Autonomous inspection and charging nest system based on unmanned aerial vehicle ad hoc network

Abstract

The invention relates to the technical field of unmanned aerial vehicle cluster energy management, and discloses an unmanned aerial vehicle ad hoc network-based autonomous inspection and charging nest system, which comprises the following steps that: each unmanned aerial vehicle in a cluster calculates and broadcasts a time dynamic scalar capable of being freely controlled in real time, and broadcasts a task and an energy opportunity packet when the scalar of a certain unmanned aerial vehicle is lower than a threshold value; according to the invention, through a distributed decision-making mechanism driven by an energy potential difference, cluster energy is enabled to dynamically flow according to a minimum resistance path like liquid, internal energy consumption caused by a fixed return path in traditional inspection is avoided, and the reliability of the unmanned aerial vehicle is improved. And meanwhile, by combining dual-channel communication arbitration and a mechanical locking unit, reliable energy cooperative transmission can still be maintained in a complex environment.

Description

Technical Field

[0001] This invention relates to an autonomous inspection and charging nest system based on a drone self-organizing network, belonging to the field of drone swarm energy management technology. Background Technology

[0002] Current requirements mandate that drones interrupt their missions and return to recharge when their remaining energy falls below a safe threshold. This design inherently binds energy replenishment to fixed physical nodes. As the inspection range expands to complex terrain or large-span scenarios, this model gradually exposes systemic limitations: inspection paths are forced to compromise due to return-to-home requirements, a large amount of flight resources are consumed in non-operational round trips, and the redundant energy from return-to-home cannot be contributed to mission execution; it is difficult to promptly dispatch drones on their return journey for sudden missions, and the swarm response capability is physically fragmented; strong electromagnetic interference or airflow disturbances can easily lead to charging interruptions. Existing solutions improve stability by overlaying sensors and algorithms, but this exacerbates the resource burden on edge devices.

[0003] Although some studies have attempted to optimize the layout of charging stations or introduce mobile charging platforms, they have not yet broken through the underlying logic of static binding between energy and tasks. Furthermore, the surge in hardware costs and increased communication load have led to secondary conflicts. Therefore, how to achieve self-organized reorganization of cluster energy and tasks in a dynamic environment, thereby eliminating return path dependence and ensuring robust execution under complex operating conditions, has become the technical problem to be solved by this invention. Summary of the Invention

[0004] This invention provides an autonomous inspection and charging nesting system based on UAV self-organizing networks. Its main purpose is to solve the problems of cluster energy stagnation and task response delay caused by fixed charging paradigms.

[0005] To achieve the above objectives, the present invention provides an autonomous inspection and charging nesting system based on a drone self-organizing network, characterized in that it includes: The energy potential assessment unit, configured in each drone in the cluster, is used to calculate and output a dynamic state scalar representing the available time in real time based on the drone's current remaining energy, drone location information, and pre-stored charging nest location information. The dynamic state scalar is the result of dividing the difference between the current remaining energy and the energy required to return to the nearest charging nest by the average cruise power consumption. The task and energy opportunity generation unit is configured in any of the first UAVs in the cluster. When the dynamic state scalar of the first UAV is lower than a preset safety threshold, it generates and broadcasts an opportunity packet containing the task information to be performed by the first UAV and the remaining flight time information to the ad hoc network. The distributed decision unit, configured in any second UAV in the ad hoc network, is used to receive opportunity packets and calculate the time cost required for the second UAV to respond to the opportunity packet based on its own state and opportunity packet information. The time cost is the sum of the time required for the second UAV to fly to the first UAV for energy replenishment and fly away, minus the dynamic state scalar of the second UAV before responding. The collaborative execution unit is configured in the second drone with the lowest time cost in the ad hoc network. It is used to respond to the first drone, drive the second drone to move to the vicinity of the first drone for non-contact wireless inductive energy replenishment, and take over the tasks to be performed by the first drone.

[0006] Preferred, dynamic state scalar The calculation method is as follows: ,in, This indicates the drone's current remaining energy. This indicates the energy required for the drone to return to its nearest charging base. This indicates the average cruise power consumption of the drone.

[0007] Preferably, the system further includes: a dual-channel communication arbitration unit, configured on each UAV, for setting up a high-bandwidth communication channel and a narrowband heartbeat communication channel; the dual-channel communication arbitration unit monitors the communication quality of the high-bandwidth communication channel, and when the communication quality of the high-bandwidth communication channel is lower than a preset threshold, it suspends the broadcast of opportunity packets and switches to the narrowband heartbeat communication channel, transmitting only heartbeat packets containing the UAV's identity identifier and a quantized dynamic status scalar level, thereby achieving basic energy replenishment and task distribution in a communication resource-constrained mode.

[0008] Preferably, the dynamic state scalar levels include a safety level, a critical level, and a danger level; in a communication resource-constrained mode, a drone with a danger level is configured to move to the nearest drone among the drones in its vicinity that meet the safety level criteria and perform a basic energy replenishment.

