Novel unmanned aerial vehicle and application method thereof

Abstract

The invention discloses an air target capture system based on a modular multi-mode architecture and an operation method thereof, and belongs to the technical field of unmanned aerial vehicle cooperative control and intelligent interception. The system systematically solves the problems of low capture efficiency, high secondary damage risk and high large-scale deployment cost in the prior art through modular hardware integration, dynamic adaptive decision and economic optimization design. The hardware architecture of the system comprises a clamp-type near battle module, a rotor wing protection structure and an umbrella-shaped dynamic capture net system, supports rapid function switching and assembly replacement, and adapts to diversified interception requirements. The invention provides an efficient, safe and economical integrated solution for air target interception, and has wide application prospect and industrialization potential.

Description

Technical Field

[0001] The present invention belongs to the technical field of autonomous control of unmanned aerial vehicles and security equipment, and particularly relates to an unmanned aerial vehicle system with multi-modal capture capabilities and an operation method thereof, which is particularly suitable for non-lethal interception and capture of moving targets in the air. The system can effectively meet the rapid disposal requirements of various targets, including out-of-control unmanned aerial vehicles, illegal unmanned aerial vehicles, aerial mooring devices, dangerous floating objects, and bird flocks, by integrating close combat physical intervention, dynamic adhesion capture, and swarm cooperation strategies, and is applicable to multi-scenario applications such as public security law enforcement, critical facility protection, air traffic control, and ecological management. Background Art

[0002] Current disposal technologies for aerial threat targets mainly include the following types and their limitations:

[0003] Capture net launch: A capture net is launched through a ground device or an unmanned aerial vehicle. However, it has high requirements for distance, low capture success rate, and the net body is easily entangled with its own equipment due to air flow interference.

[0004] Collision and destruction: A high-speed unmanned aerial vehicle is directly used to impact the target. Although the threat can be quickly eliminated, it will cause the target to explode or debris to splash, and it is easy to trigger secondary disasters in sensitive areas such as cities.

[0005] Laser weapon: High-energy laser is used to ablate the electronic components or structures of the target, but it has serious defects: The atmospheric attenuation effect is significant, and the effective range decreases by 70%-90% in rainy, snowy, and foggy weather; Precise tracking system is required for continuous irradiation, and the damage efficiency to fast-moving targets is low; It may accidentally damage the optical sensors of legal aircraft, leading to disputes.

[0006] Microwave weapon: High-power microwave is emitted to paralyze the target circuit, but there are obvious technical bottlenecks: The action range spreads in a fan shape, making it difficult to accurately focus and easily causing accidental damage to surrounding electronic devices; The system power consumption is extremely large, and it requires a vehicle-mounted power supply and is difficult to be carried by an unmanned aerial vehicle; Secondary disasters may be triggered when the target falls.

[0007] Radio interference: Blocking GPS / remote control signals forces the target to land, but there are: Completely ineffective for fiber-optic tethered unmanned aerial vehicles (control signals are transmitted through physical cables); Large-scale frequency band interference may affect other communications and cause other losses; High-end unmanned aerial vehicles have anti-jamming frequency hopping capabilities, resulting in a high interception failure rate.

[0008] Raptor capture: Training birds such as eagles and falcons to capture small unmanned aerial vehicles, but facing: Animal behavior is uncontrollable, and it may abandon the pursuit or attack non-target objects; The training period may be as long as 2 - 3 years, and the cost of a single raptor is high; It cannot cope with sharp structures such as fiberglass rotors, causing animal casualties.

[0009] Colloid spraying: Spraying quick-drying glue on the target to restrict its movement, but the defects include: The spraying distance is short, and the escape window period for high-speed targets is short; The viscosity of the colloid decreases at low temperatures, and it fails in alpine regions; It cannot selectively adhere, and may contaminate environmental objects such as buildings and vegetation.

