Unmanned aerial vehicle autonomous homeward voyage charging system based on wireless charging

Abstract

The invention discloses an unmanned aerial vehicle autonomous homeward voyage charging system based on wireless charging, and relates to the field of flight control and energy management, and the system comprises the steps: monitoring the flight path, remaining power and environmental parameters of an unmanned aerial vehicle, predicting the remaining power required for completing a task, and triggering a homeward voyage condition signal and the current position information of the unmanned aerial vehicle; generating an optimal homeward flight path of the nearest wireless charging platform based on the homeward flight triggering condition signal and the current position information of the unmanned aerial vehicle; obtaining an optimal charging point based on the generated return path, and sending a deployment instruction to the mobile charging platform; the mobile charging platform advances according to the deployment instruction path, the unmanned aerial vehicle returns according to the optimal return path, the course is corrected through inertial navigation and a visual algorithm, and the landing point of the unmanned aerial vehicle is obtained. Fine adjustment and positioning of the flight attitude are formed by using the visual identification and attitude control module, and a closed-loop flight control mechanism based on image processing and attitude adjustment amount calculation is realized.

Description

Technical Field

[0001] The present invention relates to the fields of flight control and energy management, and in particular to an autonomous return-to-home charging system for unmanned aerial vehicles (UAVs) based on wireless charging. Background Art

[0002] In the field of intelligent flight control and energy management, drones, as aerial platforms with autonomous flight capabilities, have been widely used in multiple scenarios such as inspection, surveying and mapping, logistics, and security. With the increasing complexity of tasks, how to efficiently manage drone power and extend its mission duration has become a hot issue in research and engineering practice. The mainstream methods mainly rely on preset routes combined with timed charging strategies, or point-to-point charging control through fixed charging piles and take-off and landing platforms. Some solutions attempt to introduce wireless charging devices to achieve contact or contactless charging to improve energy replenishment flexibility. Such methods focus on performing energy replenishment at the completion of the mission or at the front end of the flight, but lack a complete adaptive closed-loop mechanism for dynamic energy prediction and coordinated scheduling response of charging resources during the mission.

[0003] However, in actual applications, fixed-path return and preset charging strategies often cannot take into account the dynamic nature of mission requirements and the impact of environmental changes on energy consumption. Traditional methods lack the ability to dynamically evaluate the coupling relationship between the remaining power of the UAV mission and the return path. It is difficult to judge in real time whether there is a power basis for safe return during complex missions. The deployment of charging platforms is still mainly static, and lacks the ability to coordinate optimization with the UAV position and path, which limits the application potential of wireless charging technology in mission continuity scenarios. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an autonomous return-to-home charging system for UAVs based on wireless charging to solve the problems of insufficient dynamic evaluation of flight energy consumption and disconnection between charging platform deployment and path scheduling.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: The present invention provides a UAV autonomous return-to-home charging system based on wireless charging, which includes monitoring the UAV's flight path, remaining power and environmental parameters, predicting the remaining power required to complete the mission, triggering a return condition signal and the UAV's current location information; Based on the return condition trigger signal and the current location of the drone, the optimal return path to the nearest wireless charging platform is generated; Based on the generated return path, the optimal charging point is obtained and deployment instructions are sent to the mobile charging platform; The mobile charging platform moves according to the deployment instruction path, and the UAV returns according to the optimal return path. The course is corrected through inertial navigation and visual algorithms to obtain the landing point of the UAV; The visual recognition and attitude control modules are used to fine-tune and position the flight attitude. The coupling detection between the mobile charging platform and the drone is used to determine the position alignment. The drone battery voltage, current, and temperature are monitored to determine the drone's full charge status. Based on the drone's full power status, the attitude lock is released and the progress and remaining task list are executed according to the task scheduling instructions.

[0007] As a preferred solution of the wireless charging-based autonomous return-to-home charging system for drones of the present invention, the system monitors the flight path, remaining power, and environmental parameters of the drone, predicts the remaining power required to complete the mission, and triggers a return-to-home condition signal and the drone's current location information, including the following steps: The onboard sensors collect real-time data on flight path, remaining battery power, and environmental parameters to form the UAV flight status information; The flight status information is input into the mission energy consumption prediction model, and combined with mission parameters such as the remaining flight distance, the power required to complete the mission is obtained; The current remaining power is compared with the power required to complete the mission, and the return condition signal is triggered to obtain the current position information of the drone.

