This provides a unified spatial representation benchmark that combines electromagnetic characteristics and geographical constraints for the collaborative decision-making of drone clusters in complex terrain environments with strong interference.

[0098] 104. Based on the joint distribution characteristics, the topology of the heterogeneous cluster of drones is optimized using a distributed collaborative mechanism. During the optimization process, the relative distance constraints of the drone nodes are dynamically adjusted according to the range of line of sight obstruction caused by the terrain and the heterogeneous maneuverability of the drones.

[0099] Optionally, step 104 may specifically include the following steps: 1041. Extracting the communication attenuation gradient distribution within the terrain obstacle occlusion contour from the joint distribution feature, and calculating the communication reachable space range of the drone node affected by the terrain occlusion in combination with the current three-dimensional position of the drone node;

[0100] 1042. Determine a dynamic distance constraint threshold required for maintaining reliable communication between the UAV nodes based on the communication reachable space range and flight mode parameters of adjacent UAV nodes;

[0101] 1043. Under the distributed coordination mechanism, the UAV node broadcasts the communication reachable space range and the dynamic distance constraint threshold to neighboring nodes, receives corresponding data from the neighboring nodes, and negotiates to generate a topological connection priority list, wherein the topological connection priority list arranges node connection relationships in descending order of communication stability;

[0102] 1044. Based on the topological connection priority list, remove low-priority connections within the occlusion contour of the terrain obstacle, and add alternative communication paths across the occlusion area to complete the topological structure optimization of the heterogeneous cluster of drones.

[0103] Among them, step 1044 may specifically include the following processes: traversing the topological connection priority list, and taking the connection relationship that is ranked last in communication stability in the topological connection priority list and is located within the terrain obstacle occlusion contour as a marked connection; screening drone nodes that have a neighbor relationship with both end nodes of the marked connection outside the terrain obstacle occlusion contour as relay candidate nodes; establishing a communication connection relationship between the two end nodes and the relay candidate node, and forming an alternative path across the terrain obstacle occlusion contour based on the communication connection relationship; deleting the marked connection and adding the alternative path to complete the topological structure optimization of the drone heterogeneous cluster.

[0104] In the above scheme, the communication attenuation gradient distribution is a vector field describing the attenuation rate of the communication quality index as it changes with spatial position within an obstacle-obstructed area. The reachable communication space range is the three-dimensional airspace within which the drone node can establish stable communication at its current location, defined by the maximum communication distance and minimum beam coverage angle. The flight mode parameters are a set of parameters that characterize the drone's maneuverability, including the hovering stability coefficient, minimum turning radius, and maximum climb rate. The dynamic distance constraint threshold is the minimum and maximum distance required to ensure reliable communication between nodes. The topological connection priority list is a table of node connection relationships sorted in descending order by historical communication stability.

[0105] In this embodiment of the present application, step 1041 is first used to analyze the communication attenuation gradient arrow data in the joint distribution feature. The trajectory of the communication quality index change is tracked along the gradient direction within the obstruction contour area, and the spatial coordinate set with a unit distance attenuation value greater than -0.05 per second is recorded. The three-dimensional coordinates of the drone are used as the cone top. Combined with the current antenna beam width angle and the maximum communication distance calculation formula (maximum communication distance = (transmit power - receive sensitivity + antenna gain - path loss factor) * distance logarithm), a conical communication reachable space model with the beam centerline as the axis is generated. For example, in the canyon tower obstruction area, drone A extracts the coordinate set of the area with a attenuation of -0.08 per second from the gradient arrow field. Based on a transmit power of 27dBm, a receive sensitivity of -90dBm, an antenna gain of 10dBi, a path loss factor of 3.0, and a distance logarithm of approximately 9.7, the maximum communication distance is calculated to be 1200 meters. Combined with a 25-degree beam angle, the conical space model is generated.