[0009] Preferably, the system further includes: an energy transfer physical locking unit configured on each drone, used to perform electronic locking of the physical grappling hook and anchor point during non-contact wireless inductive energy replenishment, so as to ensure successful physical connection when the two drones approach each other; after confirming successful physical connection, the energy transfer physical locking unit authorizes the wireless inductive charging module in the cooperative execution unit to transfer energy at maximum power.

[0010] Preferably, if the energy transfer physical locking unit fails to lock after multiple attempts, it rejects the energy replenishment mission and instructs the first UAV to perform a return or local landing procedure, while broadcasting the energy transfer failure event to the ad hoc network.

[0011] Preferably, the system further includes: a micro-vibration adaptive compensation unit, configured on the first UAV, used to actively send phase detection pulses using its communication module during the energy replenishment process, and to determine the relative vibration parameters between the two UAVs based on the received carrier signal phase difference sequence, wherein the relative vibration parameters are the determined main frequency and amplitude; the micro-vibration adaptive compensation unit adjusts the rotational speed of the rotor of the first UAV based on the relative vibration parameters to generate a reverse torque, thereby suppressing the relative vibration.

[0012] Preferably, the micro-vibration adaptive compensation unit is also configured in the second UAV to adjust the hovering height of the second UAV according to the relative vibration parameters, so as to ensure that the average coupling distance between the coils of the two UAVs remains at the optimal value under vibration conditions.

[0013] Preferably, the system further includes: an energy flow residual self-diagnostic unit, configured on the drone performing energy replenishment, for real-time monitoring of the output current ripple during the contactless wireless inductive charging process; when the output current ripple exceeds a preset alarm threshold, the energy flow residual self-diagnostic unit determines that the communication channel quality of the self-organizing network has deteriorated and triggers a preset communication link protection or switching strategy.

[0014] Preferably, after taking over the task, the collaborative execution unit is further configured to subcontract part or all of the task to a third UAV in the ad hoc network with a higher dynamic state scalar and a geographical location that better meets the task requirements, based on its own dynamic state scalar and geographical location.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. By converting the energy state of each UAV into a unified dynamic scalar, the swarm system spontaneously forms an energy potential gradient. When the available time of a node is lower than a threshold, its task and remaining energy are automatically converted into an opportunity package, triggering distributed decision-making by surrounding nodes based on opportunity cost. This mechanism enables energy and tasks to flow in the swarm along the path of least resistance, thereby reducing energy consumption caused by fixed return paths and improving the overall energy utilization efficiency of the system.

[0016] 2. Narrowband heartbeat channels maintain basic energy scheduling when communication deteriorates, while the physical locking of mechanical grapples improves connection reliability during energy transmission. The combination of the two enables the system to maintain core functions through a degraded but not failed operating mode even under strong electromagnetic interference or adverse airflow conditions, avoiding systemic collapse caused by the breakage of a single technology chain.

[0017] 3. The motor counteracts the main vibration frequency through the reverse torque of the rotor, and the power supply dynamically adjusts the hovering height to maintain the optimal coupling distance. This vibration suppression mechanism based on the physical layer signal of the communication transforms the wireless charging efficiency from being passively affected by environmental stability to being able to actively suppress and compensate for dynamic disturbances. When the charging current ripple increases abnormally, the system automatically predicts the degradation of communication quality and switches the transmission channel in advance, upgrading the communication link protection mechanism from passive response to active prevention. This cross-domain perception and execution multi-path collaboration confines local faults to a single link and prevents them from spreading, significantly improving the survivability of the cluster in unpredictable scenarios. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the task coordination and energy replenishment process in the autonomous inspection and charging nest system of the present invention; Figure 2 This is a comparison chart of the relationship between UAV response distance and time cost under different terrain risk levels according to the present invention; Figure 3 This is a flowchart of the state level identification and basic energy replenishment based on the heartbeat channel of the present invention.

[0019] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It should be noted that, without conflict, the embodiments and features in the embodiments of this application can be combined with each other, and all other embodiments derived from the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0021] This application provides an autonomous inspection and charging nest system based on a drone self-organizing network. The system is built on a drone swarm, and its core operating mechanism is a distributed energy and task collaborative scheduling procedure, which mainly includes four logical modules: an energy potential assessment unit, a task and energy opportunity generation unit, a distributed decision-making unit, and a collaborative execution unit. They work together to transform the energy resources in the swarm into a dynamically transferable and reconfigurable asset, rather than being statically bound to a single drone or a charging facility with a fixed geographical location.