[0008] As a preferred solution of the wireless charging-based autonomous return-to-home charging system for drones of the present invention, the optimal return path to the nearest wireless charging platform is generated based on the return condition triggering signal and the current location information of the drone, including the following steps: By triggering the return signal and the current position of the drone, an initial path planning data packet is formed; Based on the initial path planning data package, the current geographic location and surrounding obstacle map are loaded, the environmental parameters and the power required to complete the task are imported, and the composite cost weight of each grid edge is obtained; The composite cost weight of each grid edge is processed using the Dijkstra algorithm to obtain a set of candidate paths; Filter out the optimal return path with the shortest path distance from the candidate path set.

[0009] As a preferred solution of the autonomous return-to-home charging system for UAVs based on wireless charging according to the present invention, the optimal charging point is obtained based on the generated return path, and a deployment instruction is sent to the mobile charging platform, including the following steps: Based on the optimal return path, the path node sequence and the spatial coordinates and power consumption trend information of each node are extracted; Based on the current remaining power of the UAV and the estimated cumulative energy consumption of each node on the path, a set of candidate charging nodes is screened, and the node at the end of the path is selected as the optimal charging point; Based on the optimal charging point, deployment instructions are sent to the mobile charging platform.

[0010] As a preferred solution of the wireless charging-based autonomous return-to-home charging system for drones of the present invention, wherein: the mobile charging platform moves according to the deployment instruction path, and the drone returns according to the optimal return path, the system includes the following steps: The mobile charging platform receives and parses the deployment instructions, extracting the target charging point coordinates and the travel path sequence; The mobile charging platform parses the path sequence and obtains the mobile charging platform's position information. After receiving the return command, the drone loads the optimal return path and initializes the current remaining power and position information. Based on the return path, the drone calculates the flight control instructions for each segment and performs real-time visual recognition and flight fine-tuning according to the position information of the mobile charging platform.

[0011] As a preferred solution of the wireless charging-based autonomous return-to-home charging system for drones of the present invention, the following steps are included: correcting the heading by inertial navigation and visual algorithm to obtain the landing point of the drone: The drone loads the coordinates of the endpoint of the optimal return path and initializes inertial navigation to obtain the initial pose information; Based on the initial posture, the current position and posture of the UAV are calculated in real time using the integration method to obtain the estimated state sequence; By fusing the pose information estimated by inertial navigation and the displacement measured by vision, the extended Kalman filter algorithm is used to correct the current state of the UAV and obtain the landing point of the UAV.

[0012] As a preferred solution of the autonomous return-to-home charging system for UAVs based on wireless charging described in the present invention, the visual recognition and attitude control module is used to form fine-tuning and positioning of the flight attitude, including the following steps: Use the drone camera to collect images of the landing area in real time, pre-process the images, and output enhanced images; Match the enhanced image with the drone landing point and output the matching feature pair; Based on the matching feature pairs, the attitude adjustment value is obtained and input into the PID controller to generate fine-tuning flight control instructions; Based on the execution control instructions, the flight attitude is adjusted to obtain fine-tuning and positioning of the flight attitude.

[0013] As a preferred solution of the autonomous return-to-home charging system for UAVs based on wireless charging according to the present invention, the following steps are included: The UAV receives the platform's position information and combines it with its own inertial navigation results to obtain its current position; The drone camera collects platform images and obtains the platform's visual feature locations through image processing and target recognition algorithms; Match the visual features with the platform pose and calculate the relative position error between the UAV and the platform; Based on the relative position error, a judgment is made on the position alignment.

[0014] As a preferred solution of the UAV autonomous return charging system based on wireless charging of the present invention, wherein: monitoring the voltage, current and temperature of the UAV battery to obtain the full charge status of the UAV includes the following steps: Perform weighted moving average filtering on the voltage, current, and temperature data of the drone, output the smoothed voltage, current, and temperature data, and calculate the battery's state of charge (SOC) by combining it with the battery's calibrated open-circuit voltage curve. Use the smoothed temperature data to determine whether the current battery temperature is within the safe operating range, output the temperature safety status, and make a judgment on the drone battery power; The judgment result is made based on the drone battery power, triggering the drone battery management to output a full-charge confirmation signal.

[0015] As a preferred solution of the wireless charging-based autonomous return-to-home charging system for drones of the present invention, the following steps are included: based on the drone being fully charged, releasing the attitude lock, and executing the progress and remaining task list according to the task scheduling instructions: The drone battery management detects the SOC status and outputs a full-charge confirmation signal to release the attitude lock; Based on the UAV releasing the attitude lock, the progress and remaining task list are executed according to the task scheduling instructions.

