Fire-fighting unmanned aerial vehicle quick release mechanism docking process adaptive precision control method

Abstract

The application relates to the technical field of automatic docking control of a quick-release mechanism of a fire-fighting unmanned plane, and provides a self-adaptive and accurate control method for a docking process of a quick-release mechanism of a fire-fighting unmanned plane. The quick-release mechanism of the fire-fighting unmanned plane comprises a ball-and-socket type double-end docking quick-release mechanism, the ball-and-socket type double-end docking quick-release mechanism is provided with two electric push rods, a ball socket and a ball socket flange; the method comprises the following steps: acquiring environmental data of a docking surface, constructing an adaptive control model of an output force and a pushing speed of the electric push rod; inputting the environmental data into the adaptive control model to obtain electric push rod driving parameters matched with a current docking state; according to the electric push rod driving parameters, the output force and the pushing speed of the two electric push rods are adjusted in real time; and the two electric push rods are controlled to synchronously push the ball socket and the ball socket flange to complete impact-free accurate docking. The application can adapt to various docking scenes with different loads, different wind disturbances and different initial attitudes, and solves the pain point of poor adaptability of fixed parameter driving.

Description

Technical Field

[0001] This application relates to the field of automatic docking control technology for quick-disassembly mechanisms of fire-fighting drones, and in particular to an adaptive and precise control method for the docking process of quick-disassembly mechanisms of fire-fighting drones. Background Technology

[0002] To enable rapid replacement of payloads for firefighting and reconnaissance missions, firefighting drones commonly employ a ball-and-socket type double-end docking quick-release mechanism, which uses dual electric push rods to lock the docking. Existing quick-release mechanisms often use electric push rods driven by fixed preset parameters, presenting the following clear technical problems: 1. Fixed output force and propulsion speed cannot adapt to the complex scenarios of fire fighting operations. It has extremely poor adaptability in the face of different initial docking attitude deviations, different weights of loads, and variable environmental wind disturbances: when the wind disturbance / deviation is large, the propulsion speed is too fast and it is easy to cause collision and jamming. Under heavy loads, insufficient output force is easy to cause docking failure. 2. Existing control methods cannot adjust the output force and propulsion speed of the dual electric push rods according to the real-time environment. The dual push rods have poor synchronization, are prone to misalignment during docking, and have large docking impact and low locking accuracy. 3. Existing technologies do not incorporate environmental data such as attitude deviation, load weight, and environmental wind disturbance to construct an adaptive control model. As a result, they cannot maintain an appropriate locking force after docking, which can easily lead to problems such as excessive locking damaging the mechanism or insufficient locking force causing the load to fall off.

[0003] Therefore, a method is urgently needed to solve at least one of the above problems. Summary of the Invention

[0004] This application provides an adaptive and precise control method for the docking process of a quick-release mechanism for firefighting drones, aiming to solve the problem that there is no technical solution in the existing technology that can adaptively adjust the parameters of dual electric push rods by combining multi-source environmental data to achieve impact-free and precise docking of the ball-and-socket quick-release mechanism of firefighting drones.

[0005] In a first aspect, embodiments of this application provide an adaptive and precise control method for the docking process of a quick-release mechanism for a fire-fighting drone. The quick-release mechanism for the fire-fighting drone includes a ball-and-socket type double-ended docking quick-release mechanism, wherein the ball-and-socket type double-ended docking quick-release mechanism is provided with two electric push rods, a ball socket, and a ball socket flange; the method includes: Acquire environmental data of the docking surface, including attitude deviation data, load weight data, and environmental wind disturbance data; construct an adaptive control model for the output force and propulsion speed of the electric push rod; The environmental data is input into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state; based on the electric actuator drive parameters, the output force and propulsion speed of the two electric actuators are adjusted in real time. Two electric push rods are controlled to simultaneously push the ball socket and the ball socket flange to complete a precise, impact-free connection; after the connection is completed, the electric push rods are controlled to maintain the preset locking force to achieve smooth locking.

[0006] In some embodiments, acquiring environmental data of the docking surface includes: acquiring docking surface attitude deviation data through a visual sensor, acquiring load weight data through a pressure sensor, and acquiring environmental wind disturbance data through a wind speed sensor.

[0007] In some embodiments, constructing an adaptive control model for the output force and propulsion speed of the electric push rod includes: collecting historical data of successful docking under different docking scenarios, labeling the output force and propulsion speed of the electric push rod corresponding to the historical data, and training the adaptive control model using a neural network algorithm.

[0008] In some embodiments, the step of inputting the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state includes: filtering and preprocessing the input environmental data to remove noise, inputting the preprocessed environmental data into the adaptive control model, and calculating the electric push rod drive parameters corresponding to the approach phase, docking phase, and locking phase, respectively.

[0009] In some embodiments, adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters includes: adjusting the output force and propulsion speed of each electric actuator separately, collecting the actual output force and actual propulsion speed of each electric actuator in real time, comparing the actual parameters with the drive parameters, and correcting the control signal of the electric actuator based on the comparison result.

[0010] In some embodiments, controlling the two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking includes: using a master-slave synchronous control strategy to control the synchronous movement of the two electric push rods, detecting the distance between the ball socket and the ball socket flange in real time, and reducing the pushing speed of the electric push rods when the distance is less than a preset threshold.

[0011] In some embodiments, after the docking is completed, the electric push rod is controlled to maintain a preset locking force to achieve smooth locking, including: detecting a docking completion signal, switching the electric push rod to the locking mode after receiving the docking completion signal, continuously monitoring the output locking force of the electric push rod, and increasing the output force of the electric push rod when the output locking force is lower than the preset locking force.

[0012] In some embodiments, the method further includes: identifying the current docking scenario, selecting a static docking mode or an aerial emergency docking mode based on the identification result, and inputting a preset control parameter set in the corresponding mode into the adaptive control model.

[0013] In some embodiments, the method further includes: calculating the probability of structural collision during the docking process in real time; when the probability of structural collision exceeds a preset threshold, reducing the advancing speed of the electric push rod and adjusting the output force difference between the two electric push rods to correct the attitude of the docking surface.

[0014] In some embodiments, the method further includes: detecting whether the docking is successful; when the docking is unsuccessful, automatically adjusting the attitude of the docking surface, re-executing the docking process, recording the number of docking retries, and issuing a docking failure alarm signal when the number of docking retries reaches a preset maximum.

