Emergency parachute control method for unmanned aerial vehicle, system and storage medium

Abstract

The application discloses an emergency parachute control method and system of a UAV and a storage medium, and relates to the technical field of UAVs. The method comprises the following steps: performing data analysis on initial sensing data of a target UAV to determine running state parameters and flight attitude information; when the flight attitude information is abnormal or the running state parameters are abnormal, determining whether the parachute is actively opened according to the flight attitude information, the running state parameters and parachute state information, obtaining an opening judgment result corresponding to an emergency parachute, and controlling the emergency parachute; when the flight attitude information is normal and the running state parameters are normal, identifying target abnormal data from the initial sensing data, identifying the flight state risk to determine the flight state risk level of the target UAV and determine display information, adjusting the display state corresponding to the target control key according to the display information to prompt the operator of the target UAV to operate the target control key in an emergency, and effectively reducing the loss caused by the crash of the UAV.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to an emergency parachute control method, system, and storage medium for a UAV. Background Technology

[0002] To ensure the safety of drones and minimize equipment wear and tear, modern drone technology commonly employs emergency parachute systems as a key protective measure. When an abnormal situation such as power failure, uncontrolled descent, or collision is detected, the drone automatically triggers parachute deployment at the optimal altitude. The deployed parachute canopy generates sufficient air resistance to slow the drone's descent, thus achieving a gentle landing. This effectively protects expensive onboard equipment, prevents secondary hazards such as battery explosions, significantly extends the drone's lifespan, and reduces maintenance costs, thereby minimizing drone wear and tear.

[0003] Currently, the parachute deployment triggering mechanisms for drone emergency parachute systems include manual and automatic triggering mechanisms, but both have significant limitations. In actual flight, abnormal situations are complex and varied, and automatic triggering mechanisms cannot cover all potential risks. In some abnormal situations, the parachute may not deploy automatically in time, potentially leading to accidents. Furthermore, existing automatic parachute triggering mechanisms are primarily used on multi-rotor drones. Helicopter drones, with their high-speed rotating main rotors, face greater technical challenges during parachute deployment: traditional parachutes, if deployed while the rotor is running, are prone to entanglement with the rotor blades, leading to broken parachute lines or loss of drone control. Helicopter drones typically fly at higher speeds, requiring more precise triggering and faster parachute deployment. Therefore, applying the automatic parachute deployment triggering logic from multi-rotor drones to helicopter drones fails to achieve effective deceleration or reduce potential damage. While some drones support remote manual parachute deployment by the pilot, in unexpected emergencies, operators may miss the optimal deployment time due to lack of experience, delayed reaction, or communication interference. This is especially true for drones flying at high speeds or operating at low altitudes, where even a delay of a few seconds can lead to a crash. Therefore, optimizing the emergency parachute control mechanism for drones is a pressing technical problem that needs to be solved. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes an emergency parachute control method, system, and storage medium for unmanned aerial vehicles (UAVs), which optimizes the emergency parachute control mechanism of UAVs and effectively reduces UAV crash damage.

[0005] In a first aspect, embodiments of this application provide an emergency parachute control method for a drone, applied to a controller of an emergency parachute control system for a drone. The emergency parachute control system further includes: an emergency parachute electrically connected to the controller and a remote control terminal; the remote control terminal includes a target control key for manually triggering the release of the emergency parachute; the emergency parachute control method for the drone includes:

[0006] During the flight of the target drone, the initial sensor data of the target drone is acquired in real time, and the initial sensor data is analyzed to determine the corresponding operating status parameters and flight attitude information of the target drone.

[0007] When the flight attitude information is abnormal or the operating status parameter is abnormal, the parachute is actively deployed based on the flight attitude information, the operating status parameter, and the parachute status information to obtain the deployment judgment result corresponding to the emergency parachute.

[0008] Based on the parachute deployment determination result, the emergency parachute is controlled to obtain the first control result corresponding to the emergency parachute;

[0009] When the flight attitude information is normal and the operating status parameters are normal, abnormal data identification processing is performed on the initial sensing data to obtain target abnormal data.

[0010] Based on the abnormal data of the target, flight status risk identification processing is performed to determine the flight status risk level of the target UAV;

[0011] The display information corresponding to the target control key in the remote control terminal is determined according to the flight status risk level, and the display information is sent to the remote control terminal so that the remote control terminal adjusts the display state corresponding to the target control key according to the display information, thereby prompting the operator of the target UAV to operate the target control key in an emergency.

[0012] Receive the operation result corresponding to the target control key, and obtain the second control result corresponding to the emergency parachute based on the operation result.

[0013] In some optional embodiments, the step of performing anomaly identification processing on the initial sensing data to obtain target anomaly data includes:

[0014] Window data under the target window is obtained by sliding the preset target window in the initial sensing data;

[0015] The window data is subjected to a first data calculation process to obtain interference data;

[0016] The initial sensing data is analyzed and processed using a Gaussian mixture model to obtain a first distribution result.

[0017] Based on the Gaussian mixture model, the interference data is analyzed for data distribution to obtain a second distribution result;

[0018] Based on the first distribution result and the second distribution result, a data comparison analysis is performed to determine the Gaussian distribution characteristics corresponding to the initial sensing data;

[0019] Based on the Gaussian distribution characteristics, the initial sensing data is analyzed to obtain the probability distribution corresponding to the initial sensing data;

[0020] Obtain the relevant sensing data corresponding to the maximum value of the probability distribution, and calculate the data difference between the relevant sensing data and the initial sensing data;

[0021] The target threshold is determined based on the data difference, and the initial sensor is re-assigned and compressed based on the comparison between the initial sensing data and the target threshold to obtain the target sensing data.

[0022] Density calculation is performed on each first sensing data in the target sensing data to obtain the first local density value corresponding to the first sensing data;

[0023] Obtain the reverse nearest neighbor data corresponding to the first sensing data from the target sensing data, and perform local density calculation on the reverse nearest neighbor data to obtain the second local density value;

[0024] The average density value is obtained by averaging all the second local density values, and the outlier characterization value corresponding to the first sensing data is determined based on the average density value and the first local density value.

[0025] The target anomaly data corresponding to the initial sensing data is determined based on the outlier characterization value and the target threshold.

[0026] In some optional embodiments, the first data calculation processing on the window data to obtain interference data includes:

[0027] The average data is obtained by calculating the mean of the window data;

[0028] Calculate the absolute difference between each sub-data point in the window data and the average data, and calculate the mean of all the absolute differences to obtain the deviation value;

[0029] Based on the deviation value, an interference value corresponding to the window data is generated;

[0030] The interference value is added to the window data to obtain the interference data.

[0031] In some optional embodiments, the step of performing density calculation on each first sensing data in the target sensing data to obtain a first local density value corresponding to the first sensing data includes:

[0032] Determine the target quantity corresponding to the target sensing data, and perform a logarithmic operation on the target quantity to obtain the number of nearest neighbors required for the target sensing data to perform nearest neighbor calculation;

[0033] The first sensing data is determined from the target sensing data, and the remaining sensing data corresponding to the first sensing data is obtained by removing the first sensing data from the target sensing data.

[0034] The target distance is obtained by calculating the distance between each second sensor data and the first sensor data in the remaining sensor data;

[0035] The remaining sensor data is sorted according to the target distance to obtain target sorted data, and the first neighbor sensor data corresponding to the first sensor data is obtained from the target sorted data according to the number of neighbors.

[0036] The first nearest neighbor sensing data and the relevant distance between the first sensing data are obtained from the target distance, and the first local density value corresponding to the first sensing data is obtained by reciprocal calculation based on the relevant distance.

[0037] In some optional embodiments, obtaining the reverse nearest neighbor data corresponding to the first sensing data from the target sensing data, and performing local density calculation on the reverse nearest neighbor data to obtain a second local density value, includes:

[0038] Perform nearest neighbor calculation processing on each of the second sensor data in the remaining sensor data to obtain the second nearest neighbor sensor data corresponding to the second sensor data;

[0039] When the second nearest neighbor sensing data contains the first sensing data, the second sensing data is determined as the reverse nearest neighbor data of the first sensing data;

[0040] Local density calculation is performed on each third sensor data in the reverse nearest neighbor data to obtain the second local density value corresponding to each third sensor data.

[0041] In some optional embodiments, the step of determining whether the parachute should actively deploy based on the flight attitude information, the operational status parameters, and the parachute status information, and obtaining the deployment determination result corresponding to the emergency parachute, includes:

[0042] The power-on determination result is obtained by judging the power-on status of the emergency parachute in the parachute status information;

[0043] The standby determination result is obtained by judging the standby status corresponding to the emergency parachute in the parachute status information;

[0044] Obtain the heartbeat packet corresponding to the emergency parachute, and determine the connectivity status of the emergency parachute based on the heartbeat packet to obtain the connectivity determination result;

[0045] The first judgment result corresponding to the emergency parachute is determined based on the power-on judgment result, the standby judgment result, and the connectivity judgment result.

[0046] When the first judgment result indicates that the emergency parachute is in a normal power-on, normal standby and normal connectivity state, it is then determined whether the current deployment and retraction state of the emergency parachute is the retracted state.

[0047] If the current deployment state of the emergency parachute is not the retracted state, then the deployment judgment result corresponding to the emergency parachute is determined to be the first result, and the first result is used to indicate that the emergency parachute cannot be deployed.

[0048] When the current deployment state of the emergency parachute is the retracted state, the deployment control of the emergency parachute is judged and processed according to the flight attitude information and the operating state parameters to obtain the deployment judgment result of the emergency parachute.