[0010] The prior art generally has the problem of mismatch between the disposal means and the target characteristics: It cannot cut the physical connection (such as optical fibers, cables) of the tethered target, resulting in the target continuously transmitting data; It causes excessive damage to fragile targets (such as hydrogen balloons), resulting in the loss of key evidence; After the target is captured, secondary disasters may be triggered during the fall; It lacks the ability to dispose of multiple targets, and the missed capture rate increases exponentially with the increase in the number of targets. Summary of the Invention

[0011] The present invention relates to an air target capture system based on a modular multi-modal architecture and its operation method, which is particularly suitable for efficiently, safely and economically intercepting and capturing high-speed, highly maneuverable or cluster air targets. The system systematically solves the core problems of insufficient capture efficiency, high risk of secondary damage and poor economic feasibility in the prior art by integrating multi-modal hardware components, dynamic decision-making logic and cooperative control mechanisms. In the prior art, traditional capture means lack effective coverage of the dynamic escape behavior of targets, resulting in a significant missed capture rate; collision interception or energy destruction methods are prone to trigger target explosions or debris dispersion, threatening personnel safety and environmental integrity; dedicated equipment has a single function and relies on high-cost hardware, making it difficult to achieve large-scale deployment.

[0012] In view of the above technical deficiencies, the present invention proposes an innovative solution, the core of which lies in constructing a collaborative system of a modular multi-modal hardware architecture, a dynamic adaptive capture strategy, and an economic optimization design. The hardware architecture includes a clamp-type close combat module, a rotor protection structure, and a bottom umbrella-shaped dynamic capture net system, which realizes function integration and rapid replacement through modular interfaces and adapts to diverse target types and mission scenarios. The clamp module adopts a collaborative design of dynamic clamping force control and tether cutting function, and can adjust the clamping strength and cutting parameters according to the physical properties of the target; the rotor protection structure optimizes the pore gradient of the honeycomb-shaped anti-collision cover to reduce the aerodynamic drag while ensuring the anti-impact performance; the umbrella net system is based on standardized units that can be quickly replaced, combined with environment-adaptive adhesion materials and triggering mechanisms, to improve the efficiency of non-contact capture.

[0013] The dynamic adaptive capture strategy classifies target attributes in real time and matches the optimal capture mode through multi-source sensor data fusion and machine learning algorithms. Specifically, the classification logic covers multi-dimensional detection of target volume, material, and connection attributes, and processes the edge uncertainty of irregularly shaped targets through fuzzy control theory and morphological algorithms; the mode matching combines the dual logics of physical interception and non-contact capture, and dynamically adjusts the execution parameters according to the target motion state and mission environment. In the group collaboration scenario, the system constructs a multi-layer interception system through distributed decision-making nodes and an adaptive blocking network, significantly improving the coverage ability and capture accuracy for cluster targets.

[0014] The economic optimization design reduces the operation and maintenance costs through the rapid maintenance and function switching mechanisms of modular components, and at the same time is compatible with low-cost commercial drone platforms, supporting flexible expansion from single units to large-scale swarms. The modular interface adopts a standardized protocol to ensure hardware compatibility and function scalability, and is applicable to multi-field scenarios such as public safety, airspace control, and agricultural protection. Through the synergistic effect of the above technical solutions, the present invention achieves the following technical effects: the capture efficiency is significantly improved, and the missed capture rate is reduced below the theoretical threshold; the risk of secondary damage is effectively controlled through optimizing the proportion of non-contact capture modes; the economic enhancement is reflected in the substantial reduction of maintenance costs and the feasibility of large-scale deployment. Brief Description of the Drawings

[0015] The following further detailed description will be made in conjunction with the drawings and specific embodiments.

[0016] Figure 1 It is a structural block diagram of the aerial target capture system.

[0017] Figure 2 It is a schematic diagram of the drone structure (annotation: 1. Clamp-type capture unit, 2. Honeycomb anti-collision cover, 3. Umbrella-shaped dynamic capture net)

[0018] Figure 3 It is a flow chart of the air target interception mission

[0019] Figure 4 It is a refined flow chart of the single aircraft execution process

[0020] Figure 5 It is a schematic diagram of the UAV structure (Note: 1. Pincer capture unit, 2. Honeycomb anti-collision cover, 3. Mechanical deployment structure, 4. Umbrella-shaped dynamic capture net) Specific implementation manners

[0021] The hardware architecture of the air target capture system consists of a fuselage main body, a top close combat module, a bottom umbrella-shaped dynamic capture net system and a collaborative decision-making module. Each component realizes seamless integration and function expansion through a standardized module interface. The fuselage main body adopts a lightweight modular framework design, integrating a multi-rotor propulsion system and a replaceable energy module, and ensuring stability and load adaptability in dynamic tasks through theoretical optimization of lift redundancy. The flight control module is built with a multi-sensor fusion unit, which collects and fuses multi-dimensional data such as attitude angle, acceleration, barometric altitude and environmental wind speed in real time, and realizes closed-loop adjustment of flight attitude and anti-interference ability through non-linear control algorithms. The power system is compatible with multiple energy types, including but not limited to lithium-ion batteries, lead-acid batteries or other power supply modules, and realizes quick replacement through a standardized energy interface to adapt to different task requirements and diverse operating environments.