[0015] This application addresses the uncertain interference in firefighting operations by integrating three types of environmental data—attitude deviation, load weight, and environmental wind disturbance—to construct an adaptive control model. This model outputs electric actuator drive parameters that match the current docking state, adapting to various docking scenarios with different loads, wind disturbances, and initial attitudes, thus solving the problem of poor adaptability of fixed-parameter drives. Furthermore, this application allows for real-time adjustment of the output force and propulsion speed of the dual electric actuators, ensuring synchronized action, preventing skew during docking, effectively reducing docking impact, eliminating the risk of collision and jamming, and achieving impact-free and precise docking. After docking, the electric actuators can be precisely controlled to maintain a preset locking force, preventing damage to the quick-release mechanism due to excessive locking force and preventing loosening and detachment due to insufficient locking force, significantly improving the reliability and safety of quick-release docking for firefighting drones in complex firefighting environments.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic flowchart illustrating the steps of an adaptive and precise control method for the docking process of a quick-release mechanism of a fire-fighting drone, provided in one embodiment of this application. Figure 2 This is a schematic diagram of the structure of a quick-release mechanism for a fire-fighting drone provided in one embodiment of this application; Figure 3 This is a schematic block diagram of the structure of an adaptive precision control system for the docking process of a quick-release mechanism for a fire-fighting drone, provided in one embodiment of this application. Figure 4 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the order described. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.

[0022] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, the terms "first" and "second" are used in the embodiments of the present invention to distinguish identical or similar items with essentially the same function and effect. Those skilled in the art will understand that the terms "first" and "second" do not limit the quantity or execution order, and the terms "first" and "second" are not necessarily different.

[0023] It should be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the scope of the application. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0024] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0025] To enable rapid replacement of payloads for firefighting and reconnaissance missions, firefighting drones commonly employ a ball-and-socket type double-end docking quick-release mechanism, which uses dual electric push rods to lock the docking. Existing quick-release mechanisms often use electric push rods driven by fixed preset parameters, presenting the following clear technical problems: 1. Fixed output force and propulsion speed cannot adapt to the complex scenarios of fire fighting operations. It has extremely poor adaptability in the face of different initial docking attitude deviations, different weights of loads, and variable environmental wind disturbances: when the wind disturbance / deviation is large, the propulsion speed is too fast and it is easy to cause collision and jamming. Under heavy loads, insufficient output force is easy to cause docking failure. 2. Existing control methods cannot adjust the output force and propulsion speed of the dual electric push rods according to the real-time environment. The dual push rods have poor synchronization, are prone to misalignment during docking, and have large docking impact and low locking accuracy. 3. Existing technologies do not incorporate environmental data such as attitude deviation, load weight, and environmental wind disturbance to construct an adaptive control model. As a result, they cannot maintain an appropriate locking force after docking, which can easily lead to problems such as excessive locking damaging the mechanism or insufficient locking force causing the load to fall off.

[0026] Therefore, a method is urgently needed to solve at least one of the above problems.

[0027] To solve the above problem, please refer to Figure 1 This application provides an adaptive and precise control method for the docking process of a quick-release mechanism of a fire-fighting drone, which is applied to computer equipment. The computer equipment can be deployed on a single server or server cluster, or on a handheld terminal, laptop, wearable device, or robot.

[0028] The quick-release mechanism for firefighting drones addressed in this invention is a ball-and-socket type double-end docking quick-release mechanism, as follows: Figure 2 The system includes two electric actuators (1), a ball socket (2), and a ball socket flange (3). The two electric actuators are symmetrically arranged on either side of the central connecting seat. One end of each actuator is hinged to the central connecting seat, and the other end is fixedly connected to the corresponding ball socket. The ball socket flange is fixed to the fuselage or mounting equipment of the fire-fighting drone, and the ball socket and flange mate to achieve docking and locking.

[0029] The provided adaptive precision control method for the docking process of the quick-release mechanism of the firefighting drone includes steps S101 to S103. Details are as follows: Step S101. Obtain environmental data of the docking surface, including attitude deviation data, load weight data and environmental wind disturbance data; construct an adaptive control model for the output force and propulsion speed of the electric push rod.

[0030] Specifically, step S101 involves acquiring environmental data of the docking surface and constructing an adaptive control model for the output force and propulsion speed of the electric actuator. The environmental data of the docking surface includes attitude deviation data, load weight data, and environmental wind disturbance data. Attitude deviation data refers to the angular deviation between the axis of the ball socket and the axis of the ball socket flange, and the positional deviation between the center of the ball socket and the center of the ball socket flange. Load weight data refers to the total weight of the equipment mounted on the ball socket flange. Environmental wind disturbance data refers to the wind speed and direction data at the docking location.

[0031] Constructing an adaptive control model for the output force and propulsion speed of the electric actuator is one of the core components of this invention. The model's function is to establish a mapping relationship between environmental data and the optimal output force and propulsion speed of the electric actuator. The model's inputs include attitude deviation data, load weight data, and environmental wind disturbance data; the outputs are the output force and propulsion speed parameters of the electric actuator. The model's construction requires extensive training based on real-world docking data using machine learning algorithms to ensure its adaptability to different docking scenarios and environmental conditions.

[0032] In actual execution, step S101 is divided into two parallel sub-steps: one sub-step is to acquire environmental data in real time during the docking process, and the other sub-step is to pre-build and train an adaptive control model. The real-time acquisition of environmental data is performed continuously before and during each docking to ensure the timeliness and accuracy of the data. The process of building the adaptive control model can be completed offline before docking or updated and optimized online during docking.

[0033] Step S102. Input the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state; adjust the output force and propulsion speed of the two electric push rods in real time according to the electric push rod drive parameters.

[0034] Specifically, step S102 involves inputting environmental data into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state, and then adjusting the output force and propulsion speed of the two electric actuators in real time based on these parameters. Before inputting the environmental data into the adaptive control model, the data needs to be preprocessed to remove noise and outliers, thereby improving data quality and reliability. After preprocessing, the data is input into the pre-trained adaptive control model, which calculates the optimal output force and propulsion speed parameters for each of the two electric actuators in the current docking state based on the input environmental data.

[0035] After obtaining the drive parameters of the electric linear actuators, the computer equipment converts these parameters into control signals that the electric linear actuators can recognize and sends them to the drivers of the two electric linear actuators. The drivers adjust the output force and propulsion speed of the electric linear actuators according to the control signals. During the docking process, the computer equipment continuously acquires real-time environmental data and the actual operating parameters of the electric linear actuators, and inputs this data into the adaptive control model to update the drive parameters of the electric linear actuators in real time, thereby achieving dynamic adjustment of the output force and propulsion speed of the electric linear actuators.

[0036] This real-time adjustment method can effectively cope with changes in environmental conditions during the docking process, such as sudden changes in wind speed, changes in load weight, or minor changes in the attitude of the docking surface, ensuring the stability and accuracy of the docking process.

[0037] Step S103. Control the two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking; after docking, control the electric push rods to maintain the preset locking force to achieve smooth locking.

[0038] Specifically, step S103 involves controlling two electric push rods to synchronously push the ball socket and the ball socket flange to complete a shock-free and precise docking. After docking, the electric push rods are controlled to maintain a preset locking force to achieve smooth locking. Synchronizing the movement of the two electric push rods is key to achieving shock-free and precise docking. If the movements of the two electric push rods are not synchronized, the ball socket may tilt, collide with the ball socket flange, cause structural damage, or even lead to docking failure. Therefore, a synchronous control strategy is needed to ensure that the output force and pushing speed of the two electric push rods remain consistent, allowing the ball socket to move smoothly towards the ball socket flange.