[0022] In a typical inspection mission scenario, the system's operation begins with the continuous operation of the energy potential assessment unit deployed within each drone in the cluster. To avoid the limitations of traditional solutions that rely solely on the remaining battery percentage and fail to reflect the actual mission endurance, the energy potential assessment unit is configured to base its operation on the drone's current remaining energy. The system uses the drone's real-time location information and the geographical location information of all charging stations pre-stored in the onboard database to calculate and output a dynamic state scalar representing its available free time. The specific calculation procedure for this scalar follows the formula below: ,in This represents the minimum energy required to return to the nearest charging nest, calculated based on factors such as the distance from the current location to the nearest charging nest and current environmental wind resistance. This represents the average cruise power consumption of this drone model under standard cruise conditions, either pre-calibrated or calculated from historical data. Thus, the dynamic state scalar... It is not an isolated energy value, but a comprehensive indicator that incorporates geographical and temporal dimensions. It precisely quantifies the time window during which drones can perform additional tasks or provide external support while ensuring a safe return, thereby establishing a unified and quantifiable energy potential gradient for the entire cluster.

[0023] When the dynamic state scalar calculated by the energy potential assessment unit of any first UAV in the cluster Once the drone descends and reaches a preset safety threshold, such as 120 seconds, the task and energy opportunity generation unit is triggered. This aims to prevent the drone from entering an emergency state requiring immediate return, instead preemptively transforming potential energy needs into an opportunity for network collaboration. The unit then generates a standardized data structure, the opportunity packet, and broadcasts it to neighboring drones within its communication range using the ad hoc network's broadcast channel. The data fields of this opportunity packet are precisely defined to include at least three core pieces of information: the unique identifier of the initiating drone, the task information packet to be handed over, and remaining flight time information accurate to the second. In this way, the potential mission failure risk of a single node about to fall into an energy predicament is proactively transformed into a publicly available and accessible opportunity for the entire distributed network. The collaborative request for bidding; upon receiving a broadcast opportunity packet, any second UAV in the ad hoc network immediately initiates a response evaluation procedure within its built-in distributed decision-making unit. This aims to complete the collaboration with the lowest total system cost, rather than simply selecting the nearest UAV. To achieve this goal, the unit performs a time cost calculation based on the second UAV's own real-time state and the information in the opportunity packet. This calculation procedure is defined as follows: First, predict and accumulate the total time cost required to perform this support mission, which includes the time to fly from the current location to the first UAV's location, the estimated time to perform non-contact wireless inductive power replenishment, and the time required to safely fly away to avoid flight conflicts. Then, subtract the second UAV's own dynamic state scalar before responding from this sum. The result is the time cost of this response. The logic behind this design is that a drone with abundant energy reserves has a lower intrinsic cost of participating in the rescue. By treating its own scalar as a negative cost, the system can incentivize the drone with the most abundant energy and the most suitable geographical location to respond. Ultimately, among all the drones that have received the opportunity packet and completed the calculation, the drone with the lowest calculated time cost will gain the right to execute this collaborative mission.

[0024] The second UAV, having gained execution authority, immediately activates its collaborative execution unit. This unit completes the closed loop from decision-making to physical execution. It first drives the second UAV to a safe distance from the first UAV along a pre-planned path, then performs a series of precise non-contact wireless inductive energy replenishment actions. After energy transfer, it takes over the tasks defined in the opportunity packet via the communication link and continues the inspection. Furthermore, to maximize system efficiency, the collaborative execution unit is configured to perform secondary task subpacketization after taking over a task. Based on its dynamic state scalar after taking over the task and its current geographical location, it determines whether it's necessary to subpackage part or all of the taken-over task as opportunity packets to a third UAV in the ad hoc network with a higher dynamic state scalar and a geographical location more suitable for subsequent tasks, thus achieving optimal task drift within the cluster. To cope with the complex electromagnetic and airflow environment disturbances during the inspection process, this system also integrates… This forms a triple fault-tolerant system of communication, control, and energy. At the communication level, each UAV is equipped with a dual-channel communication arbitration unit. This unit sets up and simultaneously monitors a high-bandwidth communication channel and a narrowband heartbeat communication channel. Under normal operating conditions, complex information such as opportunity packets are transmitted through the high-bandwidth channel. When the unit detects that the communication quality of the high-bandwidth channel is lower than a preset threshold, it will suspend the broadcast of opportunity packets and automatically switch to the narrowband heartbeat communication channel. In this mode, the UAV only transmits heartbeat packets containing its own identity and a quantified dynamic state scalar level. The levels are divided into safe level, critical level, and dangerous level. At this time, the UAV in the dangerous level is programmed to autonomously move to the closest UAV in its communication range that is determined to be in the safe level to perform basic energy replenishment. This mechanism ensures that even in harsh environments where communication resources are severely limited, the basic energy coordination and survivability of the cluster can be maintained.