[0016] The beneficial effects of the present invention are as follows: by generating the optimal return path to the nearest wireless charging platform based on the trigger return condition signal and the current position information of the UAV, it is possible to comprehensively consider multiple factors such as mission requirements, flight distance, obstacle information and energy consumption in complex terrain and mission scenarios, construct a composite cost model and generate the optimal return path based on this, so that the UAV can choose a safe and efficient return route under critical remaining power conditions, enhance the real-time and adaptability of the return path planning, form fine-tuning and positioning of the flight attitude by utilizing the visual recognition and attitude control modules, realize a closed-loop flight control mechanism based on image processing and attitude adjustment amount calculation, ensure that the UAV accurately docks with the mobile charging platform under dynamic conditions, improve the stability and accuracy during landing, and ensure the smooth implementation of wireless charging operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of the autonomous return-to-home charging system for UAVs based on wireless charging.

[0019] Figure 2 Schematic diagram of the charging node selection and command sending process.

[0020] Figure 3 Schematic diagram of the optimal return route and the best charging point Figure 4 Schematic diagram of the UAV charging management and mission recovery process DETAILED DESCRIPTION

[0021] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0022] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0024] Reference Figures 1 to 4 , which is the first embodiment of the present invention, provides a UAV autonomous return home charging system based on wireless charging, comprising the following steps: Monitor the drone's flight path, remaining power, and environmental parameters, predict the remaining power required to complete the mission, and trigger the return condition signal to obtain the drone's current location information.

[0025] The flight path, remaining battery power and environmental parameter data are collected in real time through onboard sensors to form the UAV flight status information.

[0026] Furthermore, the flight path of the UAV, the current remaining battery power, and environmental parameters (such as wind speed, temperature, humidity, and electromagnetic interference intensity) are collected in real time through onboard sensors to obtain the UAV flight status information.

[0027] The flight status information is input into the mission energy consumption prediction model, and combined with mission parameters such as the remaining flight distance, the power required to complete the mission is obtained.

[0028] Furthermore, we collected the flight status and environmental parameter data of the UAV, selected the physical-driven linear regression model, and constructed the energy consumption prediction model with the remaining mission distance, wind speed, temperature, humidity, load and heading angle as independent variables. We optimized the energy consumption prediction model by fitting the coefficients obtained by the least squares method, and predicted the amount of power required to complete the mission based on the current status. The reason for choosing linear regression is that this method has good computational efficiency and physical interpretability in the known parameter model. Specifically, the expression is, ; in, The remaining power required to complete the remaining tasks, is the basic energy consumption constant, is the regression coefficient of the impact of the remaining distance of the task on energy consumption, is the remaining distance of the task, is the regression coefficient of the effect of wind speed on energy consumption, is the current wind speed, is the regression coefficient of the effect of ambient temperature on energy consumption, is the current ambient temperature, is the regression coefficient of the effect of ambient humidity on energy consumption, is the relative humidity of the current environment, is the regression coefficient of the impact of UAV load on energy consumption, is the payload currently carried by the drone, is the regression coefficient of the effect of the cosine value of the angle between heading and wind direction on energy consumption, is the angle between the UAV heading and the wind direction.

[0029] The current remaining power is compared with the power required to complete the mission, and the return condition signal is triggered to obtain the current position information of the drone.

[0030] Furthermore, the current remaining power required to complete the task is compared with the power required to complete the task. If the current remaining power is less than the power required to complete the task, it is considered that the current energy is insufficient to complete the remaining task, and the return condition signal is triggered. When the return trigger condition is met, the position of the drone is predicted through the positioning device.

[0031] Based on the trigger return condition signal and the current location information of the drone, the optimal return path to the nearest wireless charging platform is generated.

[0032] By triggering the return signal and the current position of the drone, an initial path planning data packet is formed.

[0033] Furthermore, the return trigger signal and the current position coordinates of the drone, as well as the candidate wireless charging platform coordinate set, are received from the return trigger module, and the environmental obstacle map and the current environmental parameter set are obtained to form an initial path planning data packet.

[0034] Based on the initial path planning data package, the current geographic location and surrounding obstacle map are loaded, the environmental parameters and the power required to complete the task are imported, and the composite cost weight of each grid edge is obtained.

[0035] Furthermore, based on the initial path planning data packet, the current position coordinates of the UAV are extracted and the high-precision digital map data of the corresponding area is loaded to construct a three-dimensional rasterized grid environment model. The obstacle database is called, the position and size information of static and dynamic obstacles in the grid are marked, and the current environmental parameters, including wind speed, wind direction, temperature, humidity, and solar radiation, are imported. The composite cost weight of each grid edge is calculated according to the following cost function, combining the remaining battery power of the UAV with the estimated energy required to complete the mission.