[0039] Once the ball joint and flange are fully engaged, the connection is complete. At this point, the electric actuator needs to maintain a certain locking force to prevent the ball joint and flange from separating during flight. The locking force needs to be set according to the load weight and flight environment to ensure that the ball joint and flange maintain a reliable connection even under maximum load and the most severe flight conditions.

[0040] In the locked state, the computer continuously monitors the output locking force of the electric actuator. If the output locking force is lower than the preset value, it will automatically increase the output force of the electric actuator to ensure the reliability of the locking.

[0041] In some embodiments, acquiring environmental data of the docking surface includes: acquiring docking surface attitude deviation data through a visual sensor, acquiring load weight data through a pressure sensor, and acquiring environmental wind disturbance data through a wind speed sensor.

[0042] This embodiment provides a detailed description of the specific implementation method for acquiring environmental data of the mating surface in step S101. A binocular vision sensor is used, installed above the intermediate connector, facing the ball socket and ball socket flange. The binocular vision sensor's acquisition frequency is set to thirty frames per second, enabling real-time capture of images of the ball socket and ball socket flange. By performing feature extraction and stereo matching on the images, the angular deviation between the axis of the ball socket and the axis of the ball socket flange, as well as the positional deviation between the center of the ball socket and the center of the ball socket flange, are calculated. The measurement accuracy of the angular deviation is ±0.01 degree, and the measurement accuracy of the positional deviation is ±0.01 millimeters.

[0043] The pressure sensor is a resistance strain gauge type, installed at the connection between the ball flange and the mounted equipment. The pressure sensor can measure the weight of the mounted equipment in real time, with a measurement range of 0 kg to 50 kg and a measurement accuracy of ±0.01 kg.

[0044] The wind speed sensor is an ultrasonic anemometer, mounted on the upper part of the firefighting drone's fuselage, away from the airflow interference area caused by the propeller. The ultrasonic anemometer can simultaneously measure wind speed and direction, with a measurement range of 0 meters per second to 20 meters per second. The wind speed measurement accuracy is ±0.1 meters per second, and the wind direction measurement accuracy is ±1 degree.

[0045] All data collected by the sensors is transmitted to a computer via a wireless communication module, where the computer processes and analyzes the data.

[0046] In some embodiments, constructing an adaptive control model for the output force and propulsion speed of the electric push rod includes: collecting historical data of successful docking under different docking scenarios, labeling the output force and propulsion speed of the electric push rod corresponding to the historical data, and training the adaptive control model using a neural network algorithm.

[0047] This embodiment details the specific implementation of the adaptive control model for the output force and propulsion speed of the electric actuator in step S101. Successful docking historical data is collected under different docking scenarios. These scenarios include static docking and aerial emergency docking. Static docking refers to docking operations performed by the firefighting drone while it is stationary on the ground. Aerial emergency docking refers to docking operations performed by the firefighting drone while it is in flight. In each docking scenario, multiple docking experiments are conducted by changing parameters such as attitude deviation, payload weight, and environmental wind disturbance. The attitude deviation data, payload weight data, environmental wind disturbance data, and the corresponding electric actuator output force and propulsion speed data are recorded during each successful docking. At least ten thousand sets of successful docking historical data are collected in total.

[0048] The output force and advance speed of the electric actuator corresponding to the historical data are marked. For each set of historical data, the output force and advance speed of the electric actuator corresponding to three different stages are marked: the approach stage, the mating stage, and the locking stage. The approach stage refers to the stage where the distance between the ball socket and the ball socket flange is greater than 100 mm. The mating stage refers to the stage where the distance between the ball socket and the ball socket flange is between 0 mm and 100 mm. The locking stage refers to the stage after the ball socket and the ball socket flange are fully engaged.

[0049] An adaptive control model was trained using a neural network algorithm. The neural network adopted a three-layer feedforward neural network structure, including an input layer, a hidden layer, and an output layer. The input layer has three neurons, corresponding to attitude deviation data, load weight data, and environmental wind disturbance data, respectively. The hidden layer has twenty neurons, using the ReLU activation function. The output layer has four neurons, corresponding to the output force and propulsion speed of the two electric actuators during the docking phase, respectively.

[0050] The training process employs the backpropagation algorithm, dividing the collected historical data into a training set and a validation set, with the training set comprising 80% and the validation set comprising 20%. During training, the weights and biases of the neural network are continuously adjusted to minimize the model's prediction error. Training is complete when the model's prediction error on the validation set is less than 1%, yielding the final adaptive control model.

[0051] In some embodiments, the step of inputting the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state includes: filtering and preprocessing the input environmental data to remove noise, inputting the preprocessed environmental data into the adaptive control model, and calculating the electric push rod drive parameters corresponding to the approach phase, docking phase, and locking phase, respectively.

[0052] This embodiment details the specific implementation of step S102, where environmental data is input into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state. Noise is removed by filtering the input environmental data. A moving average filtering algorithm is used to filter attitude deviation data, load weight data, and environmental wind disturbance data. The sliding window size is set to ten data points. For each data point, the average of the nine data points before and after it is taken as the filtered value for that data point. This filtering algorithm can effectively remove high-frequency noise from the sensor-acquired data, improving the smoothness and stability of the data.

[0053] The preprocessed environmental data is input into the adaptive control model. The model first determines the current docking stage based on the input environmental data: approach stage, docking stage, or locking stage. Then, according to the different stages, it calls the corresponding model parameters to calculate the optimal output force and propulsion speed parameters for each of the two electric actuators at that stage.

[0054] The electric actuator drive parameters for the approach, docking, and locking phases were calculated separately. During the approach phase, the electric actuator's advance speed was set to a higher value to shorten the docking time, and the output force was set to the minimum value sufficient to overcome the electric actuator's own friction and the weight of the ball and socket. During the docking phase, the electric actuator's advance speed was set to a lower value to achieve impact-free docking, and the output force was adjusted based on attitude deviation, load weight, and environmental wind disturbance to ensure smooth mating between the ball and socket and the flange. During the locking phase, the electric actuator's advance speed was set to zero, and the output force was set to the preset locking force to ensure a reliable connection between the ball and socket and the flange.

[0055] In some embodiments, adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters includes: adjusting the output force and propulsion speed of each electric actuator separately, collecting the actual output force and actual propulsion speed of each electric actuator in real time, comparing the actual parameters with the drive parameters, and correcting the control signal of the electric actuator based on the comparison result.

[0056] This embodiment provides a detailed description of the specific implementation method for adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters in step S102. Adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters includes adjusting the output force and propulsion speed of each electric actuator separately, collecting the actual output force and actual propulsion speed of each electric actuator in real time, comparing the actual parameters with the drive parameters, and correcting the control signal of the electric actuator based on the comparison results.