[0025] At the physical control level, to address the relative displacement and connection instability issues caused by airflow disturbances during wireless charging between two drones in the air, each drone is equipped with an energy transfer physical locking unit and a micro-vibration adaptive compensation unit. Before entering the energy replenishment phase, the energy transfer physical locking unit is responsible for performing the electronic locking operation between the physical grappling hook and the corresponding anchor point. This mechanical structure ensures a reliable physical connection between the two drones after they approach each other. Only after the unit confirms a successful physical lock will it send an authorization signal to the wireless inductive charging module, allowing it to transfer energy at maximum power. If the lock fails after multiple attempts, such as three times, the unit rejects the energy replenishment task and instructs the first drone to execute a preset return-to-home or on-site safe landing procedure, while simultaneously broadcasting the energy transfer failure event to the ad hoc network. Meanwhile, the micro-vibration adaptive compensation unit continues to operate during the locking period. To suppress the impact of minor vibrations on charging efficiency, the communication module of the first drone, acting as the power recipient, actively sends phase detection pulses to the power supplier drone. Based on the phase difference sequence of the received return carrier signal, it uses a Kalman filter algorithm to invert the relative vibration main frequency and amplitude between the two drones. Subsequently, based on these vibration parameters, the unit fine-tunes the differential rotation speed of its own rotor in real time to generate a reverse compensation torque, thereby actively suppressing relative vibration. Meanwhile, the micro-vibration adaptive compensation unit of the second drone, acting as the power supplier, dynamically fine-tunes its hovering height based on the same relative vibration parameters to ensure that, under the condition of residual vibration, the average coupling distance of the wireless charging coils on the two drones always remains near the pre-calibrated optimal value. This hardware-software combined mechanism transforms the success rate of wireless charging from passively relying on environmental stability to actively being immune to dynamic disturbances.

[0026] At the level of in-depth diagnostics of energy transmission, the system also includes an energy flow residual self-diagnostic unit configured on the drone performing energy replenishment. Given that the communication quality of the ad hoc network indirectly affects the real-time performance of wireless charging control commands, and thus the stability of the charging current, this unit is configured to monitor the ripple of the output current in real time during charging using a high-frequency sampler. When the effective value of the output current ripple exceeds a preset alarm threshold, such as 5% of the nominal current, the system does not attribute it to a charging hardware failure. Instead, it judges it as a sign of deterioration or congestion in the communication channel quality of the ad hoc network. This judgment will immediately trigger preset communication link protection or switching strategies, such as notifying the dual-channel communication arbitration unit to prepare to switch to the narrowband heartbeat channel in advance. This cross-domain perception mechanism, which predicts communication domain risks by monitoring subtle changes in the energy domain, upgrades the system's fault response from passive response to proactive prevention, significantly improving the system-level resilience of the entire cluster in unpredictable scenarios.

[0027] Example 1: In an autonomous power grid inspection mission covering a large area of ​​high mountains and canyons, a first UAV is hovering and collecting data on a critical transmission tower located in a remote valley. A sudden strong local downdraft causes the UAV to rapidly consume energy to maintain stable hovering. Its energy potential assessment unit calculates the dynamic state scalar. Within a short period, the drone rapidly dropped below the preset safety threshold. At this point, not only was the distance to any preset charging station greater than the energy required for its safe return, but its data collection task, which had high priority, could not be interrupted midway. Given this situation, the first drone's task and energy opportunity generation unit was immediately triggered, broadcasting an opportunity packet to the ad hoc network containing its own identification, the sequence of remaining data collection points on the key transmission towers to be handed over, and the precise remaining flight time. Meanwhile, a second drone, performing a routine inspection task on another ridge, possessed a higher dynamic state scalar... After receiving the opportunity packet, its distributed decision-making unit determines the best responder by calculating the time cost and responds accordingly. However, when the second UAV flies into the valley area, its dual-channel communication arbitration unit detects that the communication quality of the high-bandwidth communication channel has significantly decreased due to the complex terrain. This mechanism is then triggered, and it actively switches to the narrowband heartbeat communication channel to maintain basic communication with the first UAV, thus ensuring the execution of the core objective of energy support mission under communication degradation conditions.

[0028] After the second UAV approaches and establishes a mechanical connection with the first UAV through the energy transfer physical locking unit, the non-contact wireless inductive energy replenishment process is initiated. At this moment, the energy flow residual self-diagnostic unit configured on the second UAV begins to operate. The output current ripple it monitors slightly exceeds the preset alarm threshold due to interference from the unstable communication link on the charging control command. This physical representation of the energy domain objectively provides another physical dimension to corroborate the previous judgment of the dual-channel communication arbitration unit regarding the communication domain state, forming a de facto cross-domain information verification closed loop, further enhancing the system's comprehensive perception capability of complex environments. During this process, the micro-vibration adaptive compensation units on both UAVs work together to invert the relative vibration parameters caused by the turbulent wind at the bottom of the valley using the communication carrier phase difference, and suppress and compensate for them by adjusting the rotor speed and hovering height, thereby achieving a single-system... Within the system architecture, it simultaneously addresses two interrelated challenges: energy transmission stability under physical disturbances and communication reliability in weak signal environments. After energy replenishment, the collaborative execution unit not only takes over the tasks to be performed by the first UAV, but also, because the second UAV is geographically closer to the next task point, it directly continues to perform subsequent inspection tasks, while the first UAV, having received energy replenishment, goes into standby or returns to base. By transforming potential single-point failures into a dynamic reorganization of energy and tasks within the cluster, the system provides a mechanism that allows the coverage boundary and duration of inspection operations to no longer be limited by the endurance of a single node under worst-case conditions, but rather to depend on the overall energy reserves and collaborative efficiency of the cluster. The constraint of returning to base for recharging, which had to be considered in task planning, is replaced by a dynamic and on-demand energy network during the task execution phase.