[0022] It is executed in the mission startup phase and includes the following core operations: The UAV system supports initialization and self-check processes. The multi-sensor fusion unit performs zero calibration on the attitude angle sensor, accelerometer, barometric altimeter and environmental wind speed sensor to ensure that the data acquisition error is lower than the theoretical threshold; then a comprehensive test is carried out on the actuators. The pincer capture unit performs a full-stroke test of the opening and closing action to verify the positioning accuracy of the servo motor and the output torque of the hydraulic auxiliary unit. The umbrella-shaped dynamic capture net system simulates the unfolding and folding actions to detect the response time of the electromagnetic latch and the synchronization of the pneumatic-mechanical composite deployment mechanism; finally, the end-to-end handshake protocol verification is carried out through a redundant communication link to ensure that the multi-band adaptive switching mechanism and the high-strength encryption transmission function are normal, and the packet loss rate and bit error rate meet the theoretical standards. The above process can be triggered or monitored through the control interface, and at the same time, an automated execution option is reserved to adapt to different task requirements.

[0023] The top close - combat module is connected to the fuselage main body through a quick - release interface and includes two core components: a clamp - type capture unit and a honeycomb - shaped anti - collision cover. The clamp - type capture unit is designed as an opening - and - closing mechanism. The inner side of the clamp head is integrated with multiple layers of friction - enhanced surfaces, including a micro - structure texture layer and a flexible adsorption layer, which improve the clamping stability for irregular targets and smooth materials through the theoretical optimization of surface roughness and friction coefficient. The drive mechanism adopts a dual - mode control logic, integrating a high - precision servo motor and a hydraulic auxiliary unit to achieve dynamic adjustment and rapid response of the clamping force, adapting to the capture requirements of targets with different strengths. The shear component is embedded in the rear end of the clamp head closing surface and uses a high - hardness composite coating process. Through the collaborative design of the edge geometry and material properties, it ensures efficient cutting of the mooring object while reducing energy loss. The honeycomb - shaped anti - collision cover is based on the theoretical models of aerodynamic simulation and anti - impact performance, optimizing the aperture gradient distribution and porosity parameters, and combining three - dimensional woven composite materials to improve the energy absorption efficiency and structural integrity. The modular quick - release interface supports rapid replacement to cope with structural damage in special mission environments.

[0024] The target detection and classification process is achieved through multi - source sensor fusion: The system uses radar, optical imaging, and multi - spectral detectors to collect target data, constructs a three - dimensional point cloud model of the target, and performs real - time classification. Specifically, for volume classification, the convex hull algorithm is used to calculate the minimum bounding volume of the target, and the fuzzy control theory is combined to handle the edge uncertainty of irregular - shaped targets, generating volume classification labels (small, medium, large); for material detection, the multi - spectral imaging system analyzes the reflection and transmission characteristics of the target surface to identify the material type (metal, composite material, biological material, etc.) and detect linear connection structures (such as cables, optical fibers); for motion state analysis, the Kalman filter and machine learning model are used to predict the target motion trajectory, calculate the speed, acceleration, and escape probability, generating a dynamic threat level assessment. The above classification results can be verified or corrected through the human - machine interaction interface, and at the same time, it supports the full - automatic mode to adapt to emergency mission scenarios.

[0025] The bottom umbrella-shaped dynamic capture net system is connected to the bottom of the fuselage through an interface and consists of a foldable tensile fiber net, a pressure-sensitive trigger unit, and a pneumatic-mechanical composite deployment mechanism. The foldable tensile fiber net is designed in an umbrella shape and forms a preset geometric configuration after deployment. Different mesh sizes can be replaced to adapt to the capture requirements of targets with different volumes. The surface of the net body is coated with a viscous material, and different viscous materials are used to adapt to changes in temperature, humidity, and air pressure, ensuring effective adhesion and target restraint under extreme climate conditions. The pressure-sensitive trigger unit adopts a sealed capsule structure filled with shear-thickening fluid. The feedback adjustment algorithm selects different rupture thresholds to adapt to the continuous changes in the target's struggling strength and motion acceleration, avoiding premature triggering or delayed response. The pneumatic-mechanical composite deployment mechanism triggers the deployment action in stages based on the theoretical model of the target's motion speed or activates the compressed gas injection assistance. The deployment time is negatively correlated with the target speed, ensuring the timeliness and spatial coverage accuracy of the interception action.