[0036] The composite cost weight of each grid edge is processed using the Dijkstra algorithm to obtain a set of candidate paths.

[0037] Furthermore, the composite cost weight of each grid edge is searched using the Dijkstra algorithm with weighted edges. The starting point is the current position of the drone, and the end point set is all candidate platforms. The total path cost is calculated as the accumulation of all edge costs. Specifically, the expression is ; in, For the entire path The total cost, is the path index, To indicate the path From the starting point To the end node One edge of For the edge The cost weight, is the starting node index, The end node index.

[0038] Filter out the optimal return path with the shortest path distance from the candidate path set.

[0039] Furthermore, the candidate path set is smoothed in three dimensions, and the B-spline curve interpolation algorithm is used to fit the path points into a continuous smooth curve. Specifically, the expression is: ; in, For parameters The coordinates of the curve points below, is the total number of control points minus one, For the The coordinates of the starting node, For the The starting node is at the order Lower parameters The value of .

[0040] Based on the generated return path, the optimal charging point is obtained and deployment instructions are sent to the mobile charging platform.

[0041] Based on the optimal return path, the path node sequence and the spatial coordinates and power consumption trend information of each node are extracted.

[0042] Furthermore, based on the received optimal return path, the path data packet is parsed, and the spatial coordinate information of each node in the path, including the longitude, latitude and altitude values, is extracted in turn. At the same time, combined with the current energy consumption model of the drone, the cumulative power consumption trend corresponding to each node on the path is calculated and recorded. Specifically, the power consumption function is used to estimate the energy required for each node to fly to the next node, thereby forming a path node sequence and the spatial coordinates and power consumption trend information of each node.

[0043] Based on the current remaining power of the UAV and the estimated cumulative energy consumption of each node on the path, a set of candidate charging nodes is screened out, and the node at the end of the path is selected as the optimal charging point.

[0044] Furthermore, based on the current remaining power of the drone and the estimated cumulative energy consumption of each node on the path, the path nodes are screened in turn for energy consumption comparison. The nodes with the current remaining power ≥ the estimated energy consumption are selected as the candidate charging point set. Among the candidate charging point set, the node closest to the end point of the path and ensuring that the drone can reach it safely is selected as the best charging point. Based on the optimal charging point, deployment instructions are sent to the mobile charging platform.

[0045] Furthermore, based on the determined optimal charging point location coordinates and the estimated drone arrival time window, a complete deployment instruction data packet containing the charging point spatial information, deployment time period, and wireless charging power setting parameters is generated. The deployment instruction is then sent to the mobile charging platform in real time via a wireless communication module (such as 5G or a dedicated wireless communication protocol). The mobile charging platform moves according to the deployment instruction path, and the drone returns according to the optimal return path.

[0046] The mobile charging platform receives and parses the deployment instructions, extracting the target charging point coordinates and the travel path sequence.

[0047] Furthermore, after receiving the deployment instruction, the mobile charging platform performs structured decoding on the data packet through the communication parsing module, extracts the spatial coordinate information of the target charging point (such as longitude and latitude or global positioning coordinates), as well as the location information and navigation parameters of each path node in the path sequence, and then sets the target coordinates as the end point of the path. The path sequence is loaded into the path planning module according to time or priority order for subsequent travel control module to call, so as to realize autonomous navigation and travel of the platform from the current position to the designated charging point.

[0048] The mobile charging platform parses the path sequence and obtains the position information of the mobile charging platform. After receiving the return command, the drone loads the optimal return path and initializes the current remaining power and position information.

[0049] Furthermore, after parsing the deployment instructions, the mobile charging platform calculates the spatial coordinates and travel direction angles of each node in turn based on the extracted path node sequence. Combined with the platform's current position information, the differential positioning and inertial navigation module is used to calculate the platform's current posture information, including position and attitude angle. After the drone receives the return command, it loads the optimal return path sequence from the cache, obtains the current remaining power through the onboard power monitoring module, and uses the IMU (inertial measurement unit) and visual inertial navigation system to synchronously calculate the current drone's posture information, including position vector and attitude angle information.

[0050] Based on the return path, the drone calculates the flight control instructions for each segment and performs real-time visual recognition and flight fine-tuning according to the position information of the mobile charging platform.