[0029] Example 2: To quantitatively verify the effectiveness of the micro-vibration adaptive compensation unit of the present invention in maintaining the efficiency of non-contact wireless inductive energy replenishment under dynamic disturbances, this example constructs a ground-based hardware-in-the-loop simulation test platform. This platform consists of a second UAV acting as the power supplier and a first UAV acting as the power receiver. Both UAVs are equipped with a complete energy transfer physical locking unit and a micro-vibration adaptive compensation unit. The first UAV is fixed to a six-degree-of-freedom motion platform, which can accurately reproduce the vibrations caused by airflow disturbances according to a preset program. The second UAV hovers above the first UAV and is mechanically connected to it through the physical locking unit. The test environment is also equipped with a high-precision power analyzer and a laser displacement sensor, used to record the energy transfer efficiency and the relative displacement between the charging coils of the two UAVs in real time, respectively. The core parameter of the test, namely the vibration profile simulated by the six-degree-of-freedom motion platform, is... The setting follows a decision-making logic aimed at engineering realism. Technical factors influencing the parameter setting include the frequency range of disturbances caused by wind shear in real flight environments, and the sensitivity of the UAV's rotor system to specific frequencies. The parameter setting must ensure that the applied vibration is sufficient to challenge the alignment of the wireless charging coil to test the performance of the compensation unit, while avoiding the introduction of unconventional vibrations that would trigger the flight control system to enter protection mode, so as to ensure the validity of the test results for typical application scenarios. Based on this, the decision rule is set to select the main vibration frequency in the frequency range of 0.5Hz to 5Hz. This range covers most common atmospheric turbulence frequencies in the low and medium altitudes. In this experiment, for a simulated gust of wind between high-rise buildings in a city, a sinusoidal vibration with a main frequency of 2.5Hz and an amplitude of ±1.5cm was superimposed with a low-frequency random drift signal as the benchmark challenge for the compensation system.

[0030] The experiment was conducted by comparing a control group and an experimental group. Both groups were run under the condition that the aforementioned vibration profile was activated. In the control group, the first and second UAVs had their micro-vibration adaptive compensation units disabled, relying solely on the rigid connection of the physical locking unit to counteract vibration. In the experimental group, the full compensation function was enabled. After the experiment started, in the control group, it was observed that the relative displacement between the charging coils of the two UAVs changed drastically with the platform vibration. The peak displacement recorded by the laser displacement sensor reached 14.8 mm, and the power analyzer showed that the energy transmission efficiency fluctuated significantly with the displacement, dropping to as low as 37% of the nominal value. In the experimental group, when vibration was applied, the micro-vibration adaptive compensation unit of the first UAV, which was powered, analyzed the phase difference of the communication carrier to deduce the 2.5 Hz main vibration frequency and drove the rotor to generate a reverse compensation torque. At the same time, the second UAV, which was powered, also dynamically adjusted its hovering height according to the shared vibration parameters. Under the synergistic effect of the two UAVs, the relative displacement between the coils was significantly suppressed, with the peak displacement not exceeding 3.5 mm, and the energy transmission efficiency was stably maintained above 85% of the nominal value.

[0031] Table 1: A comparison table of data sampling for some key states during the experiment.

[0032] The experimental data, as shown in Table 1, indicates a direct correlation between the stability of energy transmission efficiency in the experimental group and the activation state of the micro-vibration adaptive compensation unit. This unit actively compensates for the impact of external physical disturbances on energy transmission efficiency within a specific range. The results of this experiment confirm the engineering feasibility of the collaborative mechanism proposed in this invention, which is based on communication physical layer signal feedback and flight dynamics active compensation, and provide a technical approach for achieving efficient energy replenishment between UAVs in dynamic and non-ideal environments.