[0057] The output force and propulsion speed of each electric actuator are adjusted independently. The two electric actuators are controlled independently, each with its own driver and control circuit. The computer sends control signals to the drivers of both electric actuators to control the output force and propulsion speed of each actuator. This independent control method allows for the adjustment of their parameters based on the actual operating conditions of each electric actuator, improving the accuracy and reliability of the docking process.

[0058] The system collects real-time data on the actual output force and propulsion speed of each electric actuator. Each actuator integrates a force sensor and a speed sensor. The force sensor measures the actual output force of the actuator in real-time with an accuracy of ±1 Newton. The speed sensor measures the actual propulsion speed of the actuator in real-time with an accuracy of ±0.1 millimeter per second. The data collected by the force and speed sensors is transmitted to a computer via wired communication.

[0059] The actual parameters are compared with the drive parameters, and the control signal of the electric linear actuator is adjusted based on the comparison results. The computer equipment compares the actual output force and actual propulsion speed of the electric linear actuator collected with the drive parameters calculated by the model, and calculates the difference between the actual parameters and the drive parameters. If the difference is within the allowable range, the control signal is not adjusted. If the difference exceeds the allowable range, a proportional-integral-derivative (PID) control algorithm is used to calculate the correction amount of the control signal, which is sent to the actuator of the electric linear actuator to adjust the output force and propulsion speed of the electric linear actuator to keep the actual parameters consistent with the drive parameters.

[0060] In some embodiments, controlling the two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking includes: using a master-slave synchronous control strategy to control the synchronous movement of the two electric push rods, detecting the distance between the ball socket and the ball socket flange in real time, and reducing the pushing speed of the electric push rods when the distance is less than a preset threshold.

[0061] This embodiment details the specific implementation of step S103, which involves controlling two electric actuators to synchronously push the ball socket and flange together to achieve a precise, impact-free connection. A master-slave synchronous control strategy is employed to control the synchronous movement of the two electric actuators by designating one as the master actuator and the other as the slave actuator. The master actuator operates according to the drive parameters calculated from the model. The slave actuator collects the actual output force and actual propulsion speed of the master actuator in real time and uses these parameters as its own target parameters, following the master actuator's movement. This master-slave synchronous control strategy ensures that the movements of the two electric actuators remain highly synchronized, with a synchronization error of less than 0.1 millimeters.

[0062] Real-time detection of the distance between the ball socket and the ball socket flange. The distance is calculated in real-time using a binocular vision sensor. The measurement accuracy is ±0.1 mm.

[0063] When the distance is less than a preset threshold, the advancement speed of the electric actuator is reduced. The preset threshold is set to 50 mm. When the distance between the ball socket and the ball socket flange is greater than 50 mm, the electric actuator operates at the advancement speed of the approach phase. When the distance between the ball socket and the ball socket flange is less than or equal to 50 mm, the advancement speed of the electric actuator is reduced to one-fifth of its original speed, entering the docking phase. When the distance between the ball socket and the ball socket flange is less than or equal to 5 mm, the advancement speed of the electric actuator is further reduced to one-tenth of its original speed to achieve impact-free docking.

[0064] In some embodiments, after the docking is completed, the electric push rod is controlled to maintain a preset locking force to achieve smooth locking, including: detecting a docking completion signal, switching the electric push rod to the locking mode after receiving the docking completion signal, continuously monitoring the output locking force of the electric push rod, and increasing the output force of the electric push rod when the output locking force is lower than the preset locking force.

[0065] This embodiment provides a detailed description of the specific implementation method for controlling the electric push rod to maintain a preset locking force to achieve smooth locking after docking in step S103. The docking completion signal is detected by a contact sensor installed inside the ball socket. When the ball socket and its flange are fully engaged, the contact sensor is triggered, generating a docking completion signal. This signal is transmitted to a computer via wired communication.

[0066] Upon receiving the docking completion signal, the electric actuator is switched to locking mode. After receiving the docking completion signal, the computer equipment immediately stops the electric actuator's advance and switches it to locking mode. In locking mode, the electric actuator's advance speed is set to zero, and the output force is set to the preset locking force.

[0067] Continuously monitor the output locking force of the electric linear actuator. The output locking force of the electric linear actuator is monitored in real time via a force sensor inside the actuator. The monitoring frequency is set to ten times per second.

[0068] When the output locking force is lower than the preset locking force, the output force of the electric actuator is increased. The preset locking force is set based on the load weight and is set to three times the load weight. If the output locking force of the electric actuator is detected to be lower than the preset locking force, the computer device will automatically increase the output force of the electric actuator until the output locking force reaches the preset value. If the output locking force remains lower than the preset locking force for more than five seconds, the computer device will issue a locking abnormality alarm signal.

[0069] In some embodiments, the method further includes: identifying the current docking scenario, selecting a static docking mode or an aerial emergency docking mode based on the identification result, and inputting a preset control parameter set in the corresponding mode into the adaptive control model.

[0070] This embodiment describes in detail an additional step of the method of the present invention. The current docking scenario is identified by using the flight status data of the firefighting drone. The flight status data includes the drone's altitude, speed, acceleration, and attitude angle. If the drone's altitude is less than one meter, its speed is less than 0.1 meters per second, and its acceleration is less than 0.1 meters per square second, it is determined to be a static docking scenario. Otherwise, it is determined to be an aerial emergency docking scenario.

[0071] Select either static docking mode or aerial emergency docking mode based on the identification results. Static docking mode is suitable for ground-based docking operations, requiring high docking accuracy and offering relatively flexible docking time requirements. Aerial emergency docking mode is suitable for aerial docking operations, requiring short docking time and high docking stability.

[0072] The preset control parameter sets for the corresponding modes are input into the adaptive control model. Different preset control parameter sets are pre-set for static docking mode and aerial emergency docking mode. The preset control parameter sets include parameters such as the approach phase advance speed, the docking phase advance speed, and the locking force. In static docking mode, the approach phase advance speed is set to 50 mm / s, the docking phase advance speed is set to 5 mm / s, and the locking force is set to three times the load weight. In aerial emergency docking mode, the approach phase advance speed is set to 100 mm / s, the docking phase advance speed is set to 10 mm / s, and the locking force is set to four times the load weight. The preset control parameter sets for the corresponding modes are input into the adaptive control model, which combines the preset control parameter sets with real-time environmental data to calculate the optimal electric actuator drive parameters.

[0073] In some embodiments, the method further includes: calculating the probability of structural collision during the docking process in real time; when the probability of structural collision exceeds a preset threshold, reducing the advancing speed of the electric push rod and adjusting the output force difference between the two electric push rods to correct the attitude of the docking surface.

[0074] This embodiment details another additional step of the method of the present invention. The probability of structural collision during the docking process is calculated in real time. Based on the current attitude deviation data, the advancement speed of the electric push rod, and the output force data, a probabilistic statistical method is used to calculate the probability of a collision between the ball socket and the ball socket flange. The collision probability calculation frequency is set to five times per second.