[0106] Then, in step 1042, the flight mode parameters of the adjacent nodes are read. The hovering stability coefficient (i.e., the gyro attitude angle variance) is extracted for the rotorcraft, and the minimum turning radius (i.e., the aerodynamic model calculation value) is extracted for the fixed-wing aircraft. The minimum distance constraint is calculated as 1.5 times the sum of the minimum turning radii of the two nodes, and the maximum distance constraint is the minimum value of the intersection boundary of the two nodes' reachable communication space. If there is an altitude difference between the nodes, an additional pitch angle compensation distance is added. For example, if fixed-wing aircraft B has a turning radius of 50 meters and coordinates with rotorcraft C's turning radius of 5 meters, the minimum distance is 1.5 × (50 + 5) = 82.5 meters. If B's ​​maximum communication range is 1500 meters and C's maximum range is 800 meters, the maximum constraint is 800 meters. If B is 100 meters above C, an additional altitude compensation distance of 20 meters is added, resulting in the final threshold value being [82.5 meters, 820 meters].

[0107] At the same time, in step 1043, within the distributed communication framework, each node broadcasts its reachable space parameters and distance constraint threshold. After receiving neighbor data, it calculates a connection stability score with each neighbor, which is equal to 1 divided by the packet loss rate multiplied by the normalized signal strength value. After collecting all neighbor scores, a dynamic priority list is generated from high to low, sorting nodes with the same score in ascending order by node ID. For example, if drone D receives data from neighbors E / F / G: E has a 2% packet loss rate, a -75dBm signal, and a normalized value of 0.85, resulting in a score of 1 / 0.02 × 0.85 = 42.5; F has a 5% packet loss rate, a -80dBm signal, and a normalized value of 0.6, resulting in a score of 1 / 0.05 × 0.6 = 12.0; G has a 1% packet loss rate, a -70dBm signal, and a normalized value of 0.9, resulting in a score of 1 / 0.01 × 0.9 = 90.0, resulting in a priority order of G>E>F.

[0108] Finally, in step 1044, the priority list is traversed, and connections with scores below the stability threshold and located within the terrain occlusion contour are deleted. UAVs outside the occlusion contour that are also neighbors of the disconnected nodes are selected as relay nodes to construct a new sender-to-relay-to-receiver path. The topology is updated after verifying that the communication stability score of the new path is greater than 1.2 times the original path's score. For example, the connection from UAV H to UAV I, which has a score of 0.55 and is located in the hillshade area, is deleted. Node J outside the contour is selected as a relay candidate, and a new H→J→I path is created. The new path has a packet loss rate of 0.8%, a signal of -68 dBm, and a normalized value of 0.92. The score is 1 / 0.008 × 0.92 = 115.0, which is greater than 66.0 (1.2 times the original path's 0.55), completing the topology optimization.

[0109] In actual applications, in the electromagnetic interference canyon, through step 1041, the fixed-wing UAV "No. 1" coordinates are 118.755° east longitude, 32.038° north latitude, and 200 meters above sea level. The attenuation gradient field of the tower shadow area within the joint feature is analyzed at a negative 0.07 per second. Based on the transmission power of 30dBm, the receiving sensitivity of -95dBm, the antenna gain of 12dBi, and the path loss factor of 3.2, the maximum communication distance is calculated as (30-(-95)+12-3.2)*log10(distance)=1480 meters, combined with a 28-degree beam angle to generate a conical reachable space.

[0110] Through step 1042, "No. 1" and rotorcraft "No. 3" coordinate, with a hovering stability coefficient of 0.92. The turning radius is read as follows: 58 meters for the fixed-wing aircraft and 6 meters for the rotorcraft. The minimum distance is 1.5×(58+6)=96 meters. The maximum distance is 850 meters, the minimum of 1480 meters for "No. 1" and 850 meters for "No. 3". Because "No. 1" is 120 meters higher than the other aircraft, the pitch compensation distance is increased by 25 meters. The final dynamic constraint is set to [96 meters, 875 meters].

[0111] Through step 1043, "No. 1" broadcasts parameters and receives neighbor data: calculates the connection stability with "No. 3": packet loss rate 2.5%, signal -73dBm, normalized value 0.82, score = 1 / 0.025×0.82=32.8; calculates the connection stability with "No. 2": packet loss rate 7%, signal -81dBm, normalized value 0.58, score 1 / 0.07×0.58≈8.3; generates a priority list: "No. 3">"No. 2".

[0112] In step 1044, link 2 is deleted. Its score of 8.3 is less than the threshold of 10 and it is located in the tower projection area. Relay node 5, located outside the outline, is selected to establish a path from 1 to 5 to 3. The new path has a packet loss rate of 1.2% and a signal level of -69 dBm. The normalized value is 0.86, and the score (1 / 0.012 × 0.86 = 71.7) is greater than the original path's score of 32.8, which is 1.2 times the original path's score of 39.36. This completes the topology reconstruction.