[0033] Example 3: This example combines Figures 1 to 3 This describes an autonomous inspection and charging nesting system based on a drone self-organizing network, such as... Figure 1 As shown, firstly, UAV 1 is in normal inspection mode, and its internal energy potential assessment unit will calculate the dynamic state scalar in real time. If the calculated result is less than the safety threshold, the task and energy opportunity generation unit is activated. This unit broadcasts an opportunity packet containing task information and remaining flight time information to the ad hoc network. Upon receiving this opportunity packet, the neighboring UAVs 2, 3...N calculate and submit their response time costs based on their local state and the opportunity packet information. The system selects the UAV with the lowest time cost as the responder. The selected UAV 2 responder then activates its cooperative execution unit, driving itself to the target location, i.e., approaching UAV 1. After approaching, it performs non-contact wireless energy replenishment. This process includes physical locking and energy transfer. To ensure efficient energy replenishment, the micro-vibration adaptive compensation unit inverts vibration parameters based on communication phase difference and actively suppresses disturbances through reverse torque and altitude adjustment, further guaranteeing energy transmission efficiency. The system also includes a dual-channel communication arbitration unit, which monitors the quality of the high-bandwidth communication channel in real time. If the quality falls below a preset threshold, it automatically downgrades to a narrowband heartbeat channel to maintain basic energy coordination in harsh communication environments. After energy replenishment is completed, the system enters the mission takeover and dynamic reconfiguration phase: UAV2 takes over the mission, UAV1 receives energy, and chooses to continue executing the mission or return to standby depending on the situation, thereby completing the mission and energy coordination reconfiguration between UAVs.

[0034] like Figure 2As shown in the figure, three types of broken lines are used to identify three typical terrain environments: low-risk plains, medium-risk mountains, and high-risk canyons. On the X-axis, which represents the response distance, the distance ranges from 5km to 40km, with samples taken at 5km intervals. The Y-axis represents the corresponding time cost in seconds, used to quantify the comprehensive time cost required for the UAV to complete the support mission from receiving the opportunity packet. The solid dots in the figure represent the time cost curve in the low-risk plain environment, where the time cost increases linearly and slowly with the increase of response distance. The dashed triangles mark the medium-risk mountain environment, showing a moderate increase in cost due to the complex terrain. The dashed squares represent the high-risk canyon environment, with the steepest curve slope, indicating that the time cost of UAV response missions is significantly higher in high-risk canyon terrain than in other terrain types.

[0035] like Figure 3 As shown, firstly, the hazardous-level UAV detects the communication quality through its dual-channel communication arbitration unit. If the current high-bandwidth channel communication quality is found to be below the threshold, the system will automatically suspend opportunistic packet broadcasting and switch to a narrowband channel. Subsequently, the hazardous-level UAV broadcasts its heartbeat identity / hazard level to the network through the switched narrowband heartbeat channel. The heartbeat packet is forwarded to a nearby safe-level UAV via the narrowband channel. Upon receiving the heartbeat packet, the safe-level UAV immediately identifies the hazard level signal and confirms its own safe status. Then, it uses the algorithm logic to calculate the closest distance to determine the relative distance relationship with the hazardous-level UAV. After confirming the response, the safe-level UAV will autonomously move closer to the hazardous-level UAV and perform a basic energy replenishment operation upon approach. After replenishment, the system will update the status level and notify confirmation of replenishment completion. At this point, the communication link returns to basic maintenance to ensure the system's survivability under minimum communication resources.

[0036] Example 4: To ensure that the autonomous inspection and charging nesting system of the present invention possesses engineering determinism and environmental adaptability in its internal decision-making and evaluation procedures during specific deployment, this example discloses an offline calibration and parameter self-tuning method applied before system deployment. The application scenario of this method is to load a set of general system software onto a specific model of UAV hardware, and through a series of standardized tests, generate a set of operating parameters for this specific hardware and software combination. This method first measures the basic cruise power consumption of the UAV. Calibration was performed in an open test range under standard atmospheric conditions. The UAV was instructed to execute a closed-loop flight path with a total range of five kilometers at its typical inspection speed. The onboard power management unit recorded the instantaneous power consumption during the flight at a frequency of 1Hz. After the flight, the power consumption data during takeoff and landing were removed, and the arithmetic mean of the power consumption data points during the cruise phase was taken. This result was determined as the baseline cruise power consumption of this UAV model under standard conditions. Furthermore, the energy required for the return journey... The calculations can dynamically reflect the impact of wind resistance. During real-time flight, the system calculates the motor output power reported by the flight control system and the baseline cruise power consumption. The difference is used to calculate a dimensionless drag compensation coefficient in real time through a lookup table, which is then used for dynamic correction. The calculation results were then used to set a non-fixed safety threshold for the dynamic state scalar of the trigger opportunity packet broadcast. The principle for setting this threshold was to balance the relationship between the margin of task execution and potential risks. Specifically, during the task planning phase, the inspection area was divided into multiple grids with different risk levels based on obstacle density, electromagnetic interference intensity, and the distribution of non-landing areas. A basic time margin was set for each risk level. The basic margin for an open area marked as low risk was set to 120 seconds, while the basic margin for a complex obstacle area marked as high risk was set to 300 seconds. During the operation of the UAV, its safety threshold would be adaptively adjusted according to the risk level of its current grid.