[0051] Furthermore, after the drone receives the return path, the system first discretizes the path into a series of continuous waypoints. Combined with the currently initialized pose information, the system calculates the velocity vector and attitude adjustment angle required for each segment of the trajectory through flight control algorithms (such as linear trajectory tracking or waypoint interpolation), generating a sequence of segment-by-segment flight control commands. During flight, the system receives the pose information of the mobile charging platform in real time and uses the drone's onboard camera for visual recognition to extract the platform's spatial features (such as landmark patterns and structural edges). The recognition results are used to estimate the current relative pose error through a visual-inertial navigation fusion algorithm (such as the extended Kalman filter). Based on this error, the current control command is dynamically adjusted, and the corrected flight fine-tuning command is output to achieve stable and precise alignment with the platform.

[0052] The heading is corrected through inertial navigation and visual algorithms to obtain the landing point of the drone.

[0053] The drone loads the coordinates of the endpoint of the optimal return path and initializes inertial navigation to obtain the initial pose information.

[0054] Furthermore, the drone receives and analyzes the optimal return path, extracts the three-dimensional spatial coordinates of the path endpoint, and synchronously reads the initial data of the accelerometer and gyroscope in the inertial measurement unit (IMU). Based on the sensor data and the coordinates of the path endpoint, the drone's initial posture information, including position and attitude angles (pitch, roll, and yaw), is calculated through the inertial navigation initialization algorithm.

[0055] Based on the initial posture, the integration method is used to calculate the current position and posture of the UAV in real time to obtain the estimated state sequence.

[0056] Furthermore, based on the posture information obtained during drone initialization, the acceleration and angular velocity data from the inertial measurement unit (IMU) are collected in real time. The acceleration is integrated twice into the velocity and position using the integration method, while the angular velocity is integrated into the attitude angle. The current position and attitude changes of the drone are calculated moment by moment to form a continuous estimated state sequence.

[0057] By fusing the pose information estimated by inertial navigation and the displacement measured by vision, the extended Kalman filter algorithm is used to correct the current state of the UAV and obtain the landing point of the UAV.

[0058] Furthermore, the pose estimation obtained by inertial navigation and the displacement data measured by the visual sensor are collected in real time, and the pose estimation and displacement data measured by the visual sensor are used as the input of the extended Kalman filter (EKF). The information is fused through the state prediction and observation update steps to effectively filter out the accumulated error of inertial navigation and the noise of visual measurement, and gradually correct the current position and attitude estimation of the UAV to obtain the landing point of the UAV.

[0059] Utilize visual recognition and attitude control modules to achieve fine-tuning and positioning of flight attitude.

[0060] The drone camera is used to collect images of the landing area in real time, pre-process the images, and output enhanced images.

[0061] Furthermore, the drone camera continuously captures real-time images of the landing area at a preset frame rate, denoises the captured original images to remove environmental interference and sensor noise, and then uses image enhancement algorithms (such as histogram equalization or adaptive contrast enhancement) to improve the image clarity and detail, and outputs the enhanced image.

[0062] Match the enhanced image with the drone landing point and output the matching feature pair.

[0063] Furthermore, the enhanced image is input into the feature extraction module, and the SIFT (Scale-Invariant Feature Transform) algorithm is used to extract the key feature points in the image. The feature points are matched with the pre-stored feature templates of the corresponding area of ​​the drone landing point. The best matching pair is screened out by calculating the similarity of the feature descriptors, and the matching feature pair is output.

[0064] Based on the matching feature pairs, the attitude adjustment value is obtained, and the attitude adjustment value is input into the PID controller to generate fine-tuning flight control instructions.

[0065] Furthermore, the deviation between the current and ideal attitudes of the UAV is calculated based on the matching feature pairs, and the attitude adjustment amount (including the errors of the yaw angle, pitch angle, and roll angle) is obtained. The adjustment amount is passed as the input signal to the three-axis PID controller. The PID controller compensates for the error in real time according to the proportional, integral, and differential algorithms, and generates corresponding fine-tuning flight control instructions.

[0066] Based on the execution control instructions, the flight attitude is adjusted to obtain fine-tuning and positioning of the flight attitude.

[0067] Furthermore, after receiving the fine-tuning flight control command generated by the PID controller, the drone executes the corresponding rudder deflection adjustment operation, changing the attitude angle (yaw, pitch, roll) in real time to correct the flight attitude. The feedback loop of the flight control system continuously monitors the attitude changes to ensure that the adjusted flight attitude is accurate and stable, thereby completing the fine-tuning and positioning of the flight attitude and ensuring the safe and accurate landing of the drone.