[0075] When the probability of a structural collision exceeds a preset threshold, the advancing speed of the electric actuator is reduced. The preset threshold is set to 30%. If the calculated probability of a structural collision is greater than or equal to 30%, the advancing speed of the electric actuator is immediately reduced to half of its original speed. If the probability of a structural collision is greater than or equal to 50%, the advancing movement of the electric actuator is immediately stopped.

[0076] Adjust the output force difference between the two electric actuators to correct the docking face attitude. If the attitude deviation data shows the ball socket tilting to the left, increase the output force of the left electric actuator and decrease the output force of the right electric actuator to rotate the ball socket to the right, correcting the attitude deviation. If the attitude deviation data shows the ball socket tilting to the right, increase the output force of the right electric actuator and decrease the output force of the left electric actuator to rotate the ball socket to the left, correcting the attitude deviation. The magnitude of the output force difference is adjusted according to the magnitude of the attitude deviation; the larger the attitude deviation, the larger the output force difference. Once the attitude deviation is corrected to within the allowable range, restore the output forces of both electric actuators to their original values ​​and continue the docking operation.

[0077] In some embodiments, the method further includes: detecting whether the docking is successful; when the docking is unsuccessful, automatically adjusting the attitude of the docking surface, re-executing the docking process, recording the number of docking retries, and issuing a docking failure alarm signal when the number of docking retries reaches a preset maximum.

[0078] This embodiment describes in detail another additional step of the method of the present invention. The docking is checked for success. After docking, the success of the docking is determined by the signal from the contact sensor and the output locking force data of the electric actuator. If the contact sensor is triggered and the output locking force of the electric actuator reaches the preset locking force, the docking is considered successful. Otherwise, the docking is considered unsuccessful.

[0079] If docking fails, the attitude of the docking surface is automatically adjusted, and the docking process is repeated. If docking fails, the computer equipment will control the two electric push rods to retract to their initial positions. Then, based on the attitude deviation data recorded during the docking process, the attitude of the fire-fighting drone is adjusted to align the axis of the ball socket with the axis of the ball socket flange. After the attitude adjustment is completed, the docking process from steps S101 to S103 is repeated.

[0080] Record the number of connection retries. Each time the connection process is re-executed, the connection retrieval count is incremented by one. The initial value of the connection retrieval count is zero.

[0081] When the number of docking retries reaches the preset maximum, a docking failure alarm signal will be issued. The preset maximum number of retries is set to three. If docking still fails after three retries, the computer equipment will issue a docking failure alarm signal and send the docking failure information to the ground control station of the firefighting drone for manual intervention by ground operators.

[0082] In some embodiments, for extreme scenarios where visual sensors fail due to smoke interference in a fire, a method for acquiring attitude deviation data by multi-sensor fusion is provided. The method in this embodiment further includes: detecting the smoke concentration in the docking area in real time using a smoke concentration sensor, automatically switching the main sensor for acquiring attitude deviation data according to the smoke concentration, and using a Kalman filter algorithm to fuse the data from multiple sensors to obtain the final docking surface attitude deviation data.

[0083] The smoke concentration sensor is an optical smoke concentration sensor, installed above the middle connector, adjacent to the binocular vision sensor. It can detect the smoke concentration in the docking area in real time, with a measurement range of 0 mg / m³ to 1000 mg / m³ and a measurement accuracy of ±10 mg / m³.

[0084] Attitude deviation data acquisition employs a multi-sensor fusion scheme utilizing binocular vision sensors, infrared thermal imaging sensors, and lidar sensors. The infrared thermal imaging sensors and lidar sensors are mounted side-by-side with the binocular vision sensors, facing the ball socket and flange. The infrared thermal imaging sensor can penetrate low-concentration smoke to capture thermal radiation images of the ball socket and flange, with a measurement accuracy of ±0.5 degrees for angular deviation and ±0.5 millimeters for positional deviation. The lidar sensor can penetrate high-concentration smoke, measuring the distance and position of the ball socket and flange by emitting a laser beam, with a measurement accuracy of ±1 degree for angular deviation and ±1 millimeter for positional deviation.

[0085] The specific method for automatically switching the main sensor based on smoke concentration is as follows: when the smoke concentration is less than 50 mg / m³, the binocular vision sensor is used as the main sensor, and the infrared thermal imaging sensor and the lidar sensor are used as auxiliary sensors; when the smoke concentration is greater than or equal to 50 mg / m³ and less than 500 mg / m³, the infrared thermal imaging sensor is used as the main sensor, and the binocular vision sensor and the lidar sensor are used as auxiliary sensors; when the smoke concentration is greater than or equal to 500 mg / m³, the lidar sensor is used as the main sensor, the infrared thermal imaging sensor is used as the auxiliary sensor, and the binocular vision sensor stops working.

[0086] The specific method for fusing data from multiple sensors using the Kalman filter algorithm is as follows: State equations and observation equations for the attitude deviation data are established. The observation data from the main sensor is used as the primary input, and the observation data from the auxiliary sensors is used as correction input. Through prediction and updating steps, the estimated values ​​of the attitude deviation data are continuously optimized, ultimately yielding smooth and accurate docking surface attitude deviation data. The measurement accuracy of the fused attitude deviation data is no less than that of the main sensor, while effectively suppressing noise and interference from individual sensors.

[0087] In some embodiments, for scenarios where the load changes dynamically during docking, an online incremental update method for an adaptive control model is provided, which solves the problem that offline trained models cannot adapt to dynamic load changes.

[0088] The method in this embodiment also includes: collecting load weight change data and corresponding electric actuator actual operation data in real time during the docking process; when the load weight change exceeds a preset threshold, using an incremental learning algorithm to update the adaptive control model online; and using the new model to calculate the electric actuator drive parameters after the update is completed.

[0089] The system collects real-time data on load weight changes and corresponding actual operating data of the electric actuator. A pressure sensor mounted on the ball joint flange collects load weight data at a frequency of ten times per second. Simultaneously, force and speed sensors inside the electric actuator collect data on the actual output force and propulsion speed of the actuator at the same frequency. The load weight data, attitude deviation data, environmental wind disturbance data, and the corresponding actual output force and propulsion speed data of the electric actuator at the same moment are combined into a set of online sample data and stored in the local memory of the computer device.

[0090] A model update is triggered when the change in load weight exceeds a preset threshold. The preset threshold is set to 5% of the initial load weight. The difference between the current load weight and the load weight at the time of the last model update is calculated. If the absolute value of the difference is greater than or equal to the preset threshold, the online update process of the adaptive control model is triggered.

[0091] An incremental learning algorithm is used to update the adaptive control model online. This algorithm employs incremental backpropagation, eliminating the need to retrain the model using all historical data. Instead, it adjusts only some weights and biases using recently collected online sample data. Specifically, the update process involves using one hundred recently collected online sample data sets as the incremental training set, inputting them into the existing adaptive control model, calculating the model's prediction error, and then adjusting the weights and biases of the output layer and the last hidden layer based on this error to minimize the prediction error. This update process runs in the background and does not affect the normal operation of the current integration process.