[0113] The overall solution of step 104 above, by integrating terrain attenuation characteristics with maneuverability to dynamically calculate communication distance constraints, realizes topology self-optimization based on distributed priority evaluation, and builds a robust communication network that is resistant to obstruction in a strong interference and complex terrain environment, significantly improving the reliability of collaborative navigation of heterogeneous clusters under extreme conditions.

[0114] 105. Generate a collaborative navigation instruction set that adapts to terrain undulations and signal interference based on the optimized topological structure, and control the UAV nodes to perform three-dimensional track correction through the collaborative navigation instruction set to maintain the overall navigation continuity of the cluster.

[0115] Optionally, step 105 may specifically include the following steps: 1051. extracting the communication connection relationship between the drone nodes and the corresponding reference waypoints from the optimized topological structure, wherein the reference waypoints are located in the gaps between terrain obstacles and meet the dynamic distance constraint threshold;

[0116] 1052. Calculate the heading deflection angle, pitch and climb angle, and speed adjustment required for the UAV based on the relative orientation of the reference waypoint and the current three-dimensional position of the UAV, and generate individual track correction parameters.

[0117] 1053. After exchanging the individual track correction parameters through adjacent nodes under the communication path defined by the communication connection relationship, a cluster coordinated action sequence is generated to avoid synchronous maneuver conflicts.

[0118] 1054. Convert the cluster coordinated action sequence into flight control steering instructions and power output instructions, and drive the UAV to perform three-dimensional trajectory correction with coordinated changes in lift, yaw, and speed along the gap between the terrain obstacles through the flight control steering instructions and power output instructions.

[0119] In the above scheme, the reference waypoint is a three-dimensional coordinate point located in the traversable airspace between terrain obstacles, while also meeting the minimum and maximum limits of the dynamic distance constraint threshold. The individual track correction parameters are a set of three-dimensional motion instructions, including the horizontal heading deflection angle, vertical pitch climb angle, and velocity change. The swarm coordination action sequence is a collection of coordinated maneuver instructions for the swarm of drones sorted by timestamp, ensuring that adjacent nodes do not conflict with each other. The flight control steering commands are physical quantity commands that control the deflection angles of the drone's ailerons, elevator, and rudder. The power output command is a percentage command for adjusting motor speed or engine thrust.

[0120] In the embodiment of the present application, first, step 1051 is used to parse the adjacency matrix of the optimized topology and extract node pairs with valid communication connections. For each node pair, a spatial coordinate point that satisfies a dynamic distance constraint threshold, such as [100 meters, 800 meters], is searched within the terrain obstacle gap area. A reference waypoint with no collision risk is then selected using an airspace conflict detection algorithm. For example, in the connection relationship between fixed-wing node A and rotor-wing node B, a reference waypoint is selected at the canyon wall gap (305.7, 408.2, 150.6). This point is 520 meters from A and 280 meters from B, both within the constraint range.

[0121] Then, in step 1052, using the reference waypoint as the target location, the UAV's current attitude and velocity vector are combined to calculate the heading deflection angle as the difference between the target azimuth angle and the current yaw angle. The pitch climb angle is calculated as the inverse tangent of the target altitude difference and the horizontal distance. The speed adjustment is calculated as the difference between the target distance and the coordination time window. For example, if UAV C's current location is (300.2, 400.5, 120.0) and the reference waypoint is (310.6, 405.3, 150.0), the horizontal azimuth difference is arctan²(4.8, 10.4) ≈ 24.8°, requiring a deflection of 24.8°. The vertical altitude difference is 30 meters, the horizontal distance is 12.5 meters, and the pitch angle is arctan(30 / 12.5) ≈ 67.4°. The target distance is 15 meters, and the target is required to arrive in 2 seconds, so the speed adjustment is 7.5 meters / second.

[0122] At the same time, in step 1053, the node broadcasts its trajectory correction parameters to neighboring nodes on the topologically defined communication link. After receiving all neighboring parameters, a spatiotemporal conflict detection algorithm, such as the four-dimensional cylinder collision model, is used to verify maneuver conflicts. If the estimated trajectories of two nodes intersect in the same spatiotemporal domain, the action timestamp of the lower-priority node is delayed, generating a time-shifted coordinated action sequence. For example, if the space-time distance between the corrected trajectories of nodes D and E is less than the safety threshold at time T+3 seconds, E's maneuver is delayed by 0.5 seconds to eliminate the conflict.