[0049] In some optional embodiments, the flight attitude information includes attitude angles and control attitude errors; the operating status parameters include the target UAV's operating mode, current altitude above ground, current barometric altitude, altitude descent rate, and the duration corresponding to the altitude descent rate; the step of judging and processing the deployment and retraction control of the emergency parachute based on the flight attitude information and the operating status parameters to obtain the parachute deployment judgment result, includes:

[0050] When the working mode is the preset mode, the opening judgment result corresponding to the emergency parachute is determined to be the second result, and the second result is used to indicate that the emergency parachute does not need to be opened;

[0051] When the working mode is not the preset mode, the attitude angle is compared with the angle threshold to obtain a first comparison result; if the first comparison result indicates that the attitude angle is greater than the angle threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined to be a third result, which indicates that the emergency landing requires parachute deployment; if the first comparison result indicates that the attitude angle is not greater than the angle threshold, the current ground clearance is compared with the safe altitude threshold to obtain a second comparison result; if the second comparison result indicates that the current ground clearance is less than the safe altitude threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result.

[0052] If the second comparison result indicates that the current altitude is not less than the safe altitude threshold, the altitude descent rate is compared with a preset descent rate threshold to obtain a third comparison result, and the duration is compared with a first time threshold to obtain a fourth comparison result; if the third comparison result indicates that the altitude descent rate is greater than the preset descent rate threshold and the fourth comparison result indicates that the duration is greater than the first time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the third result.

[0053] If the third comparison result indicates that the altitude descent rate is not greater than the preset descent rate threshold, or the fourth comparison result indicates that the duration is not greater than the first time threshold, the third comparison result and the fourth comparison result are continuously monitored and updated, and the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result; and the parachute deployment judgment result is continuously updated using the updated third comparison result and the fourth comparison result; and the control attitude error is further compared with the attitude error threshold to obtain a fifth comparison result;

[0054] If the fifth comparison result indicates that the control attitude error is not greater than the attitude error threshold, the runaway count is reset to 0; if the fifth comparison result indicates that the control attitude error is greater than the attitude error threshold, the runaway count is incremented by 1, and if the runaway count is 1, the current air pressure altitude is determined as the initial air pressure altitude; if the runaway count is not 1, the latest air pressure altitude is obtained, the latest air pressure altitude is assigned to the current air pressure altitude, and the latest air pressure altitude and the initial air pressure altitude are compared to obtain the sixth comparison result.

[0055] If the sixth comparison result indicates that the latest air pressure altitude is not lower than the initial air pressure altitude, the runaway count is reset; if the sixth comparison result indicates that the latest air pressure altitude is lower than the initial air pressure altitude, the execution time of the runaway count is compared with the second time threshold to obtain a seventh comparison result.

[0056] If the seventh comparison result indicates that the execution time is not greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result; if the sixth comparison result indicates that the execution time is greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the third result.

[0057] In some optional embodiments, the emergency parachute control system of the UAV further includes: a parachute control unit, wherein the controller is electrically connected to the emergency parachute via the parachute control unit; before obtaining the first control result, or before obtaining the second control result, the emergency parachute control method of the UAV includes:

[0058] When the parachute deployment judgment result indicates that the emergency parachute needs to deploy automatically, or when the operation result is to trigger the target control key corresponding to the emergency parachute, the current flight altitude of the target UAV is compared with the preset safe altitude threshold to obtain the eighth comparison result;

[0059] If the eighth comparison result indicates that the current flight altitude is greater than the preset safe altitude threshold and the parachute control unit is locked, an unlock command and a parachute deployment command are sent to the parachute control unit so that the parachute control unit unlocks in response to the unlock command and then controls the emergency parachute to deploy in response to the parachute deployment command.

[0060] If the eighth comparison result indicates that the current flight altitude is not greater than the preset safe altitude threshold, it is determined whether the target drone has landed and whether the parachute control unit has been unlocked. If yes, a locking command is sent to the parachute control unit; otherwise, control of the parachute control unit is terminated.

[0061] In a second aspect, embodiments of this application provide an emergency parachute control system for an unmanned aerial vehicle (UAV), including a controller. The controller includes at least one processor and a memory for communicatively connecting to the at least one processor. The memory stores instructions executable by the at least one processor, which are executed by the at least one processor to enable the at least one processor to perform the emergency parachute control method for an UAV as described in any of the embodiments of the first aspect.

[0062] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions for causing a computer to perform an emergency parachute control method for a drone as described in any of the embodiments of the first aspect.

[0063] The beneficial effects of this application include: by utilizing the controller of the emergency parachute control system of a UAV, during the flight of the target UAV, the initial sensing data of the target UAV is acquired in real time, and the initial sensing data is analyzed to determine the corresponding operating status parameters and flight attitude information of the target UAV; when the flight attitude information is abnormal or the operating status parameters are abnormal, a judgment is made on whether the parachute should be actively deployed based on the flight attitude information, the operating status parameters, and the parachute status information, and a deployment judgment result corresponding to the emergency parachute is obtained; the emergency parachute is controlled according to the deployment judgment result to obtain a first control result corresponding to the emergency parachute; when the flight attitude information is normal and When the operating status parameters are normal, the initial sensing data is processed for abnormal data identification to obtain target abnormal data; based on the target abnormal data, flight status risk identification is performed to determine the flight status risk level of the target UAV; based on the flight status risk level, the display information corresponding to the target control key in the remote control terminal is determined, and the display information is sent to the remote control terminal so that the remote control terminal adjusts the display state corresponding to the target control key according to the display information, thereby prompting the operator of the target UAV to operate the target control key in an emergency; the operation result corresponding to the target control key is received, and a second control result corresponding to the emergency parachute is obtained based on the operation result. The automatic triggering mechanism implemented through comprehensive decision-making is beneficial to effectively reduce the wear and tear of the UAV; at the same time, by controlling the brightness state of the target control key through the flight status risk level of the UAV, the UAV operator is guided to quickly locate the target control key so as to respond to emergencies in a timely manner, thus optimizing the manual triggering mechanism; this application optimizes the overall emergency parachute control mechanism of the UAV by being compatible with the optimized automatic triggering mechanism and the manual triggering mechanism, which can effectively reduce the wear and tear of the UAV in a crash. Attached Figure Description

[0064] Figure 1 This is a schematic diagram of the structure of an emergency parachute control system for an unmanned aerial vehicle provided in one embodiment of this application;

[0065] Figure 2 This is a flowchart illustrating an emergency parachute control method for a drone provided in one embodiment of this application;

[0066] Figure 3This is a schematic diagram of the overall process for determining whether a parachute should actively deploy, provided in one embodiment of this application.

[0067] Figure 4 This is a schematic diagram of the hardware structure of a controller provided in one embodiment of this application.

[0068] Reference numerals: Emergency parachute control system 100, controller 101, parachute control unit 102, emergency parachute 103, remote control terminal 104, sensor 105, and target control key 106 for the unmanned aerial vehicle. Detailed Implementation

[0069] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0070] It should be noted that although a logical order is shown in the flowcharts in this application, in some cases, the steps shown or described may be performed in a different order than that shown in the flowcharts. In the description of this application, "several" means one or more, and "more" means two or more. The terms "first" and "second" are used only to distinguish technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of technical features indicated, or implicitly indicating the order in which the technical features are indicated.

[0071] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0072] First, let me explain some of the terms used in this application:

[0073] In this application, ECU (Electronic Control Unit) refers to the parachute control unit.

[0074] The embodiments of this application will be further described below with reference to the accompanying drawings.

[0075] like Figure 1As shown, the emergency parachute control system 100 for a drone includes: a controller 101, a parachute control unit 102, an emergency parachute 103 electrically connected to the controller 101, a remote control terminal 104, and various sensors 105. Specifically, the controller 101 is electrically connected to the emergency parachute 103 via the parachute control unit 102; the remote control terminal 104 includes a target control key 106 for manually triggering the opening of the drone's emergency parachute; the emergency parachute 103 is electrically connected to the remote control terminal 104. The sensors 105 are used to collect various data from the drone to provide a reference for the controller 101 to monitor the drone's status.

[0076] Specifically, the emergency parachute 103 is used to retract or release under the control of the parachute control unit 102. The emergency parachute 103 is used to release in the event of a drone malfunction, acting as a buffer to slow the drone's descent and thus achieve a gentle descent, effectively protecting the expensive drone equipment. The remote control terminal 104 provides an operation panel to the drone operator, allowing the operator to control the drone within a certain remote range. When the drone operator observes an abnormal situation, they can manually control the emergency parachute 103 to open by pressing the target control key 106, causing the drone to descend gently. The parachute control unit 102 controls the unlocking, locking, release, and retraction of the emergency parachute 103. The controller 101 implements the emergency parachute control method for drones provided in this embodiment; it can optimize the emergency parachute control mechanism of the drone and effectively reduce drone wear and tear.

[0077] Those skilled in the art will understand that the system structure shown in the figures does not constitute a limitation on the embodiments of this application, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0078] It will be understood by those skilled in the art that the system architecture and application scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. It is known by those skilled in the art that with the evolution of system architecture and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0079] Based on the above system structure, various embodiments of the emergency parachute control method for unmanned aerial vehicles (UAVs) of this application are proposed below.

[0080] like Figure 2 As shown, the emergency parachute control method for this UAV can be applied to, for example... Figure 1The illustrated emergency parachute control system for the UAV includes a controller, which further comprises an emergency parachute electrically connected to the controller and a remote control terminal. The remote control terminal includes a target control key for manually triggering the release of the emergency parachute. The emergency parachute control method for the UAV may include, but is not limited to, steps S100 to S700.

[0081] Step S100: During the flight of the target UAV, acquire the initial sensor data of the target UAV in real time, and perform data analysis on the initial sensor data to determine the corresponding operating status parameters and flight attitude information of the target UAV.

[0082] Specifically, the drone is equipped with various sensors, such as current sensors, voltage sensors, temperature sensors, ultrasonic sensors, speed sensors, acceleration sensors, position sensors, and barometers. Therefore, this application does not impose specific limitations on the number and type of sensors carried on the drone. Furthermore, each sensor is electrically connected to the controller, which can acquire the initial sensing data uploaded by each sensor.

[0083] For example, the initial sensing data includes, but is not limited to: current sensing signals, voltage sensing signals, temperature sensing signals, ultrasonic sensing signals, velocity sensing signals, acceleration sensing signals, position sensing signals, and air pressure signals, etc. Therefore, this application does not impose specific limitations on the type and number of sensing signals included in the initial sensing data.