[0092] After the update is complete, the new model is used to calculate the electric actuator drive parameters. Immediately after the model update, the new model parameters are loaded into the runtime memory, and subsequent electric actuator drive parameter calculations will use the updated model. Simultaneously, the updated model parameters are saved to local storage for use during the next docking.

[0093] In some embodiments, for scenarios involving collaborative operations of multiple drones, a distributed synchronous docking control method for a multi-drone ball-and-socket quick-release mechanism is provided, which enables precise collaborative docking between multiple firefighting drones.

[0094] The method in this embodiment also includes: establishing a multi-UAV distributed communication network, using a precise time protocol to achieve time synchronization between multiple UAVs, using a distributed consensus algorithm to achieve attitude synchronization and electric push rod parameter synchronization between multiple UAVs, and controlling the ball-and-socket quick-release mechanism of multiple UAVs to simultaneously complete docking and locking.

[0095] Establish a distributed communication network for multiple drones. Each firefighting drone is equipped with a wireless communication module, and a self-organizing network protocol is used to establish the distributed communication network. Each drone in the network can communicate directly with other drones without needing to go through a central node for forwarding. The communication distance is no less than one kilometer, and the communication latency is no more than ten milliseconds.

[0096] A precise time protocol is used to achieve time synchronization among multiple drones. One drone is selected as the master clock node, and the others are slave clock nodes. The master clock node periodically sends time synchronization messages to all slave clock nodes. The slave clock nodes adjust their local clocks according to the received time synchronization messages to keep the local clocks of all drones synchronized. The time synchronization accuracy is no more than one millisecond.

[0097] Attitude synchronization and motorized actuator parameter synchronization among multiple UAVs are achieved through a distributed consensus algorithm. Each UAV broadcasts its current attitude data, environmental data, and calculated motorized actuator parameters to all other UAVs via a distributed communication network. After receiving the data from other UAVs, each UAV uses the distributed consensus algorithm to calculate the globally consistent target attitude and motorized actuator parameters. Then, each UAV adjusts its own attitude and motorized actuator parameters to align with the globally consistent target values.

[0098] The ball-and-socket quick-release mechanism controls multiple drones to simultaneously complete docking and locking. Once the attitude and electric actuator parameters of all drones reach globally consistent target values, the master clock node issues a unified docking start command. Upon receiving the docking start command, all drones simultaneously control their electric actuators to push the ball sockets to dock with the ball socket flanges. After docking is complete, all drones simultaneously switch their electric actuators to locking mode, maintaining the preset locking force.

[0099] In some embodiments, a dynamic temperature compensation method for electric actuator parameters is provided for extreme high-temperature environments in fire scenes, which solves the problem of decreased control accuracy caused by changes in the performance of electric actuators under high-temperature environments.

[0100] The method in this embodiment further includes: installing temperature sensors at the motor, gearbox, and lead screw of each electric actuator to collect temperature data of key parts of the electric actuator in real time, and dynamically compensating the output force and propulsion speed of the electric actuator according to the pre-established temperature-parameter mapping relationship.

[0101] The temperature sensors utilize platinum resistance temperature sensors, characterized by high temperature resistance and high accuracy. The motor temperature sensor is mounted on the surface of the motor stator, the gearbox temperature sensor is mounted on the gearbox housing, and the leadscrew temperature sensor is mounted on the surface of the leadscrew nut. The temperature sensors have a measurement range of -40°C to 300°C, with a measurement accuracy of ±0.5°C. Temperature data is acquired five times per second.

[0102] A temperature-parameter mapping relationship is established beforehand. In a laboratory environment, different temperature conditions are simulated, and the deviations between the actual output force and actual propulsion speed of the electric linear actuator at different temperatures and the theoretical values ​​are measured. Mapping tables are established between motor temperature, gearbox temperature, and lead screw temperature and the deviations in output force and propulsion speed of the electric linear actuator, respectively, and these mapping tables are stored in the local memory of the computer.

[0103] The specific method for dynamically compensating the output force and propulsion speed of the electric linear actuator is as follows: Temperature data of key components of the electric linear actuator is collected in real time. Based on the temperature data, a temperature-parameter mapping table is consulted to obtain the output force deviation and propulsion speed deviation of the electric linear actuator at the current temperature. Then, the corresponding deviation values ​​are subtracted from the theoretical output force and theoretical propulsion speed calculated by the model to obtain the compensated drive parameters of the electric linear actuator. Finally, the operation of the electric linear actuator is controlled according to the compensated drive parameters.

[0104] Dynamic compensation can effectively offset the impact of high-temperature environment on the performance of electric actuator, keeping the deviation between the actual output force and actual propulsion speed of the electric actuator and the theoretical value within three percent, thus ensuring the accuracy and reliability of the docking process.

[0105] In some embodiments, a fault-tolerant docking control method is provided for scenarios where a single electric linear actuator fails, which solves the problem of docking failure caused by a single electric linear actuator failure and significantly improves the reliability of the system.

[0106] The method in this embodiment also includes: real-time monitoring of the operating status of the two electric actuators; when a fault is detected in one of the electric actuators, automatically switching to a single actuator fault-tolerant control mode; and adjusting the output force and trajectory of the remaining normal electric actuators, combined with the attitude adjustment of the fire-fighting drone itself, to complete docking and locking.

[0107] The operation status of the two electric linear actuators is monitored in real time. Monitoring parameters include the operating current, output force, propulsion speed, and communication status of the electric linear actuators. Under normal operation, the operating current of the electric linear actuators is between 50% and 120% of the rated current, the deviation of the output force and propulsion speed from the drive parameters is within the allowable range, and the communication status is normal.

[0108] The electric linear actuator is considered to be malfunctioning if any of the following conditions are detected: the operating current of the electric linear actuator exceeds 150% of the rated current for more than one second; the output force of the electric linear actuator is zero for more than one second; the advance speed of the electric linear actuator is zero for more than one second; or the status feedback data of the electric linear actuator cannot be received three times consecutively.

[0109] Automatically switches to single-actuator fault-tolerant control mode. When a fault is detected in one of the electric actuators, power supply to the faulty actuator is immediately stopped, and it is locked in its current position. At the same time, the maximum output force of the remaining normal electric actuators is increased to 150% of their rated output force to meet the force requirements of single-actuator docking.

[0110] The docking is completed by adjusting the output force and trajectory of the remaining normal electric push rods, combined with the attitude adjustment of the firefighting drone itself. Specifically, the motion of the ball socket is decomposed into translational motion along the docking axis and rotational motion around the docking axis. The remaining normal electric push rods provide thrust along the docking axis and simultaneously generate torque around the docking axis by changing the magnitude of the output force. The firefighting drone corrects the attitude deviation of the ball socket by adjusting its roll and pitch angles in conjunction with the torque of the electric push rods. Once the ball socket is fully aligned with the ball socket flange, the normal electric push rods are controlled to push the ball socket to complete the docking. After docking, the normal electric push rods are controlled to maintain twice the preset locking force to ensure reliable locking.