[0026] Capture mode matching and plan generation are dynamically executed according to the target classification result: for small and medium-sized targets or high-risk scenarios, the system activates the non-contact capture mode, predicts the spatio-temporal coordinates of the interception point based on the target's motion trajectory, optimizes the deployment timing and coverage range of the umbrella net, and the pressure-sensitive trigger unit adjusts the rupture threshold in real time to ensure that the net body completes adhesion and restraint at the moment of target contact; for large targets or moored targets, the physical interception mode is enabled, and the clamping force of the clamp-type capture unit is dynamically adjusted according to the target's material and strength. A low clamping force (theoretical lower limit value) is applied during the initial embedding stage to avoid structural damage, and then it is linearly increased to the traction threshold at the theoretical rate while monitoring the power load and attitude stability of the UAV (tilt tolerance ≤ theoretical threshold). When a mooring is detected manually, the cutting component starts the cutting stroke, and the cutting speed matches the material hardness. High-resolution secondary scanning is used to confirm the complete separation of the mooring, and the residual detection threshold is set as the theoretical diameter ratio. The priority of the plan is dynamically adjusted based on the real-time environmental risk assessment (wind speed, electromagnetic noise, visibility) to ensure that the execution logic is highly adapted to the task scenario. The user can manually adjust the mode priority or intervene in the execution parameters through the control interface while retaining the fully automatic decision-making chain to adapt to special scenarios.

[0027] The collaborative decision-making module adopts a distributed computing node architecture. Each node independently processes tasks such as target tracking, path planning, and communication encryption, and realizes global policy synchronization and fault tolerance control through a consensus algorithm. The communication protocol supports a multi-band adaptive switching mechanism, combined with a high-strength encryption standard and redundant link design, to ensure signal stability and data security in a complex electromagnetic environment. The dynamic capture strategy realizes adaptive adjustment through a closed-loop feedback mechanism, including target classification, pattern matching, capture execution, and swarm cooperation control. The target classification constructs a three-dimensional point cloud model of the target based on multi-source sensor data fusion, combines fuzzy control theory to handle the edge uncertainty of irregularly shaped targets, and improves the classification accuracy through membership functions and confidence evaluations. The pattern matching predicts the target motion trajectory through machine learning algorithms and generates multi-modal interception plans, and the priority is dynamically adjusted based on real-time environmental risk assessments, including multi-dimensional parameters such as wind speed interference, electromagnetic noise, and visibility.

[0028] The swarm cooperation capture process is realized through dynamic role assignment and path optimization: The role assignment comprehensively evaluates the remaining energy, positioning accuracy, and payload status of the UAVs based on multi-objective optimization theory, generates a priority weight matrix to select the main interceptor and auxiliary blockade aircraft; The auxiliary blockade aircraft is deployed according to the topological network model to construct a dynamic blockade sector. The blockade angle is positively correlated with the target escape probability, forming a multi-layer adaptive blockade network. The path optimization introduces a game theory framework to balance the energy consumption of the group. The UAVs distribute the task load through distributed negotiation and converge to the Nash equilibrium state; Conflict avoidance uses a spatio-temporal four-dimensional planning algorithm (3D space + time dimension) to generate the optimal path to ensure that the safety distance is always greater than the theoretical threshold (calculated based on the maximum relative speed and response time). The global policy synchronization realizes task assignment and status synchronization through a consensus algorithm (such as the Raft protocol) to ensure the consistency of swarm behavior. Users can monitor the swarm status in real time through the command terminal or intervene in key decision-making nodes, and at the same time support a fully automatic cooperation mode to adapt to high-density task requirements.

[0029] It includes capture confirmation, target handover, and maintenance processes: The system verifies the target capture status through multi-sensor fusion (such as clamping stability, integrity of the net body restraint), and after confirmation, executes a return flight or hands over to the designated disposal unit; The system reset process clears the historical task data (overwriting and writing the theoretical number of times), restores the default control parameters (PID gain, clamping force threshold, etc.), and recalibrates the sensor reference values (following the international standard protocol) to ensure the consistency of the initial state of the next task; Regular maintenance detects the wear degree of the friction surface of the clamp head (roughness ≤ theoretical threshold), evaluates the sharpness of the shear edge (blade angle tolerance ≤ theoretical value), and scans the structural integrity of the honeycomb-shaped anti-collision cover (crack detection sensitivity ≤ theoretical threshold), and replaces modular components with excessive performance decay to ensure the long-term stability of the system.