[0084] Specifically, collecting initial sensor data lays the data foundation for conducting data analysis to determine the corresponding operational status parameters and flight attitude information of the target UAV.

[0085] For example, the controller analyzes the initial sensor data to determine flight attitude information, including: acquiring basic operational information such as angular velocity, acceleration, and heading from the initial sensor data; and determining the target UAV's flight attitude information by performing data fusion and attitude calculation on the basic operational information. Determining the target UAV's flight attitude information through the fusion of multiple sensor data improves the reliability of attitude prediction. This application does not impose specific limitations on how the UAV's flight attitude information is determined. Specifically, flight attitude information includes, but is not limited to: attitude angles and control attitude errors; attitude angles include: roll attitude angle values ​​and pitch attitude angle values.

[0086] Specifically, the operational status parameters include, but are not limited to: the target drone's descent speed, current altitude, descent rate, duration of descent, and current barometric altitude.

[0087] For example, the current altitude of the target UAV can be determined by a position sensor signal; the rate of descent can be determined by a velocity sensor signal over a continuous time period; and the current barometric altitude can be determined by conversion based on the current barometric pressure sensor signal. This application does not impose specific limitations on the method of obtaining the operational status parameters.

[0088] In this step, during the flight of the UAV, the controller acquires the initial sensor data, operating status parameters and flight attitude information of the target UAV in real time, laying the data foundation for subsequent abnormal data identification and processing, flight status risk identification and processing, and comprehensive judgment and control processing.

[0089] It is understandable that in step S100, after analyzing the initial sensing data to determine the corresponding operating status parameters and flight attitude information of the target UAV, it is possible to quickly determine whether the flight attitude information and initial sensing data are normal. When the flight attitude information is abnormal or the operating status parameters are abnormal, steps S200 to S300 provided in the following embodiment are executed; when the flight attitude information is normal and the operating status parameters are normal, steps S400 to S700 provided in the following embodiment are executed.

[0090] Step S200: When the flight attitude information is abnormal or the operating status parameters are abnormal, determine whether the parachute should be actively deployed based on the flight attitude information, operating status parameters, and parachute status information, and obtain the deployment judgment result corresponding to the emergency parachute.

[0091] In this step, when the controller detects that the flight attitude information or the operating status parameters are abnormal, the controller will determine whether the parachute should be actively deployed based on the acquired flight attitude information, operating status parameters, and parachute status information, and obtain the deployment judgment result corresponding to the emergency parachute, providing a reliable reference for subsequent active control of the emergency parachute.

[0092] Specifically, the parachute status information includes, but is not limited to: the power-on status of the emergency parachute and the standby status of the emergency parachute.

[0093] For example, when the flight attitude information does not meet the preset attitude, such as rotation attitude, or the operating status parameters do not meet the preset data, but the parachute information meets the preset status, the emergency parachute deployment judgment result is determined to be that the parachute needs to be deployed; when the flight attitude information does not meet the preset attitude or the operating status parameters do not meet the preset data, but the parachute information does not meet the preset status, the emergency parachute deployment judgment result is determined to be that the parachute cannot be deployed; when the flight attitude information meets the preset attitude or the operating status parameters meet the preset data, the emergency parachute deployment judgment result is determined to be that the parachute does not need to be deployed.

[0094] In some optional embodiments, step S200 is further described, wherein a determination is made on whether the parachute should be actively deployed based on flight attitude information, operational status parameters, and parachute status information, and the deployment determination result corresponding to the emergency parachute is obtained, including but not limited to steps S210 to S270.

[0095] Step S210: Determine the power-on status of the emergency parachute based on the parachute status information to obtain the power-on judgment result.

[0096] For example, after the controller obtains the power-on status of the emergency parachute from the parachute status information, it makes a judgment based on the power-on status to obtain a power-on judgment result; specifically, the power-on judgment result is that the emergency parachute is in normal power-on or that the emergency parachute is in abnormal power-on. Obtaining the power-on judgment result lays the foundation for subsequently determining the first judgment result.

[0097] Specifically, the controller can determine the power-on status of the emergency parachute by detecting the output current or voltage of the power module electrically connected to the emergency parachute, or by detecting changes in the current or voltage of the parachute control unit. This application does not impose specific limitations on how the parachute's power-on status is obtained.

[0098] Step S220: Determine the standby status based on the standby status of the emergency parachute in the parachute status information to obtain the standby judgment result.

[0099] For example, after the controller obtains the standby status corresponding to the emergency parachute from the parachute status information, it judges the standby status to obtain a standby judgment result; specifically, the standby judgment result is either that the emergency parachute is in normal standby or that the emergency parachute is in abnormal standby. Obtaining the standby status corresponding to the emergency parachute lays the foundation for subsequently determining the first judgment result.

[0100] Specifically, in one embodiment, the standby judgment result can be determined by obtaining the standby state identifier; for example, when the standby state identifier is 0, the standby judgment result is determined to be an emergency parachute standby malfunction; when the standby state identifier is 1, the standby judgment result is determined to be an emergency parachute standby function. In another embodiment, the controller and the parachute control unit communicate in real time, and the standby state is obtained by parsing the response to determine the standby judgment result. Therefore, this application does not limit the specific process of how to determine the standby judgment result.

[0101] Step S230: Obtain the heartbeat packet corresponding to the emergency parachute, and determine the connectivity status of the emergency parachute based on the heartbeat packet to obtain the connectivity determination result.

[0102] Specifically, a heartbeat packet refers to a data packet that is periodically and actively sent by the parachute control unit. The controller communicates with the parachute control unit to obtain the heartbeat packets sent by the parachute control unit; the heartbeat packets are parsed to obtain the connectivity judgment result; specifically, the connectivity judgment result is either "connectivity is normal" or "connectivity is abnormal". Obtaining the connectivity judgment result lays the foundation for subsequently determining the first judgment result.

[0103] Step S240: Determine the first judgment result corresponding to the emergency parachute based on the power-on judgment result, the standby judgment result, and the connectivity judgment result.

[0104] For example, after completing steps S210 to S230, corresponding power-on judgment results, standby judgment results, and connectivity judgment results are obtained, thereby determining the first judgment result corresponding to the emergency parachute. Obtaining the first judgment result can provide a reference for whether to continue to the next step of judging the current deployment and retraction status of the emergency parachute.

[0105] For example, if any one of the power-on judgment result, standby judgment result, or connectivity judgment result is abnormal, then the first judgment result corresponding to the emergency parachute is determined to be abnormal. When all three judgment results are normal, then the first judgment result corresponding to the emergency parachute is determined to be normal. If the first judgment result indicates that the emergency parachute is experiencing a power-on abnormality, a standby abnormality, or a connectivity abnormality, i.e., the first judgment result is abnormal, then the controller determines that the emergency parachute cannot deploy and can directly determine the deployment judgment result corresponding to the emergency parachute as the first result. The first result is used to indicate that the emergency parachute cannot deploy.

[0106] Step S250: When the first judgment result indicates that the emergency parachute is in a normal power-on, normal standby and normal connection state, then determine whether the current deployment state of the emergency parachute is the retracted state.

[0107] For example, when the first judgment result indicates that the emergency parachute is in a normal power-on, normal standby, and normal connection state, that is, when the first judgment result is all normal, the controller proceeds to the next judgment to determine whether the current deployment and retraction state of the emergency parachute is in the retracted state; that is, to continue to determine whether the emergency parachute is currently in the open state.

[0108] Step S260: If the current deployment state of the emergency parachute is not the retracted state, then the deployment judgment result of the emergency parachute is determined as the first result. The first result is used to characterize that the emergency parachute cannot be deployed.

[0109] For example, when the controller determines that the current deployment state of the emergency parachute is not the retracted state (i.e., the emergency parachute has been released), it determines that the emergency parachute cannot open and directly determines the opening judgment result of the emergency parachute as the first result.

[0110] Step S270: When the current deployment state of the emergency parachute is the retracted state, the deployment control of the emergency parachute is judged and processed according to the flight attitude information and operating status parameters to obtain the corresponding parachute deployment judgment result.

[0111] For example, when the current deployment state of the emergency parachute is the retracted state, that is, when the current deployment state of the emergency parachute is that the parachute is not opened, the emergency parachute is identified by an anomaly warning using machine learning algorithms or deep learning algorithms based on flight attitude information and operational status parameters, thereby making judgments and processing on the deployment control of the emergency parachute, and thus obtaining the corresponding parachute opening judgment result.

[0112] For example, when a high risk is determined based on flight attitude information and operational status parameters, the parachute deployment judgment result is determined as the third result, which is used to characterize that parachute deployment is required for emergency landing; when a low risk is determined based on flight attitude information and operational status parameters, the parachute deployment judgment result is determined as the second result, which is used to characterize that parachute deployment is not required for emergency landing.

[0113] In some optional embodiments, step S270 is further described. Specifically, the flight attitude information includes attitude angle and control attitude error; the operating status parameters include the target UAV's operating mode, current altitude above ground, current barometric altitude, altitude descent rate, and the duration corresponding to the altitude descent rate. The emergency parachute deployment and recovery control is processed based on the flight attitude information and operating status parameters to obtain the corresponding parachute deployment judgment result, including but not limited to steps S271 to S277.

[0114] Step S271: When the working mode is the preset mode, the parachute deployment judgment result corresponding to the emergency parachute is determined as the second result. The second result is used to indicate that the emergency parachute does not need to be deployed.

[0115] For example, the preset modes include: aerobatic mode and flip mode. When the controller determines that the target drone is in aerobatic mode or flip mode, since the target drone may perform actions such as diving, sudden takeoff, or sharp flips, if an active parachute deployment judgment is made at this time, there is a high probability of misjudgment and incorrect deployment of the emergency parachute, resulting in the inability to complete the aerobatic performance or flip normally. Therefore, when the target drone is in aerobatic mode or flip mode, this application determines that the emergency parachute does not need to be deployed, ending the overall judgment of whether the parachute should be actively deployed, thus reducing misjudgment.