[0123] Finally, step 1054 performs rudder command conversion: the heading angle is input into the rudder control channel and converted to rudder deflection using the proportional coefficient Kψ: Rudder deflection = heading angle × Kψ; the pitch and climb angle is input into the elevator control channel and converted to elevator deflection using the proportional coefficient Kθ: Elevator deflection = pitch and climb angle × Kθ. For fixed-wing drones, an additional aileron coordination factor is introduced to prevent sideslip. For example, if a 30° heading angle is required, Kψ = 0.8, which results in a 24° rudder deflection; if a 15° pitch angle is required, Kθ = 1.0, which results in a 15° elevator upward deflection. Power command conversion: The speed adjustment is input into the throttle control channel, and the power output increment is calculated using the thrust coefficient KT: Throttle increment = speed adjustment × KT. For multi-rotor drones, the speed difference between the motors on each axis is synchronously adjusted to achieve attitude coordination. For example, if acceleration is required to 3 m / s, KT = 5% → throttle increase by 15%. For forward flight, the front motor slows down by 10%, while the rear motor speeds up by 10%. Three-dimensional trajectory coordination: Control surface deflection and power adjustments are synchronized within a preset time window. Inertial navigation feedback on attitude angle change rate dynamically compensates for wind disturbances and electromagnetic interference. The aircraft maintains a safe distance between terrain obstacles.

[0124] In actual applications, in an urban environment with electronic interference, through step 1051, a connection is established between nodes "A" and "B" in the drone cluster optimization topology, and a safe passage between two 40-story buildings is extracted from the urban building model. The coordinates are: 118.76892° East longitude, 32.05731° North latitude, and 220 meters above sea level. This point is 650 meters away from "A", within the dynamic constraint of [120 meters, 800 meters], and 180 meters away from "B", which is greater than the minimum distance of 120 meters. The passage width of 45 meters satisfies the dual-machine parallel operation.

[0125] Through step 1052, the current position of "B" is 118.76501° East longitude and 32.05512° North latitude, with an altitude of 180 meters. The horizontal distance to the reference waypoint is [(76892-76501)²+(5731-5512)²]^(1 / 2)≈450 meters; the relative azimuth is arctan2(219,391)≈29.1°, and it needs to turn right 29.1°; the altitude difference is 40 meters, and the pitch and climb angle is arctan(40 / 450≈5.1°; speed adjustment: the current speed is 15 m / s, and the target will be reached in 3 seconds, so it needs to accelerate to 450 / 3=150 m / s. If it exceeds the limit, it will be adjusted to the maximum speed of 25 m / s.

[0126] Through step 1053, "B" broadcasts the correction parameters, "turn right 29.1° / climb 5.1° / accelerate 10 m / s" to the neighboring node; detects that the spatial distance with the trajectory of node "C" is 22 meters < the safety threshold of 30 meters at T+6 seconds, and adopts priority arbitration: "B" task priority 7 > "C" priority 5, delaying the "C" action by 0.8 seconds; generates a collaborative sequence: T+0.0 second "B" starts the maneuver, T+0.8 second "C" starts the maneuver.

[0127] Through step 1054, the command conversion is as follows: Rudder deflection = 29.1° × 0.82 = 23.9°; Elevator deflection = 5.1° × 1.05 = 5.4°; Throttle increment = 10 m / s × 4.8% = 48% → Total throttle increased to 100% + 48% = 148%. Monitoring execution: T+2 seconds, the inertial navigation feedback indicates insufficient yaw rate, and the rudder is automatically increased to 26.5°. T+3 seconds, the millimeter-wave radar detects a distance of 38 meters from the left side of the high-rise building, and the heading is fine-tuned to 31.5°. T+4 seconds, the infrared sensor identifies a high-voltage power line (at an altitude of 230 meters), and the pitch is urgently increased to 8.2° + 160% throttle. Crossing completion: T+18 seconds, the reference waypoint is reached with an error of less than 0.5 meters, and the track aligns with the centerline of the building gap, with a maximum deviation of less than 3 meters.