[0111] In some embodiments, to address the issue of limited battery capacity in firefighting drones, an energy optimization adaptive scheduling method for the docking process is provided, which minimizes energy consumption during the docking process while ensuring docking success rate and docking time.

[0112] The method in this embodiment also includes: establishing an energy consumption model for the docking process, and adaptively scheduling the advancement speed and output force of the electric push rod based on the current remaining battery power, docking distance, environmental wind disturbance and load weight, thereby optimizing the energy consumption of the docking process.

[0113] An energy consumption model for the docking process was established. The energy consumption during docking mainly includes the kinetic energy consumption of the electric actuator, the energy consumption to overcome wind resistance, and the energy consumption to maintain attitude. Through extensive experimental data, the relationships between the electric actuator's propulsion speed and kinetic energy consumption, the ambient wind speed and energy consumption to overcome wind resistance, and the load weight and energy consumption to maintain attitude were established. Finally, a model was obtained showing the relationship between the total energy consumption of the docking process and the electric actuator's propulsion speed, ambient wind speed, load weight, and docking distance.

[0114] Based on the current remaining battery power, docking distance, environmental wind disturbance, and load weight, the advancement speed and output force of the electric actuator are adaptively scheduled. Specifically, the scheduling strategy is as follows: when the remaining battery power is greater than 50%, priority is given to ensuring docking time, using a higher electric actuator advancement speed; when the remaining battery power is between 20% and 50%, a balance is struck between docking time and energy consumption, using a medium electric actuator advancement speed; when the remaining battery power is less than 20%, priority is given to ensuring energy consumption, using a lower electric actuator advancement speed. Simultaneously, the output force of the electric actuator is dynamically adjusted according to the magnitude of environmental wind disturbance and load weight, ensuring docking stability while avoiding energy waste caused by excessive output force.

[0115] The energy optimization adaptive scheduling method can reduce energy consumption during the docking process by 15% to 30% while ensuring a docking success rate of no less than 99%, effectively extending the endurance of firefighting drones and improving their operational capabilities in fire scenes.

[0116] Please see Figure 3 As shown, Figure 3 This is a schematic diagram of the structure of the adaptive precision control system 200 for the docking process of the quick-disassembly mechanism of the fire-fighting drone provided in this embodiment. The adaptive precision control system 200 is used to execute the steps of the adaptive precision control method for the docking process of the quick-disassembly mechanism of the fire-fighting drone shown in the above embodiments. The adaptive precision control system 200 can be a single server or a server cluster, or it can be a terminal, such as a handheld terminal, a laptop computer, a wearable device, or a robot.

[0117] like Figure 3 As shown, the adaptive precision control system 200 for the quick-disassembly mechanism docking process of the firefighting drone includes: The data acquisition unit 201 is used to acquire environmental data of the docking surface, including attitude deviation data, load weight data and environmental wind disturbance data; and to construct an adaptive control model for the output force and propulsion speed of the electric push rod.

[0118] The parameter input unit 202 is used to input the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state; and to adjust the output force and propulsion speed of the two electric push rods in real time according to the electric push rod drive parameters.

[0119] The docking control unit 203 is used to control two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking; after docking, the electric push rods are controlled to maintain the preset locking force to achieve smooth locking.

[0120] In some embodiments, acquiring environmental data of the docking surface includes: acquiring docking surface attitude deviation data through a visual sensor, acquiring load weight data through a pressure sensor, and acquiring environmental wind disturbance data through a wind speed sensor.

[0121] In some embodiments, constructing an adaptive control model for the output force and propulsion speed of the electric push rod includes: collecting historical data of successful docking under different docking scenarios, labeling the output force and propulsion speed of the electric push rod corresponding to the historical data, and training the adaptive control model using a neural network algorithm.

[0122] In some embodiments, the step of inputting the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state includes: filtering and preprocessing the input environmental data to remove noise, inputting the preprocessed environmental data into the adaptive control model, and calculating the electric push rod drive parameters corresponding to the approach phase, docking phase, and locking phase, respectively.

[0123] In some embodiments, adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters includes: adjusting the output force and propulsion speed of each electric actuator separately, collecting the actual output force and actual propulsion speed of each electric actuator in real time, comparing the actual parameters with the drive parameters, and correcting the control signal of the electric actuator based on the comparison result.

[0124] In some embodiments, controlling the two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking includes: using a master-slave synchronous control strategy to control the synchronous movement of the two electric push rods, detecting the distance between the ball socket and the ball socket flange in real time, and reducing the pushing speed of the electric push rods when the distance is less than a preset threshold.

[0125] In some embodiments, after the docking is completed, the electric push rod is controlled to maintain a preset locking force to achieve smooth locking, including: detecting a docking completion signal, switching the electric push rod to the locking mode after receiving the docking completion signal, continuously monitoring the output locking force of the electric push rod, and increasing the output force of the electric push rod when the output locking force is lower than the preset locking force.

[0126] In some embodiments, the method further includes: identifying the current docking scenario, selecting a static docking mode or an aerial emergency docking mode based on the identification result, and inputting a preset control parameter set in the corresponding mode into the adaptive control model.

[0127] In some embodiments, the method further includes: calculating the probability of structural collision during the docking process in real time; when the probability of structural collision exceeds a preset threshold, reducing the advancing speed of the electric push rod and adjusting the output force difference between the two electric push rods to correct the attitude of the docking surface.

[0128] In some embodiments, the method further includes: detecting whether the docking is successful; when the docking is unsuccessful, automatically adjusting the attitude of the docking surface, re-executing the docking process, recording the number of docking retries, and issuing a docking failure alarm signal when the number of docking retries reaches a preset maximum.

[0129] It should be noted that those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the adaptive precision control system and its modules for the quick-disassembly mechanism docking process of the fire-fighting drone described above can be found in the corresponding embodiments of the adaptive precision control method for the quick-disassembly mechanism docking process of the fire-fighting drone, and will not be repeated here.

[0130] The aforementioned adaptive and precise control method for the quick-disassembly mechanism docking process of firefighting drones can be implemented as a computer program, which can, for example... Figure 3 It runs on the device shown.

[0131] Please see Figure 4 , Figure 4 This is a schematic block diagram of the structure of a computer device provided in an embodiment of this application. The computer device includes a processor, a memory, and a network interface connected via a device bus, wherein the memory may include a storage medium and internal memory.

[0132] The storage medium can store operating devices and computer programs. The computer program includes program instructions that, when executed, cause the processor to perform an adaptive and precise control method for the docking process of any quick-release mechanism of a fire-fighting drone.

[0133] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0134] The internal memory provides an environment for the execution of computer programs in non-volatile storage media. When the computer program is executed by the processor, it enables the processor to execute any adaptive and precise control method for the docking process of the quick-release mechanism of a fire-fighting drone.

[0135] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the terminal to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0137] In one embodiment, the processor is configured to run a computer program stored in memory to perform the following steps: Acquire environmental data of the docking surface, including attitude deviation data, load weight data, and environmental wind disturbance data; construct an adaptive control model for the output force and propulsion speed of the electric actuator.