[0116] Step S272: When the working mode is not the preset mode, the attitude angle is compared with the angle threshold to obtain the first comparison result; if the first comparison result indicates that the attitude angle is greater than the angle threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined as the third result, which is used to indicate that the emergency landing requires parachute deployment; if the first comparison result indicates that the attitude angle is not greater than the angle threshold, the current ground clearance is compared with the safe altitude threshold to obtain the second comparison result; if the second comparison result indicates that the current ground clearance is less than the safe altitude threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined as the second result.

[0117] For example, the attitude angles include: roll attitude angle value and pitch attitude angle value; the angle threshold is specifically 50 degrees; the safe altitude threshold is specifically 10 meters. It is understood that the values ​​of the angle threshold and the safe altitude threshold can be set according to actual needs, and this application does not make specific settings for them. If the controller determines that the target drone's operating mode is neither acrobatic mode nor roll mode, it is necessary to further determine whether the parachute needs to be deployed based on other parameters besides the operating mode. First, the attitude angle is compared with the angle threshold. If the attitude angle is greater than the angle threshold, it indicates that the target drone is in an abnormal or extreme flight state, which may lead to loss of control or structural damage. Based on this, a decision is made that the emergency parachute needs to be deployed. If the attitude angle is not greater than the angle threshold, it indicates that the target drone is not in an abnormal or extreme flight state. Based on other parameters (besides the operating mode and attitude angle), a further determination is made whether the parachute needs to be deployed. Then, the controller compares the target drone's current ground clearance with the safe altitude threshold. If the current ground clearance is less than the safe altitude threshold, it indicates that the target drone is within a safe and controllable altitude range, and a decision is made that the emergency parachute does not need to be deployed.

[0118] Step S273: If the second comparison result indicates that the current altitude is not less than the safe altitude threshold, continue to compare the altitude descent rate with the preset descent rate threshold to obtain the third comparison result, and at the same time compare the duration with the first time threshold to obtain the fourth comparison result; if the third comparison result indicates that the altitude descent rate is greater than the preset descent rate threshold and the fourth comparison result indicates that the duration is greater than the first time threshold, then determine the parachute deployment judgment result corresponding to the emergency parachute as the third result.

[0119] For example, the preset descent rate threshold is specifically 0.8 m / s, and the first time threshold is specifically 0.8 seconds; it is understood that the values ​​of the preset descent rate threshold and the first time threshold can be set according to actual needs, and this application does not make specific settings in this regard.

[0120] For example, if the controller determines that the current altitude above the ground is not less than a safe altitude threshold, it needs to make a subsequent judgment on whether the parachute needs to be deployed based on other parameters (besides the operating mode, attitude angle, and current altitude above the ground). First, it continues to compare the rate of descent with a preset descent rate threshold, and at the same time, it compares the duration with a first time threshold. Then, if the rate of descent is greater than the preset descent rate threshold and the duration is greater than the first time threshold, it indicates that the target drone may be falling at an abnormal speed due to abnormal conditions such as power system failure, sudden power loss, or structural damage, and there is a risk of crashing. Based on this, a judgment is made that the emergency parachute needs to be deployed, and the subsequent deployment operation is carried out to protect the target drone.

[0121] Step S274: If the altitude descent rate represented by the third comparison result is not greater than the preset descent rate threshold or the duration represented by the fourth comparison result is not greater than the first time threshold, continuously monitor and update the third and fourth comparison results, and determine the parachute deployment judgment result corresponding to the emergency parachute as the second result; and continuously update the parachute deployment judgment result using the updated third and fourth comparison results; and continue to compare the control attitude error with the attitude error threshold to obtain the fifth comparison result.

[0122] For example, the attitude error threshold is specifically 30 degrees; it is understood that the value of the attitude error threshold can be set according to actual needs, and this application does not make a specific setting for it.

[0123] For example, if the controller determines that the altitude descent rate is not greater than a preset descent rate threshold or the duration is not greater than a first time threshold, it indicates that the target drone is temporarily in a safe and controllable flight state, and it is determined that the emergency parachute does not need to be deployed. However, it is still necessary to continuously monitor and update the third and fourth comparison results, and continuously update the parachute deployment judgment result based on the updated third and fourth comparison results, so as to promptly detect situations where the target drone is falling at an abnormal speed and has a risk of crashing, and to respond to the risk in a timely manner, making the parachute deployment judgment result that the emergency parachute needs to be deployed, thereby performing subsequent parachute deployment operations to protect the target drone. Furthermore, the controller continues to compare the control attitude error with the attitude error threshold to obtain a fifth comparison result, so as to further assist in determining whether the parachute needs to be deployed to protect the target drone, effectively reducing the damage caused by the target drone crashing.

[0124] Step S275: If the fifth comparison result indicates that the control attitude error is not greater than the attitude error threshold, reset the runaway count to 0; if the fifth comparison result indicates that the control attitude error is greater than the attitude error threshold, increment the runaway count by 1, and if the runaway count is 1, determine the current air pressure altitude as the initial air pressure altitude; if the runaway count is not 1, obtain the latest air pressure altitude, assign the latest air pressure altitude to the current air pressure altitude, and compare the latest air pressure altitude with the initial air pressure altitude to obtain the sixth comparison result.

[0125] It should be noted that the initial barometric altitude and the latest barometric altitude are parameters obtained through sensor measurements. They refer to the altitude of the target drone determined by measuring atmospheric pressure.

[0126] For example, if the controller determines that the control attitude error of the target UAV is greater than the attitude error threshold, it determines that the target UAV is abnormal and increments the out-of-control count by 1; if the controller determines that the control attitude error of the target UAV is not greater than the attitude error threshold, it determines that the target UAV is not out of control and resets the out-of-control count to 0; this is to facilitate recording the number of out-of-control situations. Further, if the out-of-control count is 1, the current air pressure altitude is determined as the initial air pressure altitude and used as the judgment standard; if the out-of-control count is not 1, the latest air pressure altitude is assigned to the current air pressure altitude, and the latest air pressure altitude and the initial air pressure altitude are compared to obtain the sixth comparison result.

[0127] Step S276: If the sixth comparison result indicates that the latest air pressure altitude is not lower than the initial air pressure altitude, reset the runaway count; if the sixth comparison result indicates that the latest air pressure altitude is lower than the initial air pressure altitude, continue to compare the execution time of the runaway count with the second time threshold to obtain the seventh comparison result.

[0128] For example, the second time threshold is specifically 1 second; it is understood that the value of the second time threshold can be set according to actual needs, and this application does not make a specific setting for it. Specifically, the control attitude error refers to the difference between the actual attitude of the UAV and the attitude expected or commanded under flight control.

[0129] For example, if the controller determines that the latest atmospheric pressure altitude is not lower than the initial atmospheric pressure altitude, it means that the target drone is in a region with lower atmospheric pressure after normal climbing, i.e., a higher position in the air. At this time, the runaway count will be reset. If the latest atmospheric pressure altitude is lower than the initial atmospheric pressure altitude, it means that the target drone has fallen to a lower control position. It is necessary to continue to compare the execution time of the runaway count with the second time threshold to obtain the seventh comparison result, so as to make a more accurate and reliable parachute opening judgment.

[0130] Step S277: If the execution time of the seventh comparison result is not greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined as the second result; if the execution time of the sixth comparison result is greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined as the third result.

[0131] For example, if the controller determines that the execution time is not greater than the second time threshold, and determines that the duration of the target drone's loss of control is within a controllable range and the target drone is in a relatively safe and stable operating state, then the controller will determine that the emergency parachute does not need to be deployed. Conversely, if the controller determines that the execution time is greater than the second time threshold, and determines that the duration of the target drone's loss of control is not within a controllable range and the target drone is in a dangerous loss of control operating state, then the controller will determine that the emergency parachute needs to be deployed, so as to protect the target drone in a timely manner.

[0132] This embodiment of the application realizes the judgment and processing of emergency parachute deployment and retrieval control through steps S271 to S277, and realizes a comprehensive decision-making mechanism based on multiple judgment and processing to make a more reliable decision on whether to release or not to release the emergency parachute.

[0133] This embodiment of the application determines whether the parachute should actively deploy through steps S210 to S270. In this process, a comprehensive judgment is first made based on the emergency parachute's power-on status, standby status, and heart rate status. If the emergency parachute is in a normal power-on, standby, and connectivity state, then it is further determined whether the current deployment / retraction state of the emergency parachute is in a retracted state. If the emergency parachute is not in a retracted state, it is determined that the emergency parachute cannot deploy. If the emergency parachute is in a retracted state, further control and judgment processing for the deployment / retraction of the emergency parachute is performed. During the control and judgment processing of the emergency parachute deployment / retraction, a comprehensive decision-making mechanism based on multiple judgment processes is achieved by judging and analyzing information such as attitude angle, current altitude above ground, descent rate, duration, control attitude error, current barometric altitude, and loss of control count time with corresponding thresholds. This more reliably makes the decision to release or not release the emergency parachute, thereby optimizing the automatic triggering mechanism of the emergency parachute.

[0134] Step S300: Control the emergency parachute according to the parachute deployment judgment result to obtain the first control result corresponding to the emergency parachute.

[0135] Specifically, the parachute deployment determination results include: a first result indicating that the emergency parachute cannot deploy, a second result indicating that the emergency parachute does not need to deploy, and a third result indicating that the emergency landing requires parachute deployment. The controller controls the emergency parachute based on the deployment determination results, and will obtain the corresponding first control result.

[0136] For example, when the parachute deployment judgment result is the first result, the controller controls the emergency parachute to deploy according to the first result. The corresponding first control result is that the emergency parachute cannot be deployed. The controller also returns the abnormality in the power-on judgment result, standby judgment result, and connectivity judgment result of the emergency parachute to the remote control terminal so that the user can obtain the abnormal status of the emergency parachute in a timely manner through the remote control terminal.

[0137] For example, when the parachute deployment determination result is the second result, the controller controls the emergency parachute to deploy based on the second result. The corresponding first control result is: the emergency parachute remains in its current state and does not need to be deployed.