[0128] The overall solution of step 105 above, through three-dimensional trajectory planning guided by reference waypoints, distributed collaborative conflict resolution and precise physical control conversion, drives the drone cluster to achieve safe and coherent collaborative track correction in strong electromagnetic interference and complex terrain environments, effectively ensuring the continuous execution capability of the cluster's overall navigation mission.

[0129] The following is a complete embodiment of steps 101 to 105: Figure 2 As shown in the figure, assume that a drone swarm enters the 3-kilometer-radius interference zone of a high-voltage power station. In step 101, fixed-wing node "A" broadcasts a 128-byte detection packet at a 10Hz frequency. After receiving it, rotor-wing node "B" analyzes the signal strength and finds it to be -87dBm, with an RMS voltage of 247.49mV, a power of 1224.99mW, and a logarithmic conversion. Simultaneously, a communication interruption is determined by detecting the loss of three response packets within a 500ms window. A millimeter-wave radar scans the power station cooling tower, acquiring 15,680 frames of point cloud data. Combined with the inertial navigation attitude angles of 2.1° pitch and -0.8° roll, the relative position is calculated: horizontal azimuth angle 62.3°, pitch angle -5.7°, and distance 325 meters. A weighting model for terrain obstruction and electromagnetic interference contribution is established. Of the 23dB signal attenuation in the cooling tower projection area, terrain contributes 78%, while electromagnetic interference contributes 22%.

[0130] Through step 102, the boundary coordinates of the area with communication attenuation greater than 15 dB (X=305120-305310, Y=408050-408230) were analyzed, and DBSCAN clustering was used to identify the cooling tower shadow area. "A" and "B" were selected as the worst node pair, with a comprehensive degradation index of 0.86. The relative azimuth angle was calculated to be 48.2°, with a horizontal range of 43.2°-53.2°; the elevation angle difference was -12.5°, with a vertical range of -15.5°--9.5°. The phased array antenna at end "A" switched its main lobe to 48.2° / -12.5°, narrowing the beamwidth to 8°. End "B" synchronously pointed at 228.2° / +12.5°, and a bidirectional narrow beam reflected from the curved surface of the cooling tower established a 5.2 Mbps link. The signal strength increased from -90 dBm to -72 dBm, and the bit error rate dropped to 3e-7.

[0131] The directional link status was quantified through step 103, with 2 interruptions and a stability of 0.333. The signal strength was -72 dBm, the normalized value was 0.76, and the composite communication quality index was 0.52. The "B" coordinate was bound to 118.769° East longitude, 32.061° North latitude, and an altitude of 185 meters, generating a heat map with a dark orange center and a yellow-green 5-meter boundary. The cooling tower point cloud was projected to obtain a circular vertical surface, and the north boundary index mutation was identified, with an inner 0.38 → outer 0.91, a difference of 53 dB. A joint feature model was constructed: the mutation boundary was filled with a red-yellow gradient field, the outer thermal map was retained, and the gray translucent shell of the cooling tower was superimposed.

[0132] Step 104 extracts a gradient value of -0.09 per second in the cooling tower shadow area from the joint features. "A" calculates a maximum communication range of 1350 meters based on a transmit power of 28 dBm and an antenna gain of 14 dBi, generating a reachable space with a 25° cone angle. Based on the hovering coefficient of the rotorcraft "Peregrine-C" of 0.88 and a turning radius of 6 meters, the dynamic constraint threshold [1.5 × (58 + 6) = 96 meters, 1350 meters] is calculated. After distributed broadcasting, "A" calculates neighbor scores: "B" scores 38.6, packet loss rate 1.8%, and signal -70 dBm; "C" scores 12.1, packet loss rate 7%, and signal -82 dBm. "C"'s low-priority connection is deleted, and a new path "A → relay node D → B" is added. The new path score is 52.3, which is 1.2 times greater than the original path.

[0133] Through step 105, the reference waypoint between "A" and "B" is extracted from the optimized topology, located at the gap between the power station pipelines, with coordinates 305215.7, 408152.3, and 195.6. The current position of "B" (305000.0, 408000.0, 180.0) requires a heading deviation of 32.5° and a climb of 4.2°, with an altitude difference of 15.6 meters and a horizontal distance of 212 meters. Acceleration should be 8.3 meters per second, and the target should be reached in 5 seconds. After exchanging parameters with the neighbor, maneuver "A" is delayed by 0.6 seconds to avoid conflict. Conversion instructions are: rudder 26°, elevator 4.4°, and throttle increment 40%. During execution, the radar detects the high-voltage line at an altitude of 202 meters, triggering an emergency climb: elevator +12°, throttle 150%, and finally passing through the gap between the pipelines.