[0068] The position alignment is determined through coupling detection between the mobile charging platform and the drone.

[0069] The UAV receives the platform's posture information and combines it with its own inertial navigation results to obtain its current posture.

[0070] Furthermore, after the mobile charging platform obtains the current position and attitude data, the drone receives the platform's position information in real time through the wireless communication module. As the current environmental reference coordinate system, the drone calls its inertial navigation and uses the integral calculation method to obtain the current position and attitude based on the linear acceleration and angular velocity collected by the built-in accelerometer and gyroscope.

[0071] The drone camera collects platform images and obtains the platform's visual feature locations through image processing and target recognition algorithms.

[0072] Furthermore, the drone camera continuously collects real-time image sequences of the mobile charging platform and inputs the images into a pre-designed image processing module. First, preprocessing operations such as grayscale conversion, noise filtering and edge detection are performed to enhance image quality. The target recognition algorithm based on convolutional neural network (CNN) is used to identify and extract unique visual landmarks or feature points on the platform, including color, shape and texture features. Through feature matching and positioning algorithms, the two-dimensional coordinate positions of these visual features on the image plane are calculated, and the camera intrinsic parameter matrix and extrinsic parameter matrix are combined to perform three-dimensional spatial positioning conversion to output the spatial position of the platform's visual features.

[0073] Match the visual features with the platform pose and calculate the relative position error between the UAV and the platform.

[0074] Furthermore, after obtaining the spatial position of the platform's visual features, spatial alignment is performed with the real-time posture information broadcast by the platform through its own inertial navigation or positioning. First, the platform's posture is converted to a unified reference coordinate system through a coordinate transformation matrix, and then the drone's own current posture is also mapped to the same coordinate system, and the visual feature position is projected to this coordinate system through an external parameter matrix to obtain the spatial displacement error vector.

[0075] Based on the relative position error, a judgment is made on the position alignment.

[0076] Furthermore, after obtaining the six-degree-of-freedom relative position error between the UAV and the platform, the current position and attitude are judged according to the set alignment threshold to determine whether they meet the docking requirements. The position error threshold is defined as ; ; If the UAV is considered to have achieved spatial alignment with the platform, the status flag is output to indicate "alignment completed". Otherwise, 0 is output to indicate "not aligned" and the signal is fed back to the flight controller for continuing to execute fine-tuning instructions. The judgment logic relies on the position and attitude errors calculated in the previous step as input, and the output is an alignment status signal used for control judgment and next action decision.

[0077] Monitor the drone battery voltage, current, and temperature to get the drone's full charge status.

[0078] The voltage, current and temperature data of the drone are filtered using a weighted moving average method to output the smoothed voltage, current and temperature data. Combined with the open circuit voltage curve of the battery calibration, the battery state of charge (SOC) is calculated.

[0079] Specifically, the collected voltage, current and temperature data are aggregated into multi-source data packets, and the weighted moving average filtering algorithm is applied to each of the multi-source data packets for smoothing. The expression is: ; in, for The temperature data after smoothing at the moment is the current moment The weighted average of The first The original measurement value at the moment, For the The corresponding weight coefficients are is the sum of the weighted data packets from multiple sources. An index for multi-source data.

[0080] Furthermore, based on the temperature-corrected smoothed voltage value and the battery's calibrated open circuit voltage-state of charge (OCV-SOC) curve, the current battery state of charge is estimated by looking up a table or using a fitting model (such as cubic spline interpolation). The estimated value serves as a key parameter for determining whether the battery is fully charged.

[0081] The smoothed temperature data is used to determine whether the current battery temperature is within the safe operating range, output the temperature safety status, and make a judgment on the battery power of the drone.

[0082] Furthermore, after obtaining the smoothed battery temperature data, the battery temperature data is compared with the battery safe operating temperature range to determine whether the current temperature is within the safe range. When the smoothed temperature data is less than the battery safe operating temperature range, the output temperature safety status is TRUE, indicating that the current temperature is suitable for battery charging and discharging operations. Otherwise, the output temperature safety status is FALSE, indicating that charging needs to be suspended or temperature control processing needs to be performed. Only when the temperature safety status is TRUE, the filtered voltage and current data are combined with the calibrated open circuit voltage-state of charge (OCV-SOC) mapping curve to calculate the current state of charge (SOC) of the battery, so as to achieve accurate assessment and status confirmation of the drone battery power.

[0083] The judgment result is made based on the drone battery power, triggering the drone battery management to output a full-charge confirmation signal.