[0138] For example, when the parachute deployment determination result is the third result, the controller controls the emergency parachute deployment based on the third result, and the corresponding first control result is: emergency parachute release.

[0139] Step S400: When the flight attitude information is normal and the operating status parameters are normal, perform abnormal data identification processing on the initial sensing data to obtain target abnormal data.

[0140] For example, when the flight attitude information is normal and the operating status parameters are normal, an anomaly identification model, such as a neural network model, is used to perform anomaly data identification processing on the initial sensing data, thereby obtaining the target anomaly data corresponding to the initial sensing data.

[0141] In some optional embodiments, step S400 is further described, wherein the initial sensing data is subjected to abnormal data identification processing to obtain target abnormal data, including but not limited to steps S401 to S412.

[0142] Step S401: Obtain window data under the target window by sliding the preset target window in the initial sensing data.

[0143] Specifically, the preset target window is a fixed-length frame. The length of the preset target window is pre-configured and can be set according to actual needs. Therefore, this application does not impose specific restrictions on the length of the preset target window.

[0144] For example, the controller first generates a preset target window and uses this target window to slide and sample the initial sensing data, acquiring multiple window data of equal length from the initial sensing data that is continuously generated over time, so as to facilitate subsequent data processing in units of the target window. For example, a preset target window with a set window length of 5 is used to slide and acquire window data in the current sensing signal; the window data includes five sampling points: 1.02, 1.05, 0.98, 1.00, and 0.99.

[0145] Step S402: Perform the first data calculation on the window data to obtain interference data.

[0146] For example, a random number is generated by a random number generator, and then the random number is superimposed on the window data to obtain the interference data corresponding to the window data, wherein the first data calculation process includes generating random numbers and data superposition processing.

[0147] In some optional embodiments, step S402: perform a first data calculation process on the window data to obtain interference data, including but not limited to steps S4021 to S4024.

[0148] Step S4021: Calculate the mean of the window data to obtain the average data.

[0149] Specifically, the arithmetic mean calculation refers to the operation of summing the data and then taking the average. If the window data includes five sampling points: 1.02, 1.05, 0.98, 1.00, and 0.99, the arithmetic mean of the five sampling points is: (1.02 + 1.05 + 0.98 + 1.00 + 0.99) / 5 = 5.04 / 5 = 1.008; that is, the average data corresponding to the window data is 1.008.

[0150] In this step, after the controller acquires the window data, it calculates the mean of multiple data points included in the window data to obtain the average data. This average data represents the average level of all data within this window and serves as the baseline for subsequently calculating the deviation of each data point within this window.

[0151] Step S4022: Calculate the absolute difference between each sub-data point in the window data and the average data, and calculate the mean of all absolute differences to obtain the deviation value.

[0152] Specifically, when the original window data includes: 1.02, 1.05, 0.98, 1.00, 0.99; and the average data is 1.008, the absolute differences between each sub-data and the average data are calculated to be: 0.012, 0.042, 0.028, 0.008, 0.018. The arithmetic mean of these 5 absolute differences is: (0.012+0.042+0.028+0.008+0.018) / 5=0.0216, and 0.0216 is the deviation value.

[0153] In this step, after calculating the average data, the controller calculates the deviation value of each sub-data point relative to the average data. The deviation value is a statistic that quantifies the volatility or dispersion within the window data; a large deviation value indicates severe data fluctuation, while a small deviation value indicates stable data. The deviation value provides a reliable reference for generating interference values ​​in the next step.

[0154] Step S4023: Generate the interference value corresponding to the window data based on the deviation value.

[0155] In this step, after calculating the deviation value, the controller creates an interference value that matches the inherent volatility of the window data, which is equivalent to simulating a controllable noise signal to facilitate the next step of interference processing.

[0156] Step S4024: Add the interference value to the window data to obtain interference data.

[0157] In this step, the controller adds generated interference values ​​to the window data. These interference values ​​are matched to the inherent volatility of the window data, thus largely preserving the main trends and shapes of the original window data. At the same time, data amplification is also performed to obtain reliable interference data, which is then used for the next step of data analysis.

[0158] Step S403: Perform data distribution analysis on the initial sensing data based on the Gaussian mixture model to obtain the first distribution result.

[0159] Specifically, in this step, the Gaussian Mixture Model (GMM) is a statistical model that describes complex data distributions by combining multiple Gaussian distributions. Its core idea is to view data as a mixture of multiple Gaussian distributions, each representing a cluster, and to estimate model parameters by maximizing the likelihood function. GMMs are primarily used for cluster analysis and probability density estimation, capable of handling multimodal data and adapting to datasets of different shapes.

[0160] In some embodiments, the expectation-maximization algorithm is used in the first data distribution analysis process to obtain multiple Gaussian distributions.

[0161] For example, after the controller performs data distribution analysis on the initial sensing data based on the Gaussian mixture model, the first distribution result includes multiple first Gaussian distributions. In this way, the powerful multimodal distribution fitting capability of the Gaussian mixture model (GMM) is utilized to accurately characterize the complex statistical distribution characteristics of the initial sensing data, providing an accurate probability model basis for subsequent analysis.

[0162] Step S404: Perform data distribution analysis on the interference data based on the Gaussian mixture model to obtain the second distribution result.

[0163] It is understandable that the data distribution analysis process in step S404 is the same as that in step S403, but the objects being processed are different.

[0164] For example, after the controller performs data distribution analysis on the interference data based on the Gaussian mixture model, the second distribution result includes multiple second Gaussian distributions. In this way, the powerful multimodal distribution fitting capability of the Gaussian mixture model (GMM) is utilized to quantitatively evaluate the distribution characteristics of the window data after being disturbed, providing a reference for comparative analysis.

[0165] Step S405: Based on the first distribution result and the second distribution result, perform data comparison and analysis to determine the Gaussian distribution characteristics corresponding to the initial sensing data.

[0166] For example, after obtaining the first and second distribution results, the controller compares the similarity between the first Gaussian distribution in the first distribution result and the second Gaussian distribution in the second distribution result. After removing the first Gaussian distributions similar to the second Gaussian distribution, the remaining first Gaussian distribution is the Gaussian distribution feature corresponding to the initial sensing data. This is equivalent to removing the first Gaussian distributions similar to the second Gaussian distribution of the interfering data, leaving the remaining first Gaussian distribution as a pure and meaningful data distribution feature, providing a reliable data foundation for further data processing. In other words, this step, by comparing the distribution differences between the original data and the interfering data, can identify stable and significant Gaussian distribution features in the initial data. This process helps to focus on the most representative and less affected essential features in the data, enhancing the reliability of subsequent processing.

[0167] Step S406: Perform data distribution analysis on the initial sensing data based on the Gaussian distribution characteristics to obtain the probability distribution corresponding to the initial sensing data.

[0168] Understandably, according to conventional mathematical knowledge, the Gaussian distribution is one of the most important distributions in probability theory and statistics. The probability density function of the Gaussian distribution is completely determined by its two parameters: mean (μ) and standard deviation (σ). The mean represents the average value of all data points, indicating the central location of the distribution; the standard deviation represents the average degree to which all data points deviate from the mean, indicating the dispersion of the distribution. Therefore, the characteristics of the Gaussian distribution in this step include: mean (μ) and standard deviation (σ).

[0169] For example, after obtaining the Gaussian distribution characteristics of the initial sensing data, the controller constructs a probability density function based on the Gaussian distribution characteristics (mean and standard deviation), and performs an integral operation on the probability density function to obtain the probability distribution corresponding to the initial sensing data. Using the identified Gaussian distribution characteristics, the initial sensing data is re-probabilized, filtering out irrelevant noise or interference patterns to obtain a purer probability distribution that better reflects the inherent patterns of the data; this provides a basis for subsequently determining the target threshold used to compress the initial sensing data.

[0170] Step S407: Obtain the relevant sensing data corresponding to the maximum value of the probability distribution, and calculate the data difference between the relevant sensing data and the initial sensing data.

[0171] For example, the relevant sensing data corresponding to the maximum value of the probability distribution is the data that is most concentrated and appears most frequently in the initial sensing data. The relevant sensing data corresponding to the maximum value of the probability distribution is obtained as the classic reference benchmark value; the data difference between the classic reference benchmark value and the initial sensing data can be calculated to quantify the degree of deviation between each actual data point in the initial sensing data and the classic reference benchmark value. The data difference is the key basis for subsequent threshold processing.

[0172] Step S408: Determine the target threshold based on the data difference, and based on the comparison results between the initial sensing data and the target threshold, reassign the initial sensor data and perform data compression processing to obtain the target sensing data.

[0173] For example, after obtaining the data differences, if there are multiple data differences, the controller calculates the average of these differences to obtain the mean information. This mean information represents the average distance of all original data points from the reference baseline. Compared to directly configuring the target threshold, setting the mean information obtained through a series of rigorous derivations and calculations as the target threshold can reduce misjudgments and improve the reliability of subsequent compression of initial sensor data. After determining this mean information as the target threshold, if the comparison result shows that the initial sensor data is greater than the target threshold, the initial sensor data greater than the target threshold remains unchanged; if the comparison result shows that the initial sensor data is less than or equal to the target threshold, the initial sensor data less than or equal to the target threshold is reassigned (it can be assigned a value of 0 or other values) to obtain the target deleted data; the target deleted data is deleted, and the remaining initial sensor data is determined as the target sensor data. In this way, the target threshold is adaptively determined based on the degree of data deviation. By deleting data with small fluctuations (below the threshold), lossy compression is achieved, which significantly reduces the amount of data and transmission overhead, while retaining significant changes. Thus, data compression and denoising can be completed efficiently while maintaining the key statistical characteristics of the data.

[0174] Step S409: Calculate the density of each first sensing data in the target sensing data to obtain the first local density value corresponding to the first sensing data.

[0175] For example, clustering calculations are performed on the target sensing data to obtain the target clustering result, and then the sub-cluster of each first sensing data in the target clustering result is obtained. The first local density value corresponding to the first sensing data is determined according to the number of data corresponding to the sub-cluster and the total number of target sensing data.