[0134] Through a fifth-order collaborative mechanism, this application achieves the precise separation of the communication impact weights of terrain obstruction and electromagnetic interference in heterogeneous clusters of drones in extreme environments coupled with strong electromagnetic interference and complex terrain, laying the foundation for anti-interference decision-making; utilizes the physical characteristics of terrain reflection to establish a highly reliable obstacle avoidance link, breaking through the limitations of line-of-sight transmission; integrates a three-dimensional model of communication capabilities and geographical constraints to provide a basis for environmental adaptive decision-making; dynamically constrains distances based on heterogeneous characteristics, and distributes and reconstructs an interruption-resistant communication network; and collaboratively corrects three-dimensional tracks guided by reference waypoints to ensure that the cluster continuously crosses narrow airspace.

[0135] Figure 3 The present invention provides a schematic diagram of a heterogeneous UAV cluster collaborative navigation system for complex terrain. Figure 3 As shown, the system includes: a collection module 31, which is used to collect communication link quality data between drone nodes in the heterogeneous drone cluster when the drone cluster flies to an area with both terrain undulations and signal interference, and synchronously obtain position information of each drone relative to terrain obstacles. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication;

[0136] a reconstruction module 32 for dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node based on the distribution of the communication attenuation area indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna;

[0137] an association module 33 for associating the connection state of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment;

[0138] an optimization module 34 for optimizing the topology of the heterogeneous cluster of drones using a distributed collaborative mechanism based on the joint distribution characteristics, and dynamically adjusting the relative distance constraints of the drone nodes according to the range of line of sight obstruction caused by terrain undulations and the heterogeneous maneuverability of the drones during the optimization process;

[0139] The correction module 35 is used to generate a collaborative navigation instruction set that adapts to terrain undulations and signal interference based on the optimized topological structure, and controls the UAV nodes to perform three-dimensional track correction through the collaborative navigation instruction set to maintain the overall navigation continuity of the cluster.

[0140] Figure 3 The UAV heterogeneous cluster collaborative navigation system for complex terrain can be implemented Figure 1 The implementation principle and technical effects of the method for cooperative navigation of a heterogeneous swarm of unmanned aerial vehicles (UAVs) in complex terrain described in the illustrated embodiment will not be elaborated on here. The specific manner in which each module and unit performs operations in the above-mentioned embodiment of a cooperative navigation system for a heterogeneous swarm of UAVs in complex terrain has been described in detail in the embodiments of the method and will not be elaborated on here.

[0141] In one possible design, Figure 3 The embodiment shown is a complex terrain oriented UAV heterogeneous cluster collaborative navigation system that can be implemented as a computing device, such as Figure 4 As shown, the computing device may include a storage component 41 and a processing component 42;

[0142] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 42 .

[0143] The processing component 42 is used for the above Figure 1 The embodiment provides a method for cooperative navigation of heterogeneous clusters of unmanned aerial vehicles (UAVs) in complex terrain.

[0144] The processing component 42 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0145] The storage component 41 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0146] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0147] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0148] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0149] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0150] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The illustrated embodiment provides a method for collaborative navigation of heterogeneous clusters of UAVs in complex terrain.

[0151] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0152] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0153] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion 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 computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0154] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for cooperative navigation of heterogeneous clusters of unmanned aerial vehicles in complex terrain, characterized by: include: When a heterogeneous UAV cluster flies to an area with both undulating terrain and signal interference, the communication link quality data between UAV nodes in the heterogeneous UAV cluster is collected, and the position information of each UAV relative to the terrain obstacles is simultaneously obtained. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication; Dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node according to the distribution of communication attenuation areas indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna; Associating the connection state of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment; Based on the joint distribution characteristics, a distributed collaborative mechanism is used to optimize the topology of the heterogeneous cluster of drones. During the optimization process, the relative distance constraints of the drone nodes are dynamically adjusted according to the range of line of sight obstruction caused by the undulating terrain and the heterogeneous maneuverability of the drones. A collaborative navigation instruction set that adapts to terrain undulations and signal interference is generated based on the optimized topological structure, and the collaborative navigation instruction set is used to control the UAV nodes to perform three-dimensional track correction to maintain the overall navigation continuity of the cluster.