[0030] Specific implementation examples are asFigure 4 The single - unit capture process shown is as follows:

[0031] t1 (Takeoff and initial state calibration): After the system starts, the drone executes the autonomous takeoff procedure, synchronously activates the multi - sensor fusion unit, and completes the real - time calibration of attitude angle, acceleration, barometric altitude, and ambient wind speed. The flight control module adjusts the lift redundancy based on the non - linear control algorithm to ensure the fuselage stability and load adaptability. The self - inspection process covers the opening and closing accuracy of the clamp - type capture unit, the deployment synchronization of the umbrella - shaped net system, and the communication link encryption function, ensuring that the initialization error of each module is lower than the theoretical threshold.

[0032] t2 (Tether detection and cutting): After the target is locked, the optical imaging and multi - spectral sensors cooperate to scan the target surface to identify whether there are tethers (such as cables, optical fibers). If a tether structure is detected, the system preferentially activates the shearing component: The high - hardness composite blade embedded in the rear end of the clamp - type module matches the cutting speed according to the material hardness, and simultaneously verifies the complete separation of the tether through high - resolution secondary scanning, with the residual diameter ratio strictly controlled below the theoretical threshold. If no tether is detected, it directly enters the capture stage.

[0033] t3 (Target classification and capture execution): Based on the real - time three - dimensional point cloud model and fuzzy control classification results: If it is a large target: Activate the clamp - type close - combat module, and the dynamic clamping force control algorithm adjusts the initial clamping force (lower limit value) according to the target material strength (metal / composite material), and then increases it to the traction threshold at a linear rate while monitoring the drone tilt stability (tolerance ≤ 5°). During the clamping process, the multi - layer friction - enhanced surface (micro - texture + flexible adsorption layer) ensures the grasping stability of the irregular target. If it is a small or medium - sized target: Enable the bottom umbrella - shaped dynamic capture net system, which can be mechanically deployed / pneumatically deployed, predict the target trajectory based on the Kalman filter, and optimize the space - time coordinates of the umbrella net deployment. The pressure - sensitive trigger unit dynamically adjusts the rupture threshold according to the target struggling intensity, and the viscous coating (adapting to temperature and humidity changes) ensures non - contact binding when the net body touches, and the mesh size is preset according to the target volume (error ≤ 2cm).

[0034] t4 (Target handover and return flight): After the capture is completed, the multi - sensor fusion verifies the clamping pressure (≥80% of the theoretical threshold) or the net body tension (≥90% of the rated value) to confirm the integrity of the target binding. The drone executes the pre - planned path and transports the target to the designated disposal unit (such as a ground recovery platform or an air transfer station). During the handover process, the communication module synchronizes the target status and coordinates in real - time through redundant multi - band links. If sudden interference occurs (such as electromagnetic noise ≥ 120dB), it automatically switches to the anti - interference frequency band and starts the encrypted re - transmission protocol. After the task is completed, the drone returns to the base station, triggers the system reset process, clears the historical data, and restores the default parameters to prepare for the next task initialization.

[0035] Meanwhile, the above process supports full-automatic mode operation, allowing manual intervention at key nodes (such as switching the capture mode and adjusting the handover path) through the control interface to ensure flexibility and reliability in complex scenarios.

Claims

1. A drone system with multi-modal capture capability, characterized in that: include: The fuselage body integrates the flight control module, power module and multi-rotor propulsion assembly; The top melee module consists of an openable and closable clamp-type capture unit, a mooring shear assembly, and a honeycomb crash shield; The clamp-type capture unit includes a symmetrically distributed clamp head structure, the inner side of which is provided with an adaptive pressure sensing surface, which can dynamically adjust the clamping force range; The shearing assembly is integrated at the rear end of the occlusal surface of the pliers head and comprises a retractable high-hardness shearing blade, the cutting direction of which is parallel to the closing plane of the pliers head; The honeycomb anti-collision cover covers the rotor assembly and is made of porous lightweight composite material, and has the functions of impact resistance and airflow guidance; Bottom umbrella dynamic capture net system, including: A folded tensile fiber mesh with a sticky polymer coating on the surface of the mesh; The pressure-sensitive trigger unit is composed of sealed capsules distributed at the nodes of the mesh body. The capsules are filled with viscous fluid and rupture and release when the external pressure exceeds a preset threshold. A pneumatic-mechanical composite deployment mechanism is used to deploy the net into a preset geometric shape within 0.1-0.8 seconds after the trigger command is issued; The collaborative decision-making module is used to generate multi-machine collaborative paths and capture strategies according to the dynamic behavior of the target in swarm mode.