[0176] In some optional embodiments, step S409 includes, but is not limited to, steps S4091 to S4095.

[0177] Step S4091: Determine the number of targets corresponding to the target sensing data, and perform a logarithmic operation on the number of targets to obtain the number of nearest neighbors required for the target sensing data to perform nearest neighbor calculation. For example, suppose the number of data points of the target sensing data is M = 1000; then k = lgM = lg1000 = 3; where lg is the logarithmic operator in mathematics, representing taking the logarithm to the base 10, which is also a commonly used logarithmic operation; other bases can also be used for the operation.

[0178] It should be noted that the nearest neighbor calculation refers to the k-Nearest Neighbor (KNN) algorithm. The KNN algorithm is a basic classification and regression method. Its core idea is: in a dataset, calculate the distance between a target data point and all remaining data points, and select the k nearest neighbors to the target data point as the k-nearest range.

[0179] For example, the controller determines the number of targets corresponding to the target sensing data, determines the logarithm to be used, and performs a logarithmic operation on the number of targets to obtain the number of nearest neighbors k. The number of nearest neighbors k then provides a reliable reference for subsequent k-nearest neighbor calculation processing.

[0180] Step S4092: Determine the first sensing data from the target sensing data, and remove the first sensing data from the target sensing data to obtain the remaining sensing data corresponding to the first sensing data.

[0181] For example, after the controller determines the first sensing data from the target sensing data, it removes the first sensing data from the target sensing data to obtain the remaining sensing data, which includes multiple second sensing data. The first sensing data and the second sensing data are distinguished to facilitate the subsequent calculation of the first local density value corresponding to the first sensing data.

[0182] Step S4093: Calculate the distance between each second sensor data and the first sensor data in the remaining sensor data to obtain the target distance.

[0183] For example, after obtaining the remaining sensor data including multiple second sensor data, the controller uses the first sensor data as a reference and calculates the distance between each second sensor data and the first sensor data using the Euclidean distance formula, thereby obtaining multiple target distances.

[0184] Step S4094: Sort the remaining sensor data according to the target distance to obtain target sorted data, and obtain the first nearest neighbor sensor data corresponding to the first sensor data from the target sorted data according to the number of nearest neighbors.

[0185] For example, the controller sorts the remaining sensor data using the target distance as a reference. The smaller the target distance, the higher the corresponding second sensor data is ranked; conversely, the larger the target distance, the lower the corresponding second sensor data is ranked. After traversing all target distances, the controller synchronously sorts each second sensor data in the remaining sensor data to obtain target sorted data. Following a front-to-back order, the controller retrieves the first nearest neighbor sensor data corresponding to the first sensor data from the target sorted data. The number of first nearest neighbor sensor data is equal to the number of neighbors, k. For example, when the number of neighbors, k, is four, the first four second sensor data are selected from the target sorted data as the first nearest neighbor sensor data corresponding to the first sensor data.

[0186] Step S4095: Obtain the first nearest neighbor sensing data and the correlation distance between the first sensing data from the target distance, and calculate the first local density value corresponding to the first sensing data by reciprocal calculation based on the correlation distance.

[0187] For example, after obtaining the first nearest neighbor sensing data, the controller determines the maximum distance between the first nearest neighbor sensing data and the first sensing data (that is, the distance between the k-th first nearest neighbor sensing data and the first sensing data) from multiple target distances as the correlation distance between the first nearest neighbor sensing data and the first sensing data. Then, the reciprocal of the correlation distance is taken to obtain the first local density value corresponding to the first sensing data. The controller performs local density estimation on the compressed data (target sensing data) to evaluate the degree of clustering of the compressed target sensing data in the feature space. The results provide a metric for the sparsity of data distribution for subsequent data outlier analysis.

[0188] Step S410: Obtain the reverse nearest neighbor data corresponding to the first sensing data from the target sensing data, and perform local density calculation on the reverse nearest neighbor data to obtain the second local density value.

[0189] In some optional embodiments, step S410 includes, but is not limited to, steps S4101 to S4103.

[0190] Step S4101: Perform nearest neighbor calculation processing on each second sensor data in the remaining sensor data to obtain the second nearest neighbor sensor data corresponding to the second sensor data.

[0191] It should be noted that the nearest neighbor calculation process in step S4101 uses the number of nearest neighbors calculated in step S4091. Nearest neighbor calculation can effectively reveal the local structure and distribution characteristics of data in the feature space, providing a basis for identifying the mutual influence between data points.

[0192] For example, after obtaining the remaining sensor data including multiple second sensor data, the controller first divides the multiple second sensor data into a target data point and remaining data points; then, it calculates the distance between the target data point and each remaining data point, and selects the k nearest neighbor points to the target data point based on the distance and the number of nearest neighbors k, and uses the k nearest neighbor points to determine the k-nearest range, which are the second nearest neighbor sensor data; when traversing each second sensor data, the above operation is repeated with each second sensor data as the target data point to obtain the second nearest neighbor sensor data corresponding to each second sensor data, so as to further determine the reverse nearest neighbor data of the first sensor data.

[0193] Step S4102: When the second nearest neighbor sensing data contains the first sensing data, the second sensing data is determined as the reverse nearest neighbor data of the first sensing data.

[0194] For example, when the second nearest neighbor sensing data contains the first sensing data, the second sensing data corresponding to the second nearest neighbor sensing data is determined as the reverse nearest neighbor data of the first sensing data; thus, data with reverse nearest neighbor relationships to the first sensing data are found more comprehensively.

[0195] Step S4103: Perform local density calculation processing on each third sensor data in the reverse nearest neighbor data to obtain the second local density value corresponding to each third sensor data.

[0196] Specifically, the local density calculation process performed when calculating the second local density value is the same as the density calculation process performed when calculating the first local density value (i.e., steps S4091 to S4095), but the objects being processed are different.

[0197] For example, step S4103 includes: obtaining the number of nearest neighbors required for nearest neighbor calculation; determining a target data point and multiple remaining data points from multiple third-sensor data; calculating the distance between the target data point and each remaining data point to obtain a target distance; sorting the remaining data points according to the target distance to obtain target sorted data; obtaining the third nearest neighbor sensor data corresponding to the target data point from the target sorted data according to the number of nearest neighbors; obtaining the correlation distance between the third nearest neighbor sensor data and the target data point from the target distance; and taking the reciprocal of the correlation distance to obtain the second local density value corresponding to the target data point; performing the above steps on each third-sensor data to obtain the second local density value of each third-sensor data. The controller, by performing density evaluation on the reverse nearest neighbor set of the first sensor data, can quantify the degree of clustering of data points within the set. This provides a metric for the sparsity of data distribution for subsequent data outlier analysis.

[0198] Step S411: Calculate the average density value by averaging all the second local density values, and determine the outlier characterization value corresponding to the first sensing data based on the average density value and the first local density value.

[0199] For example, the controller calculates the average density value by averaging all the obtained second local density values, which is the average density of the reverse nearest neighbor set of the first sensing data, thus establishing an objective density baseline for the entire local area; then the ratio of the average density value to the first local density value is determined as the outlier characterization value, which provides a standardized and quantifiable screening criterion.

[0200] Step S412: Determine the target anomaly data corresponding to the initial sensing data based on the outlier characterization value and the target threshold.

[0201] For example, the controller uses outlier values ​​as a filtering criterion for abnormal data. From the initial sensor data, data with outlier values ​​no greater than a target threshold are identified as normal data, while data with outlier values ​​greater than the target threshold are identified as target abnormal data during the UAV's flight. This completes the overall abnormal data identification.

[0202] In this embodiment of the application, abnormal data identification and processing is completed through a series of rigorous data calculations in step S400, and the target abnormal data is determined, providing a data foundation for subsequent flight status risk identification and processing.

[0203] Step S500: Perform flight status risk identification processing based on the target anomaly data to determine the flight status risk level of the target UAV.

[0204] For example, the controller performs flight status risk identification processing based on abnormal target data to determine the flight status risk level of the UAV, so as to determine the display information corresponding to the target control keys in the remote control terminal. There are several specific methods for flight status risk identification processing. Firstly, the flight status risk level can be determined based on the threshold range of the abnormal target data volume; the larger the abnormal data volume, the higher the flight status risk level. For example, a risk level mapping table can be pre-configured, including: a first threshold range corresponding to a first risk level, and a second threshold range corresponding to a second risk level; the first risk level is higher than the second risk level. When the abnormal target data volume is within the second threshold range, the flight status risk level is directly determined to be the second risk level. Secondly, the flight status risk level can also be further determined based on the importance of the sensor that generated the abnormal target data. For example, the importance of sensors can be graded; the higher the importance of the sensor that generated the abnormal data, the higher the corresponding flight status risk level; the lower the importance of the sensor that generated the abnormal data, the lower the corresponding flight status risk level. Thirdly, a comprehensive judgment can be made by combining the above two methods to determine the flight status risk level. Therefore, this application does not impose specific restrictions on the method of determining the flight status risk level of a UAV through flight status risk identification processing.

[0205] It is understandable that the classification of flight status risk levels can be determined based on actual circumstances, and this application does not impose specific restrictions on this.

[0206] Step S600: Determine the display information corresponding to the target control key in the remote control terminal according to the risk level of the flight status, and send the display information to the remote control terminal so that the remote control terminal can adjust the display state corresponding to the target control key according to the display information, thereby prompting the operator of the target UAV to operate the target control key in an emergency.