2. The method according to claim 1, characterized in that The method includes dynamically reconstructing the radiation direction of the airborne antenna of the selected UAV node according to the communication attenuation area distribution indicated by the communication link quality data, and establishing a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna. The method includes: parsing the boundary coordinates of an area where the signal attenuation value exceeds a preset threshold in the communication link quality data, and identifying a communication node pair within the boundary coordinates of the area that is most severely blocked by the terrain obstacle; For the communication node pair, extract the relative azimuth and elevation angle difference between the transmitting and receiving drones, determine the horizontal angle range in which the beam main lobe needs to be deflected based on the relative azimuth, and determine the vertical angle range in which the beam main lobe needs to be raised or lowered based on the elevation angle difference; In the radiation direction of the airborne antenna of the transmitting UAV, the main lobe of the beam is switched to the intersection of the horizontal angle range and the vertical angle range, and the beam width is simultaneously narrowed to a preset narrow beam mode to focus the transmission energy on the direction of the receiving UAV; The receiving-end drone synchronously performs switching and contraction operations, so that the narrow beams of the transmitting-end and receiving-end antennas establish a directional communication link around the terrain obstacle on the side of the terrain obstacle.

3. The method according to claim 1, characterized in that The method of optimizing the topology of the heterogeneous cluster of drones using a distributed collaborative mechanism based on the joint distribution characteristics and dynamically adjusting the relative distance constraints of the drone nodes according to the range of line of sight obstruction caused by terrain undulations and the heterogeneous maneuverability of the drones during the optimization process includes: Extracting the communication attenuation gradient distribution within the terrain obstacle occlusion contour from the joint distribution feature, and calculating the communication reachable spatial range of the drone node affected by the terrain occlusion in combination with the current three-dimensional position of the drone node; Determining a dynamic distance constraint threshold required to maintain reliable communication between the UAV nodes based on the communication reachable space range and flight mode parameters of adjacent UAV nodes; Under the distributed collaborative mechanism, the UAV node broadcasts the communication reachable space range and the dynamic distance constraint threshold to neighboring nodes, receives corresponding data from the neighboring nodes, and negotiates to generate a topological connection priority list, wherein the topological connection priority list arranges the node connection relationships in descending order of communication stability; According to the topological connection priority list, low-priority connections within the occlusion contour of the terrain obstacle are removed, and alternative communication paths across the occlusion area are added to complete the topological structure optimization of the heterogeneous cluster of drones.

4. The method according to claim 3, characterized in that The method generates a collaborative navigation instruction set adapted to terrain fluctuations and signal interference based on the optimized topological structure, and controls the UAV nodes to perform three-dimensional track correction to maintain the overall navigation continuity of the cluster through the collaborative navigation instruction set, including: Extracting the communication connection relationship between the drone nodes and the corresponding reference waypoints from the optimized topology structure, wherein the reference waypoints are located in the gaps between terrain obstacles and meet the dynamic distance constraint threshold; Calculate the heading angle, pitch angle, and speed adjustment required for the UAV based on the relative orientation of the reference waypoint and the current three-dimensional position of the UAV, and generate individual track correction parameters; After exchanging the individual track correction parameters through adjacent nodes under the communication path defined by the communication connection relationship, a cluster coordinated action sequence is generated to avoid synchronous maneuver conflicts; The cluster coordinated action sequence is converted into a flight control steering quantity instruction and a power output instruction, and the flight control steering quantity instruction and the power output instruction are used to drive the UAV to perform three-dimensional track correction with coordinated changes in lift, yaw and speed along the gap between the terrain obstacles.