2. The drone system according to claim 1, characterized in that: The startup logic of the clamp capture unit is:

1. When the target is classified as a large untethered target, the clamp head is controlled to partially unfold and embed into the target surface, and the target is moved by the power of the drone; 2. When a physical connection structure is detected on the target, the clamp head is closed to clamp the tether and the shear assembly is triggered to cut the connection; 3. During the capture of small and medium-sized untethered targets, the clamp capture unit remains inactive.

3. The drone system according to claim 1, characterized in that: The activation of the shear component must meet the following conditions:

1. The clamping pressure of the clamp head is ≥200N and the clamping time is ≥0.2 seconds; 2. Confirm that the material of the tether is shearable by multispectral sensor or laser reflection analysis.

4. The drone system according to claim 1, characterized in that: The structural parameters of the honeycomb anti-collision cover meet the following requirements: The porosity is between 40% and 60%, the pore size distribution is 3-8mm, and the overall thickness is 1 / 15-1 / 10 of the rotor diameter; The vertical distance between the anti-collision cover and the rotor assembly is ≥10mm to avoid airflow interference.

5. The UAV system according to claim 1, characterized in that: The collaborative strategy of the collaborative decision-making module includes: Dynamic allocation of master and slave roles: Dynamically designate the master interceptor and the auxiliary blockade according to the real-time distance between the drone and the target, the remaining power, and the load status; Path prediction blocking: The auxiliary blocking machine takes the current position of the target as the origin and is distributed in a fan-shaped manner on both sides of its movement trend direction. The blocking angle θ satisfies: θ=arctan(vtarget / vdrone)+Δθ Among them, vtarget is the target moving speed, vdrone is the maximum speed of the drone, and Δθ is the safety redundancy angle (5°-15°).

6. An operating method of the UAV system according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. Target attribute determination:

1. Analyze the target volume through radar reflection cross section or optical imaging and classify it into large targets or small and medium-sized targets; 2. Detect whether the target has a tethered connection through multi-spectral scanning; S2. Capture mode execution:

1. For non-tethered targets:

1. If it is a small or medium-sized target, calculate the target's motion trajectory, control the drone to move to the intersection of the trajectories, and trigger the umbrella-shaped dynamic capture net; 2. If it is a large target, the clamp-type capture unit is controlled to partially unfold and embed into the target surface, and the target is towed to the designated area by the power of the drone; 2. For the target of mooring:

1. Control the clamp-type capture unit to close and clamp the tethered object, triggering the shear assembly to cut the connection; 2. After the cut-off is completed, re-determine the target attributes and execute the corresponding capture process; S3. Swarm Collaboration Enhancement (Optional):

1. When the single-machine capture fails or the target triggers escape behavior, the collaborative decision-making module is activated; 2. While the main interceptor is executing step S2, the auxiliary blocking aircraft continues to compress the target's moving space until the target enters the effective capture area of ​​the main interceptor. At the same time, the main interceptor and auxiliary blocking aircraft are dynamically designated based on the real-time distance between the drone and the target, the remaining power and the load status.

7. The operating method according to claim 6, characterized in that: The calculation formula of the trajectory intersection point in step S2.1 is: P_intercept = P_target + v_target × t_delay + ½ a_target × t_delay² Among them, P_target is the current position of the target, v_target is the velocity vector, a_target is the acceleration vector, and t_delay is the system response delay (including mesh expansion time + adhesion effective time).

8. The operating method according to claim 6, characterized in that: The determination conditions of the escape behavior in step S3.1 include: The angle between the target velocity vector and the drone pointing vector continues to expand, and the expansion rate is ≥10° / second; Can be determined by on-site operators.

9. The operating method according to claim 6, characterized in that: The priority strategy for tether cutting described in step S2.2.1 is determined by the operator.

10. Application of the UAV system according to claim 1 in events such as interception of targets invading a no-fly zone, recovery of high-altitude tethered equipment, or removal of animals from dangerous areas.

Images