[0207] Specifically, the displayed information includes, but is not limited to, target brightness, display color, and display mode. First, the controller determines the target brightness corresponding to the target control key in the remote control terminal based on the flight status risk level. Different flight status risk levels correspond to different target brightness levels; the higher the flight status risk level, the brighter the target brightness, and vice versa. Second, the controller determines the display color corresponding to the target control key in the remote control terminal based on the flight status risk level. Different flight status risk levels correspond to different display colors. For example, when the flight status risk levels include: a first risk level, a second risk level, and a third risk level from high to low, the display color corresponding to the first risk level is determined to be red, the display color corresponding to the second risk level is determined to be orange, and the display color corresponding to the third risk level is determined to be yellow. Finally, the display mode is determined based on the flight status risk level; the display mode is either a flashing mode or a constant-on mode. For example, the first risk level corresponds to a constant-on mode, while the second and third risk levels correspond to a flashing mode. Based on the target brightness, display color, and display mode, the system determines the display information corresponding to the target control keys and sends this information to the remote control terminal. The remote control terminal then adjusts the display state of the target control keys accordingly. For example, if the display information includes target brightness, a red display color, and a constant-on mode, the brightness of the target control keys is adjusted to the target brightness, and the constant-on display is set to red. This allows the operator of the target drone to activate the target control keys and deploy the parachute in emergency situations, protecting the drone, reducing crash damage, and optimizing the manually triggered parachute deployment mechanism.

[0208] For example, if the onboard sensors detect abnormal data during the flight of a target drone, but the flight attitude information remains normal, the brightness or intensity of the manual parachute deployment button can be dynamically adjusted. This provides a visual cue to warn the pilot of potential flight risks in advance, eliminating the need for the pilot to actively monitor the sensor status. The system automatically identifies anomalies and triggers alerts. By highlighting the button or flashing the cue, the pilot is guided to quickly locate the manual deployment switch, shortening the decision-making time and releasing the parachute. This protects the drone, reduces damage from crashes, and optimizes the manual parachute deployment mechanism.

[0209] This application, through steps S400 to S600, eliminates the need for the drone operator to actively monitor sensor status during the target drone's flight. Instead, it automatically identifies anomalies and triggers alerts when the flight attitude information and operational parameters are normal. After determining the flight status risk level by performing anomaly data identification processing and flight status risk identification processing on the initial sensor data, the brightness of the parachute button is dynamically adjusted based on the flight status risk level. This visual cue warns the drone operator in advance of potential flight risks. Furthermore, by highlighting the button or flashing the prompts, the application guides the drone operator to quickly locate the manual trigger switch, shortening the drone operator's decision-making time in the face of emergencies and enabling timely deployment of the parachute to protect the drone.

[0210] Step S700: Receive the operation result corresponding to the target control key, and obtain the second control result corresponding to the emergency parachute based on the operation result.

[0211] Specifically, the operation results corresponding to the target control key include: the target control key is operated and the operation is successful, the target control key is operated and the operation fails, and the target control key is not operated.

[0212] For example, if the received operation result corresponding to the target control key is that the target control key was operated and the operation was successful, then in response to the operation result, the emergency parachute is controlled, and the corresponding first control result is: the emergency parachute is released. If the received operation result corresponding to the target control key is that the target control key was operated and the operation failed, or the target control key was not operated, then in response to the operation result, the emergency parachute is controlled, and the corresponding first control result is: the emergency parachute is not released and remains in the retracted state.

[0213] In some optional embodiments, the emergency parachute control system of the UAV further includes: a parachute control unit, wherein the controller is electrically connected to the emergency parachute via the parachute control unit; before obtaining a first control result, or before obtaining a second control result, the emergency parachute control method of the UAV includes the following steps:

[0214] Step S800: When the parachute opening judgment result indicates that the emergency parachute needs to open automatically, or the operation result is that the target control key corresponding to the emergency parachute is triggered, the current flight altitude of the target UAV is compared with the preset safe altitude threshold to obtain the eighth comparison result.

[0215] Step S900: If the eighth comparison result indicates that the current flight altitude is greater than the preset safe altitude threshold and the parachute control unit is locked, send an unlock command and a parachute deployment command to the parachute control unit so that the parachute control unit unlocks in response to the unlock command and then controls the emergency parachute to be released in response to the parachute deployment command.

[0216] Step S1000: If the eighth comparison result indicates that the current flight altitude is not greater than the preset safe altitude threshold, determine whether the target UAV has landed and whether the parachute control unit has been unlocked. If yes, send a locking command to the parachute control unit; otherwise, end the control of the parachute control unit.

[0217] Specifically, the unlock command instructs the parachute control unit to unlock the emergency parachute, the deploy command instructs the parachute control unit to deploy the emergency parachute, and the lock command instructs the parachute control unit to retract and lock the emergency parachute.

[0218] Specifically, the preset safety height threshold is 10 meters. The preset safety height threshold can be set according to actual conditions; this application does not impose specific restrictions on the value of the preset safety height threshold.

[0219] Through steps S800 to S1000, when an emergency parachute needs to be released due to an abnormal situation, the emergency parachute is released by controlling the parachute control unit to ensure timely release of the emergency parachute to protect the drone; after the drone lands on the ground, the emergency parachute is automatically and promptly retrieved to prevent the parachute from being accidentally damaged by other objects after landing, thus protecting the emergency parachute.

[0220] To give a specific example, combined with Figure 3 The complete process of determining whether the parachute actively deploys in step S200 of this application embodiment is described below. The complete process of determining whether the parachute actively deploys includes, but is not limited to, steps S1 to S18.

[0221] Step S1: Determine if the emergency parachute on the target drone is powered; if yes, proceed to step S2; if no, the parachute cannot be automatically triggered, proceed to step S18: End.

[0222] Step S2: Determine if the emergency parachute is in standby mode; if yes, proceed to step S3; if no, proceed to step S18: End.

[0223] Step S3: Determine if the parachute is functioning correctly based on the heart rate data; if yes, proceed to step S4; otherwise, proceed to step S18: End.

[0224] Step S4: Determine if the parachute has been released; if yes, proceed to step S5 and step S18 in sequence: End; if no, proceed to step S6.

[0225] Step S5: Lock the motor.

[0226] Step S6: Determine whether the target drone is in stunt mode or flip mode; if yes, proceed to step S18: end; if no, proceed to step S7.

[0227] Step S7: Determine whether the roll attitude is greater than 50 degrees or the pitch attitude is greater than 50 degrees; if yes, proceed to step S18: end; if no, proceed to step S8.

[0228] Step S8: Determine whether the current altitude of the target drone is less than the safe altitude threshold; if yes, determine that the target drone is within the safe altitude range and proceed to step S18: End; if no, proceed to step S9.

[0229] Step S9: Determine if the descent rate is greater than 8 m / s and the duration is greater than 0.8 s; if yes, continue to step S10; if no, return to step S6 to achieve continuous monitoring and judgment. Step S11 is performed simultaneously with step S9.

[0230] Step S10: Determine to deploy the emergency parachute.

[0231] Step S11: Determine whether the control attitude error is greater than 30 degrees; if yes, proceed to step S12; if no, proceed to step S13.

[0232] Step S12: Increase the runaway count; then proceed to step S14.

[0233] Step S13: Reset the runaway count to 0; then continue to step S14.

[0234] Step S14: Determine if the runaway count is equal to 1; if yes, proceed to step S15; if no, proceed to step S16.

[0235] Step S15: Record the initial air pressure altitude; then continue to step S16.

[0236] Step S16: Determine whether the current air pressure altitude is lower than the initial air pressure altitude; if yes, proceed to step S17; if no, proceed to step S13.

[0237] Step S17: Determine if the out-of-control count time is greater than 1 second; if yes, proceed to step S10; if no, proceed to step S18: End.

[0238] Step S18: End.

[0239] The technical effects that can be obtained from steps 1 to 18 in this embodiment are described in the above embodiment, and will not be repeated here.

[0240] The beneficial effects of this application include: by utilizing the controller of the emergency parachute control system of an unmanned aerial vehicle (UAV), during the flight of the target UAV, the initial sensor data of the target UAV is acquired in real time, and the initial sensor data is analyzed to determine the corresponding operating status parameters and flight attitude information of the target UAV; when the flight attitude information or operating status parameters are abnormal, the parachute is actively deployed based on the flight attitude information, operating status parameters, and parachute status information to obtain the deployment judgment result corresponding to the emergency parachute; the emergency parachute is controlled according to the deployment judgment result to obtain the first control result corresponding to the emergency parachute; when the flight attitude information and operating status parameters are normal, the initial sensor data is processed for abnormal data identification to obtain the target abnormal data; the flight status risk is identified based on the target abnormal data to determine the flight status risk level of the target UAV; the display information corresponding to the target control key in the remote control terminal is determined according to the flight status risk level, and the display information is sent to the remote control terminal so that the remote control terminal adjusts the display status corresponding to the target control key according to the display information, thereby prompting the operator of the target UAV to operate the target control key in an emergency; the operation result corresponding to the target control key is received, and the second control result corresponding to the emergency parachute is obtained according to the operation result. The automatic triggering mechanism achieved through comprehensive decision-making helps to effectively reduce the wear and tear of drones. At the same time, by controlling the brightness of the target control key based on the risk level of the drone's flight status, the drone operator is guided to quickly locate the target control key, so as to respond to emergencies in a timely manner, thus optimizing the manual triggering mechanism. This application optimizes the overall emergency parachute control mechanism of drones by being compatible with the optimized automatic triggering mechanism and the manual triggering mechanism, which can effectively reduce the wear and tear of drones in crashes.

[0241] This application provides an emergency parachute control system for a drone, such as... Figure 4 As shown, the emergency parachute control system of this drone includes: a controller; the controller includes:

[0242] The processor 401 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0243] The memory 402 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 402 can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 402 and is called by the processor 401 to execute the emergency parachute control method for the UAV according to the embodiments of this application.

[0244] Input / output interface 403 is used to implement information input and output;

[0245] The communication interface 404 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0246] Bus 405 transmits information between various components of the device (e.g., processor 401, memory 402, input / output interface 403, and communication interface 404);

[0247] The processor 401, memory 402, input / output interface 403 and communication interface 404 are connected to each other within the device via bus 405.

[0248] This application embodiment also provides a storage medium, which is a computer-readable storage medium, storing a computer program that, when executed by a processor, implements the above-described emergency parachute control method for the UAV.

[0249] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate, and may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0250] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0251] The above provides a detailed description of the preferred embodiments of this application. However, this application is not limited to the above-described embodiments. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of this application. All such equivalent modifications or substitutions are included within the scope defined by this application.