[0138] The environmental data is input into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state; based on the electric actuator drive parameters, the output force and propulsion speed of the two electric actuators are adjusted in real time.

[0139] Two electric push rods are controlled to simultaneously push the ball socket and the ball socket flange to complete a precise, impact-free connection; after the connection is completed, the electric push rods are controlled to maintain the preset locking force to achieve smooth locking.

[0140] In some embodiments, acquiring environmental data of the docking surface includes: acquiring docking surface attitude deviation data through a visual sensor, acquiring load weight data through a pressure sensor, and acquiring environmental wind disturbance data through a wind speed sensor.

[0141] In some embodiments, constructing an adaptive control model for the output force and propulsion speed of the electric push rod includes: collecting historical data of successful docking under different docking scenarios, labeling the output force and propulsion speed of the electric push rod corresponding to the historical data, and training the adaptive control model using a neural network algorithm.

[0142] In some embodiments, the step of inputting the environmental data into the adaptive control model to calculate the electric push rod drive parameters that match the current docking state includes: filtering and preprocessing the input environmental data to remove noise, inputting the preprocessed environmental data into the adaptive control model, and calculating the electric push rod drive parameters corresponding to the approach phase, docking phase, and locking phase, respectively.

[0143] In some embodiments, adjusting the output force and propulsion speed of the two electric actuators in real time according to the electric actuator drive parameters includes: adjusting the output force and propulsion speed of each electric actuator separately, collecting the actual output force and actual propulsion speed of each electric actuator in real time, comparing the actual parameters with the drive parameters, and correcting the control signal of the electric actuator based on the comparison result.

[0144] In some embodiments, controlling the two electric push rods to synchronously push the ball socket and the ball socket flange to complete the impact-free and precise docking includes: using a master-slave synchronous control strategy to control the synchronous movement of the two electric push rods, detecting the distance between the ball socket and the ball socket flange in real time, and reducing the pushing speed of the electric push rods when the distance is less than a preset threshold.

[0145] In some embodiments, after the docking is completed, the electric push rod is controlled to maintain a preset locking force to achieve smooth locking, including: detecting a docking completion signal, switching the electric push rod to the locking mode after receiving the docking completion signal, continuously monitoring the output locking force of the electric push rod, and increasing the output force of the electric push rod when the output locking force is lower than the preset locking force.

[0146] In some embodiments, the method further includes: identifying the current docking scenario, selecting a static docking mode or an aerial emergency docking mode based on the identification result, and inputting a preset control parameter set in the corresponding mode into the adaptive control model.

[0147] In some embodiments, the method further includes: calculating the probability of structural collision during the docking process in real time; when the probability of structural collision exceeds a preset threshold, reducing the advancing speed of the electric push rod and adjusting the output force difference between the two electric push rods to correct the attitude of the docking surface.

[0148] In some embodiments, the method further includes: detecting whether the docking is successful; when the docking is unsuccessful, automatically adjusting the attitude of the docking surface, re-executing the docking process, recording the number of docking retries, and issuing a docking failure alarm signal when the number of docking retries reaches a preset maximum.

[0149] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to implement the steps of the adaptive and precise control method for the docking process of the quick-release mechanism of a fire-fighting drone as provided in any embodiment of this application.

[0150] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiments, such as the hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, SmartMedia Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the computer device.

[0151] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An adaptive and precise control method for the docking process of a quick-release mechanism for a fire-fighting drone, wherein the quick-release mechanism for the fire-fighting drone includes a ball-and-socket type double-ended docking quick-release mechanism, the ball-and-socket type double-ended docking quick-release mechanism being provided with two electric push rods, a ball socket, and a ball socket flange; characterized in that, include: Acquire environmental data of the docking surface, including attitude deviation data, load weight data, and environmental wind disturbance data; construct an adaptive control model for the output force and propulsion speed of the electric push rod; The environmental data is input into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state; based on the electric actuator drive parameters, the output force and propulsion speed of the two electric actuators are adjusted in real time. Two electric push rods are controlled to simultaneously push the ball socket and the ball socket flange to complete a precise, impact-free connection; after the connection is completed, the electric push rods are controlled to maintain the preset locking force to achieve smooth locking.

2. The method according to claim 1, characterized in that, The acquisition of environmental data of the docking surface includes: Visual sensors acquire docking surface attitude deviation data, pressure sensors acquire load weight data, and wind speed sensors acquire environmental wind disturbance data.

3. The method according to claim 1, characterized in that, The adaptive control model for the output force and propulsion speed of the electric actuator includes: Collect historical data of successful docking under different docking scenarios, label the output force and propulsion speed of the electric push rod corresponding to the historical data, and use neural network algorithm to train an adaptive control model.

4. The method according to claim 1, characterized in that, The step of inputting the environmental data into the adaptive control model to calculate the electric actuator drive parameters that match the current docking state includes: The input environmental data is filtered and preprocessed to remove noise. The preprocessed environmental data is then input into the adaptive control model to calculate the electric actuator drive parameters for the approach, docking, and locking phases, respectively.

5. The method according to claim 1, characterized in that, The step of adjusting the output force and propulsion speed of the two electric push rods in real time according to the electric push rod drive parameters includes: The output force and propulsion speed of each electric actuator are adjusted separately. The actual output force and actual propulsion speed of each electric actuator are collected in real time. The actual parameters are compared with the drive parameters, and the control signal of the electric actuator is corrected based on the comparison results.

6. The method according to claim 1, characterized in that, The control of two electric push rods to synchronously push the ball socket and the ball socket flange to complete a precise, impact-free connection includes: A master-slave synchronous control strategy is adopted to control the synchronous movement of two electric push rods. The distance between the ball socket and the ball socket flange is detected in real time. When the distance is less than a preset threshold, the advancing speed of the electric push rod is reduced.

7. The method according to claim 1, characterized in that, After the docking is completed, the electric push rod is controlled to maintain a preset locking force to achieve smooth locking, including: The system detects the docking signal and switches the electric push rod to the locking mode upon receiving the signal. It continuously monitors the output locking force of the electric push rod and increases the output force when the output locking force is lower than the preset locking force.

8. The method according to claim 1, characterized in that, The method further includes: Identify the current docking scenario, select either static docking mode or aerial emergency docking mode based on the identification result, and input the preset control parameter set of the corresponding mode into the adaptive control model.

9. The method according to claim 1, characterized in that, The method further includes: The probability of structural collision during the docking process is calculated in real time. When the probability of structural collision exceeds a preset threshold, the advancing speed of the electric push rod is reduced, and the output force difference between the two electric push rods is adjusted to correct the attitude of the docking surface.

10. The method according to claim 1, characterized in that, The method further includes: The system detects whether the docking was successful. If the docking fails, it automatically adjusts the attitude of the docking surface, re-executes the docking process, records the number of docking retries, and issues a docking failure alarm signal when the number of docking retries reaches the preset maximum.

Images