[0084] Furthermore, a full charge threshold is set based on the previous state of charge value, and the current state of charge value is compared with the set full charge threshold, which is generally set to 95%-100% to ensure safety redundancy. When the conditions are met and the battery temperature is within a safe range, the current battery is determined to be fully charged, and the battery management unit automatically triggers the output of a full charge confirmation signal. The full charge confirmation signal serves as the basis for starting the subsequent flight control module or mission scheduling system, and is also recorded in the mission log for closed-loop control and status tracking of the charging mission, thereby realizing the logical closed-loop linkage of battery status judgment, charging control and flight preparation.

[0085] Based on the drone's full power status, the attitude lock is released and the progress and remaining task list are executed according to the task scheduling instructions.

[0086] The drone battery management detects the SOC status and outputs a full-charge confirmation signal to release the attitude lock.

[0087] Furthermore, the battery voltage, current and temperature data are collected and obtained in real time, and the raw data is smoothed using a weighted moving average filtering algorithm. The filtered voltage, current and temperature are used to estimate the battery's current state of charge. The SOC estimation method is based on the open circuit voltage (OCV) curve, and the current battery SOC value is obtained through a table lookup and matching method. If the SOC value exceeds the full charge threshold (for example, 95%), the battery management outputs a full charge confirmation signal. After receiving the signal, the flight control immediately executes the attitude lock release operation and sends a motor unlocking command to the attitude control module to enable the motor to rotate; reset the angular velocity and acceleration bias items in the attitude estimator, and update the navigation status flag to the "flyable" state to prepare for the subsequent mission launch.

[0088] Based on the UAV releasing the attitude lock, the progress and remaining task list are executed according to the task scheduling instructions.

[0089] Furthermore, after the drone receives the full-charge confirmation signal output by the battery management and successfully releases the attitude lock, the task management module immediately calls the task scheduler to obtain the current task execution progress status (indicated by the task number and completion flag, such as the task execution progress structure and the remaining task list, and reads the execution flag of the previous round of tasks in the flight control record. If it is in the "completed" status, the task is removed from it and the queue is updated. Otherwise, the task is retained and marked as "to be continued". Based on the current remaining task list, the scheduler selects the next task according to the task priority (which can be weighted according to multiple factors such as time urgency, task type, and distance to the task area) and encapsulates the task parameters (including target point coordinates, task type, required flight altitude, etc.) into an execution instruction structure.

[0090] In summary, the present invention generates the optimal return path to the nearest wireless charging platform based on the trigger return condition signal and the current position information of the UAV, and realizes comprehensive consideration of multiple factors such as mission requirements, flight distance, obstacle information and energy consumption in complex terrain and mission scenarios, constructs a composite cost model and generates the optimal return path based on it, so that the UAV can choose a safe and efficient return route under critical remaining power conditions, enhances the real-time and adaptability of return path planning, and forms fine-tuning and positioning of the flight attitude by utilizing the visual recognition and attitude control modules, realizing a closed-loop flight control mechanism based on image processing and attitude adjustment amount calculation, ensuring that the UAV accurately docks with the mobile charging platform under dynamic conditions, improving the stability and accuracy during landing, and ensuring the smooth implementation of wireless charging operations.

[0091] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A UAV autonomous return-to-home charging system based on wireless charging, characterized by: include, Monitor the drone's flight path, remaining battery power, and environmental parameters, predict the remaining battery power required to complete the mission, and trigger return condition signals and the drone's current location information; Based on the return condition trigger signal and the current location of the drone, the optimal return path to the nearest wireless charging platform is generated; Based on the generated return path, the optimal charging point is obtained and deployment instructions are sent to the mobile charging platform; The mobile charging platform moves according to the deployment instruction path, and the UAV returns according to the optimal return path. The course is corrected through inertial navigation and visual algorithms to obtain the landing point of the UAV; The visual recognition and attitude control modules are used to fine-tune and position the flight attitude. The coupling detection between the mobile charging platform and the drone is used to determine the position alignment. The drone battery voltage, current, and temperature are monitored to determine the drone's full charge status. Based on the drone's full power status, the attitude lock is released and the progress and remaining task list are executed according to the task scheduling instructions.

2. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 1, wherein: Monitor the flight path, remaining power and environmental parameters of the drone, predict the remaining power required to complete the mission, trigger the return condition signal and the current location information of the drone, including the following steps: The onboard sensors collect real-time data on flight path, remaining battery power, and environmental parameters to form the UAV flight status information; The flight status information is input into the mission energy consumption prediction model, and combined with mission parameters such as the remaining flight distance, the power required to complete the mission is obtained; The current remaining power is compared with the power required to complete the mission, and the return condition signal is triggered to obtain the current position information of the drone.

3. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 2, wherein: Based on the trigger return condition signal and the current location information of the UAV, the optimal return path to the nearest wireless charging platform is generated, including the following steps: By triggering the return signal and the current position of the drone, an initial path planning data packet is formed; Based on the initial path planning data package, the current geographic location and surrounding obstacle map are loaded, the environmental parameters and the power required to complete the task are imported, and the composite cost weight of each grid edge is obtained; The composite cost weight of each grid edge is processed using the Dijkstra algorithm to obtain a set of candidate paths; Filter out the optimal return path with the shortest path distance from the candidate path set.

4. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 3, characterized in that: Based on the generated return path, the optimal charging point is obtained and a deployment instruction is sent to the mobile charging platform, including the following steps: Based on the optimal return path, the path node sequence and the spatial coordinates and power consumption trend information of each node are extracted; Based on the current remaining power of the UAV and the estimated cumulative energy consumption of each node on the path, a set of candidate charging nodes is screened, and the node at the end of the path is selected as the optimal charging point; Based on the optimal charging point, deployment instructions are sent to the mobile charging platform.

5. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 4, characterized in that: The mobile charging platform moves according to the deployment instruction path, and the UAV returns according to the optimal return path. The following steps are included: The mobile charging platform receives and parses the deployment instructions, extracting the target charging point coordinates and the travel path sequence; The mobile charging platform parses the path sequence and obtains the mobile charging platform's position information. After receiving the return command, the drone loads the optimal return path and initializes the current remaining power and position information. Based on the return path, the drone calculates the flight control instructions for each segment and performs real-time visual recognition and flight fine-tuning according to the position information of the mobile charging platform.

6. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 5, characterized in that: Correct the course through inertial navigation and visual algorithm to obtain the landing point of the drone. The following steps are included: The drone loads the coordinates of the endpoint of the optimal return path and initializes inertial navigation to obtain the initial pose information; Based on the initial posture, the current position and posture of the UAV are calculated in real time using the integration method to obtain the estimated state sequence; By fusing the pose information estimated by inertial navigation and the displacement measured by vision, the extended Kalman filter algorithm is used to correct the current state of the UAV and obtain the landing point of the UAV.

7. The UAV autonomous return-to-home charging system based on wireless charging according to claim 6, characterized in that: Utilize visual recognition and attitude control modules to achieve fine-tuning and positioning of flight attitude. The following steps are included: Use the drone camera to collect images of the landing area in real time, pre-process the images, and output enhanced images; Match the enhanced image with the drone landing point and output the matching feature pair; Based on the matching feature pairs, the attitude adjustment value is obtained and input into the PID controller to generate fine-tuning flight control instructions; Based on the execution control instructions, the flight attitude is adjusted to obtain fine-tuning and positioning of the flight attitude.

8. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 7, characterized in that: The coupling detection between the mobile charging platform and the UAV is used to determine the position alignment, including the following steps: The UAV receives the platform's position information and combines it with its own inertial navigation results to obtain its current position; The drone camera collects platform images and obtains the platform's visual feature locations through image processing and target recognition algorithms; Match the visual features with the platform pose and calculate the relative position error between the UAV and the platform; Based on the relative position error, a judgment is made on the position alignment.

9. The wireless charging-based autonomous return-to-home charging system for unmanned aerial vehicles according to claim 8, characterized in that: Monitor the drone battery voltage, current and temperature to get the drone's full charge status, including the following steps: Perform weighted moving average filtering on the voltage, current, and temperature data of the drone, output the smoothed voltage, current, and temperature data, and calculate the battery's state of charge (SOC) by combining it with the battery's calibrated open-circuit voltage curve. Use the smoothed temperature data to determine whether the current battery temperature is within the safe operating range, output the temperature safety status, and make a judgment on the drone battery power; The judgment result is made based on the drone battery power, triggering the drone battery management to output a full-charge confirmation signal.

10. The UAV autonomous return-to-home charging system based on wireless charging according to claim 9, characterized in that: Based on the drone's full power status, release the attitude lock, execute the progress and remaining task list according to the task scheduling instructions, The following steps are included: The drone battery management detects the SOC status and outputs a full-charge confirmation signal to release the attitude lock; Based on the UAV releasing the attitude lock, the progress and remaining task list are executed according to the task scheduling instructions.

Images