[0037] For the calculation of response time cost in the distributed decision-making unit, the estimated energy replenishment time is determined as follows: the energy difference required for the requesting first UAV to recover to the safety threshold of its location is divided by the wireless charging power of that model of UAV under nominal conditions; the result is the estimated energy replenishment time. For the secondary subcontracting of tasks that the collaborative execution unit may perform, the triggering logic is determined as follows: when the second UAV takes over the task, if its own dynamic state scalar... The secondary packet splitting mechanism is activated only when the system detects a third drone in the network with a better geographical location and a dynamic state scalar value that is 150% higher than the current scalar value of the second drone. Finally, the linkage threshold of the dual-channel communication arbitration unit and the energy flow residual self-diagnosis unit is calibrated collaboratively. In a shielded laboratory, the signal-to-noise ratio (SNR) of the high-bandwidth communication channel between drones is gradually reduced through a programmable signal attenuator. At each SNR level, a standard power wireless charging process is executed, and the effective value of the charging output current ripple is recorded. Through this process, a mapping relationship database between channel SNR and current ripple value is established. Subsequently, an SNR value defined as the communication quality critical point, i.e., 12dB, is selected. Its corresponding current ripple value in the database, i.e., 4.5% of the nominal current, is fixed as the alarm threshold of the energy flow residual self-diagnosis unit. Thus, when the drone detects that the current ripple reaches this threshold during actual flight, it can infer from the database relationship that the communication channel quality has deteriorated to the critical point and trigger the preset communication link switching strategy.

[0038] Example 5: In a specific deployment scenario, the procedure for generating the geospatial risk level map upon which the system of this invention depends for its operation is as follows: First, acquire the digital elevation model and three-dimensional obstacle data of the target inspection area, and calculate the ground complexity and airspace clearance parameters of each geographic grid through geometric analysis; Second, dispatch a surveying UAV equipped with a spectrum analyzer to fly along a predetermined grid path, collect and record the broadband electromagnetic interference field strength at different locations, and form an electromagnetic environment interference layer; Finally, normalize the raw data of the three dimensions of terrain complexity, obstacle density, and electromagnetic interference intensity, and calculate a comprehensive risk index for each geographic grid through a preset weighted summation function. The risk indices of all grids together constitute the risk level map of the area, which is then distributed to each UAV in the cluster.

[0039] To address the potential situation where no node in the network is capable of safely performing support tasks, the system incorporates a collaborative task veto and self-rescue escalation procedure. After calculating the response time cost, a second UAV must perform an additional safety self-assessment, which predicts its expected dynamic state scalar after recharging the first UAV and returning to its own safe zone. If the expected scalar value is lower than the safety threshold of its own location, the second UAV will actively abandon the response; if the first UAV does not receive a response from any UAV that has passed the safety self-assessment within a specific time window after broadcasting the opportunity package, i.e., 60 seconds, its mission and energy opportunity generation unit will automatically cancel the opportunity package and immediately trigger the highest priority autonomous return or emergency landing procedure. This mechanism provides a definite final action plan for the system when it faces a dilemma that cannot be resolved by coordinated energy replenishment.

[0040] Example 6: To specifically determine the geometric parameters of the energy transfer physical locking unit and the control law of the micro-vibration adaptive compensation unit, the following procedure was adopted in a hardware-in-the-loop simulation and control parameter calibration scenario: The anchor point of the locking unit was designed as an inverted conical funnel with an interior angle of 30 degrees. Its depth and the engagement length of the hook were determined in an offline optimization program. This program, through simulation calculation, under a given set of initial relative displacement and velocity disturbances, sought the geometric dimension combination that maximized the success rate of physical capture, and finally determined that the funnel depth was five centimeters and the effective engagement length of the hook was three centimeters; the control law used to suppress relative vibration in the micro-vibration adaptive compensation unit was... During calibration, the control law was determined to be a proportional-integral-derivative (PID) controller. Its input is the relative displacement of the coil obtained by inverting the phase difference of the communication carrier, and its output is the adjustment amount of the UAV rotor speed. During the calibration process, the six-degree-of-freedom motion platform was instructed to apply a series of harmonic vibrations at a single frequency, covering a frequency range of 0.5Hz to 5Hz. At each frequency point, the proportional, integral, and derivative gain parameters of the controller were iteratively adjusted, and the root mean square value of the residual relative displacement measured by the laser displacement sensor was used as the performance index to find a set of optimal gain combinations that minimized this index. This set of optimal gain parameters was then embedded into the flight control software of the UAV.