5. The method according to claim 1, wherein The associating the connection state of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment includes: Quantifying the connection status of the directional communication link as a communication quality index including link stability and signal strength values, wherein the link stability is calculated inversely proportional to the number of consecutive communication interruptions, and the signal strength value is obtained from the RF signal sampling value of the receiving drone; Mapping the communication quality index to the three-dimensional spatial coordinates of the corresponding drone position point to generate a communication quality spatial distribution map with the corresponding drone position point as the carrier; Marking a vertical projection plane of the terrain obstacle in the communication quality spatial distribution map, identifying a mutation boundary where the difference in the communication quality index on both sides of the vertical projection plane exceeds a preset drop, wherein the mutation boundary represents the range of the terrain blocking effect on the communication link; The three-dimensional spatial relationship between the communication quality spatial distribution map and the mutation boundary is superimposed to form a joint distribution feature that simultaneously reflects the communication capability attenuation gradient distribution of the heterogeneous cluster of drones and the occlusion contour of the terrain obstacle.

6. The method according to claim 2, characterized in that The method includes switching the main lobe direction of the beam to the intersection of the horizontal angle range and the vertical angle range in the radiation direction of the airborne antenna of the transmitting UAV, and simultaneously shrinking the beam width to a preset narrow beam mode to focus the transmission energy on the direction of the receiving UAV, including: Determining the azimuth of a three-dimensional target to which the main lobe of the beam is to be directed based on the center value of the horizontal angle range and the center value of the vertical angle range; driving the antenna radiation unit of the transmitting UAV to rotate to a horizontal deflection angle corresponding to the orientation of the three-dimensional space target, and adjusting the antenna radiation unit to a vertical pitch angle corresponding to the orientation of the three-dimensional space target after the rotation is completed; Narrowing the excitation phase distribution range of the adjusted antenna radiating element to a preset narrow beam mode, so that the electromagnetic wave energy is concentrated to cover a conical area centered on the three-dimensional target orientation through the narrowing process; Detect the radio frequency signal strength at the location of the receiving drone, and dynamically fine-tune the horizontal deflection angle and the vertical pitch angle so that the conical area completely covers the location of the receiving drone.

7. The method according to claim 3, characterized in that The method of removing low-priority connections within the terrain obstacle occlusion outline according to the topological connection priority list and adding alternative communication paths across the occlusion area to complete the topological structure optimization of the heterogeneous cluster of drones includes: Traversing the topology connection priority list, and taking the connection relationship that is ranked last in the topology connection priority list in terms of communication stability and is located within the terrain obstacle occlusion outline as a marked connection; Selecting drone nodes outside the terrain obstacle occlusion outline that have neighbor relationships with both end nodes connected to the marker as relay candidate nodes; Establishing a communication connection relationship between the two end nodes and the candidate relay node, and forming an alternative path that crosses the occlusion outline of the terrain obstacle based on the communication connection relationship; The marked connection is deleted and the alternative path is added to complete the topology optimization of the heterogeneous cluster of drones.

8. A UAV heterogeneous cluster collaborative navigation system for complex terrain, characterized by: include: A collection module is used to collect communication link quality data between drone nodes within a heterogeneous drone cluster when the cluster flies to an area with both terrain undulations and signal interference, and simultaneously obtain the position information of each drone relative to terrain obstacles. The communication link quality data reflects the dual impact of terrain obstruction and signal interference on cluster communication; a reconstruction module, configured to dynamically reconstruct the radiation direction of the airborne antenna of the selected UAV node based on the distribution of the communication attenuation area indicated by the communication link quality data, and establish a directional communication link that bypasses the terrain obstacle by switching the main lobe direction of the beam or shrinking the beam width in the radiation direction of the airborne antenna; an association module, configured to associate the connection status of the directional communication link with the location information to construct a joint distribution feature reflecting the cluster communication capability and spatial configuration under a terrain obstruction environment; an optimization module for optimizing the topology of the heterogeneous cluster of drones using a distributed collaborative mechanism based on the joint distribution characteristics, and dynamically adjusting the relative distance constraints of the drone nodes according to the range of line of sight obstruction caused by terrain undulations and the heterogeneous maneuverability of the drones during the optimization process; The correction module is used to generate a collaborative navigation instruction set that adapts to terrain undulations and signal interference based on the optimized topological structure, and to control the UAV nodes to perform three-dimensional track correction to maintain the overall navigation continuity of the cluster through the collaborative navigation instruction set.

9. A computing device, characterized in that It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a heterogeneous cluster collaborative navigation method for complex terrain of unmanned aerial vehicles as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the method for cooperative navigation of a heterogeneous cluster of unmanned aerial vehicles for complex terrain as described in any one of claims 1 to 7 is implemented.

Images