[0041] In the same simulation environment, the Kalman filter algorithm used by the micro-vibration adaptive compensation unit defines its internal state vector as a four-dimensional vector containing the relative position and relative velocity on the two-dimensional plane between the charging coils. Its state transition matrix is ​​constructed based on a discrete-time model of Newton's laws of motion, which assumes that the relative acceleration is Gaussian white noise in a short time. Its measurement matrix defines the linear relationship between the observed value of the communication carrier phase difference and the relative position component in the state vector. To cope with the abnormal situation of complete communication interruption that may occur during energy replenishment, the system sets up a two-way heartbeat protection mechanism. During the physical locking process between the two machines, if either party does not receive the other party's heartbeat packet within 500 milliseconds, it is judged that a complete communication interruption has occurred. At this time, the two machines will immediately stop the locking and energy transmission actions and simultaneously trigger a preset emergency separation procedure that does not require communication coordination. That is, the receiving drone climbs vertically by five meters while the supplying drone descends vertically by five meters, thereby establishing a safe distance in the shortest possible time.

[0042] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0043] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An autonomous inspection and charging nest system based on unmanned aerial vehicle ad hoc network, characterized in that, include: The energy potential assessment unit, configured in each drone in the cluster, is used to calculate and output a dynamic state scalar representing the available time in real time based on the drone's current remaining energy, drone location information, and pre-stored charging nest location information. The dynamic state scalar is the result of dividing the difference between the current remaining energy and the energy required to return to the nearest charging nest by the average cruise power consumption. The task and energy opportunity generation unit is configured in any of the first UAVs in the cluster. When the dynamic state scalar of the first UAV is lower than a preset safety threshold, it generates and broadcasts an opportunity packet containing the task information to be performed by the first UAV and the remaining flight time information to the ad hoc network. The distributed decision unit, configured in any second UAV in the ad hoc network, is used to receive opportunity packets and calculate the time cost required for the second UAV to respond to the opportunity packet based on its own state and opportunity packet information. The time cost is the sum of the time required for the second UAV to fly to the first UAV for energy replenishment and fly away, minus the dynamic state scalar of the second UAV before responding. The collaborative execution unit is configured in the second drone with the lowest time cost in the ad hoc network. It is used to respond to the first drone, drive the second drone to move to the vicinity of the first drone for non-contact wireless inductive energy replenishment, and take over the tasks to be performed by the first drone.

2. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, Dynamic state scalar The calculation method is as follows: ,in, This indicates the drone's current remaining energy. This indicates the energy required for the drone to return to its nearest charging base. This indicates the average cruise power consumption of the drone.

3. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, The system also includes a dual-channel communication arbitration unit, configured on each UAV, used to set up a high-bandwidth communication channel and a narrowband heartbeat communication channel; the dual-channel communication arbitration unit monitors the communication quality of the high-bandwidth communication channel, and when the communication quality of the high-bandwidth communication channel is lower than a preset threshold, it suspends the broadcast of opportunity packets and switches to the narrowband heartbeat communication channel, transmitting only heartbeat packets containing the UAV's identity and quantized dynamic status scalar level.

4. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 3, characterized in that, The dynamic state scalar levels include safety level, critical level, and danger level; in communication resource-constrained mode, a drone with a danger level is configured to move to the nearest drone among those that meet the safety level criteria and perform basic energy replenishment.

5. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, The system also includes: an energy transfer physical locking unit, configured on each drone, used to perform electronic locking of the physical grappling hook and anchor point during non-contact wireless inductive energy replenishment, to ensure a successful physical connection when the two drones approach each other; after confirming a successful physical connection, the energy transfer physical locking unit authorizes the wireless inductive charging module in the cooperative execution unit to transfer energy at maximum power.

6. The autonomous inspection and charging nesting system based on a drone self-organizing network according to claim 5, characterized in that, If the energy transfer physical locking unit fails to lock after multiple attempts, it rejects the energy replenishment mission and instructs the first UAV to perform a return-to-home or on-site landing procedure, while broadcasting the energy transfer failure event to the ad hoc network.

7. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, The system also includes: a micro-vibration adaptive compensation unit, configured on the first UAV, used to actively send phase detection pulses using its communication module during the energy replenishment process, and to determine the relative vibration parameters between the two UAVs based on the phase difference sequence of the received carrier signal. The relative vibration parameters are the determined main frequency and amplitude. The micro-vibration adaptive compensation unit adjusts the rotational speed of the rotor of the first UAV to generate a reverse torque based on the relative vibration parameters.

8. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 7, characterized in that, The micro-vibration adaptive compensation unit is also configured on the second UAV to adjust the hovering height of the second UAV according to the relative vibration parameters.

9. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, The system also includes: an energy flow residual self-diagnostic unit, configured on the drone performing energy replenishment, used to monitor the output current ripple during the contactless wireless inductive charging process in real time; when the output current ripple exceeds the preset alarm threshold, the energy flow residual self-diagnostic unit determines that the communication channel quality of the self-organizing network has deteriorated and triggers the preset communication link protection or switching strategy.

10. The autonomous inspection and charging nesting system based on UAV self-organizing network according to claim 1, characterized in that, After taking over the task, the collaborative execution unit is also configured to subcontract part or all of the task to a third UAV in the ad hoc network with a higher dynamic state scalar and a geographical location that better meets the task requirements, based on its own dynamic state scalar and geographical location.

Images