Claims

1. A method for controlling an emergency parachute for an unmanned aerial vehicle (UAV), characterized in that, A controller for an emergency parachute control system for a drone, the emergency parachute control system further comprising: an emergency parachute electrically connected to the controller, and a remote control terminal; the remote control terminal including a target control key for manually triggering the release of the emergency parachute; the emergency parachute control method for the drone comprising: During the flight of the target drone, the initial sensor data of the target drone is acquired in real time, and the initial sensor data is analyzed to determine the corresponding operating status parameters and flight attitude information of the target drone. When the flight attitude information is abnormal or the operating status parameter is abnormal, the parachute is actively deployed based on the flight attitude information, the operating status parameter, and the parachute status information to obtain the deployment judgment result corresponding to the emergency parachute. Based on the parachute deployment determination result, the emergency parachute is controlled to obtain the first control result corresponding to the emergency parachute; When the flight attitude information is normal and the operating status parameters are normal, abnormal data identification processing is performed on the initial sensing data to obtain target abnormal data. Based on the abnormal data of the target, flight status risk identification processing is performed to determine the flight status risk level of the target UAV; The display information corresponding to the target control key in the remote control terminal is determined according to the flight status risk level, and the display information is sent to the remote control terminal so that the remote control terminal adjusts the display state corresponding to the target control key according to the display information, thereby prompting the operator of the target UAV to operate the target control key in an emergency. Receive the operation result corresponding to the target control key, and obtain the second control result corresponding to the emergency parachute based on the operation result; The step of performing abnormal data identification processing on the initial sensing data to obtain target abnormal data includes: acquiring window data under the target window by sliding a preset target window in the initial sensing data; performing a first data calculation on the window data to obtain interference data; performing data distribution analysis processing on the initial sensing data based on a Gaussian mixture model to obtain a first distribution result; performing data distribution analysis processing on the interference data based on the Gaussian mixture model to obtain a second distribution result; performing data comparison analysis based on the first distribution result and the second distribution result to determine the Gaussian distribution characteristics corresponding to the initial sensing data; performing data distribution analysis on the initial sensing data based on the Gaussian characteristics to obtain the probability distribution corresponding to the initial sensing data; obtaining the relevant sensing data corresponding to the maximum value of the probability distribution, and calculating the relevant sensing data. The target sensing data is obtained by: 1) determining a target threshold based on the data difference between the initial sensing data and the target sensing data; 2) reassigning and compressing the initial sensing data based on the comparison between the initial sensing data and the target threshold; 3) calculating the density of each first sensing data point in the target sensing data to obtain a first local density value; 4) obtaining the reverse nearest neighbor data corresponding to the first sensing data from the target sensing data, and calculating the local density of the reverse nearest neighbor data to obtain a second local density value; 5) calculating the mean of all the second local density values ​​to obtain an average density value, and determining the outlier representation value corresponding to the first sensing data based on the average density value and the first local density value; and 6) determining the target abnormal data corresponding to the initial sensing data based on the outlier representation value and the target threshold. The step of determining whether the parachute should actively deploy based on the flight attitude information, the operational status parameters, and the parachute status information, and obtaining the deployment determination result corresponding to the emergency parachute, includes: determining the power-on status of the emergency parachute in the parachute status information to obtain a power-on determination result; determining the standby status of the emergency parachute in the parachute status information to obtain a standby determination result; obtaining the heartbeat packet corresponding to the emergency parachute, and determining the connectivity status of the emergency parachute based on the heartbeat packet to obtain a connectivity determination result; and determining the deployment result corresponding to the emergency parachute based on the power-on determination result, the standby determination result, and the connectivity determination result. First judgment result: When the first judgment result indicates that the emergency parachute is in a normal power-on, normal standby, and normal connectivity state, it is determined whether the current deployment state of the emergency parachute is in the retracted state; when the current deployment state of the emergency parachute is not in the retracted state, the opening judgment result corresponding to the emergency parachute is determined as the first result, which is used to indicate that the emergency parachute cannot open; when the current deployment state of the emergency parachute is in the retracted state, the deployment control of the emergency parachute is judged and processed according to the flight attitude information and the operating state parameters to obtain the opening judgment result corresponding to the emergency parachute.

2. The emergency parachute control method for a UAV according to claim 1, characterized in that, The step of performing a first data calculation on the window data to obtain interference data includes: The average data is obtained by calculating the mean of the window data; Calculate the absolute difference between each sub-data point in the window data and the average data, and calculate the mean of all the absolute differences to obtain the deviation value; Based on the deviation value, an interference value corresponding to the window data is generated; The interference value is added to the window data to obtain the interference data.

3. The emergency parachute control method for a UAV according to claim 1, characterized in that, The step of performing density calculation on each first sensing data in the target sensing data to obtain a first local density value corresponding to the first sensing data includes: Determine the target quantity corresponding to the target sensing data, and perform a logarithmic operation on the target quantity to obtain the number of nearest neighbors required for the target sensing data to perform nearest neighbor calculation; The first sensing data is determined from the target sensing data, and the remaining sensing data corresponding to the first sensing data is obtained by removing the first sensing data from the target sensing data. The target distance is obtained by calculating the distance between each second sensor data and the first sensor data in the remaining sensor data; The remaining sensor data is sorted according to the target distance to obtain target sorted data, and the first neighbor sensor data corresponding to the first sensor data is obtained from the target sorted data according to the number of neighbors. The first nearest neighbor sensing data and the relevant distance between the first sensing data are obtained from the target distance, and the first local density value corresponding to the first sensing data is obtained by reciprocal calculation based on the relevant distance.

4. The emergency parachute control method for a UAV according to claim 3, characterized in that, The step of obtaining the reverse nearest neighbor data corresponding to the first sensing data from the target sensing data, and performing local density calculation on the reverse nearest neighbor data to obtain a second local density value, includes: Perform nearest neighbor calculation processing on each of the second sensor data in the remaining sensor data to obtain the second nearest neighbor sensor data corresponding to the second sensor data; When the second nearest neighbor sensing data contains the first sensing data, the second sensing data is determined as the reverse nearest neighbor data of the first sensing data; Local density calculation is performed on each third sensor data in the reverse nearest neighbor data to obtain the second local density value corresponding to each third sensor data.

5. The emergency parachute control method for a UAV according to claim 1, characterized in that, The flight attitude information includes attitude angles and control attitude errors; the operating status parameters include the target UAV's operating mode, current altitude above ground, current barometric altitude, altitude descent rate, and the duration corresponding to the altitude descent rate; the judgment and processing of the emergency parachute deployment and recovery control based on the flight attitude information and the operating status parameters to obtain the parachute deployment judgment result includes: When the working mode is the preset mode, the opening judgment result corresponding to the emergency parachute is determined to be the second result, and the second result is used to indicate that the emergency parachute does not need to be opened; When the working mode is not the preset mode, the attitude angle is compared with the angle threshold to obtain a first comparison result; if the first comparison result indicates that the attitude angle is greater than the angle threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined to be a third result, which indicates that the emergency parachute needs to be deployed; if the first comparison result indicates that the attitude angle is not greater than the angle threshold, the current ground clearance is compared with the safe altitude threshold to obtain a second comparison result; if the second comparison result indicates that the current ground clearance is less than the safe altitude threshold, the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result. If the second comparison result indicates that the current altitude is not less than the safe altitude threshold, the altitude descent rate is compared with a preset descent rate threshold to obtain a third comparison result, and the duration is compared with a first time threshold to obtain a fourth comparison result; if the third comparison result indicates that the altitude descent rate is greater than the preset descent rate threshold and the fourth comparison result indicates that the duration is greater than the first time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the third result. If the third comparison result indicates that the altitude descent rate is not greater than the preset descent rate threshold, or the fourth comparison result indicates that the duration is not greater than the first time threshold, the third comparison result and the fourth comparison result are continuously monitored and updated, and the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result; and the parachute deployment judgment result is continuously updated using the updated third comparison result and the fourth comparison result; and the control attitude error is further compared with the attitude error threshold to obtain a fifth comparison result; If the fifth comparison result indicates that the control attitude error is not greater than the attitude error threshold, the runaway count is reset to 0; if the fifth comparison result indicates that the control attitude error is greater than the attitude error threshold, the runaway count is incremented by 1, and if the runaway count is 1, the current air pressure altitude is determined as the initial air pressure altitude; if the runaway count is not 1, the latest air pressure altitude is obtained, the latest air pressure altitude is assigned to the current air pressure altitude, and the latest air pressure altitude and the initial air pressure altitude are compared to obtain the sixth comparison result. If the sixth comparison result indicates that the latest air pressure altitude is not lower than the initial air pressure altitude, the runaway count is reset; if the sixth comparison result indicates that the latest air pressure altitude is lower than the initial air pressure altitude, the execution time of the runaway count is compared with the second time threshold to obtain a seventh comparison result. If the seventh comparison result indicates that the execution time is not greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the second result; if the sixth comparison result indicates that the execution time is greater than the second time threshold, then the parachute deployment judgment result corresponding to the emergency parachute is determined to be the third result.

6. The emergency parachute control method for a UAV according to any one of claims 1 to 5, characterized in that, The emergency parachute control system of the UAV further includes: a parachute control unit, wherein the controller is electrically connected to the emergency parachute via the parachute control unit; before obtaining the first control result, or before obtaining the second control result, the emergency parachute control method of the UAV includes: When the parachute deployment judgment result indicates that the emergency parachute needs to deploy automatically, or when the operation result is to trigger the target control key corresponding to the emergency parachute, the current flight altitude of the target UAV is compared with the preset safe altitude threshold to obtain the eighth comparison result; If the eighth comparison result indicates that the current flight altitude is greater than the preset safe altitude threshold and the parachute control unit is locked, an unlock command and a parachute deployment command are sent to the parachute control unit so that the parachute control unit unlocks in response to the unlock command and then controls the emergency parachute to deploy in response to the parachute deployment command. If the eighth comparison result indicates that the current flight altitude is not greater than the preset safe altitude threshold, it is determined whether the target drone has landed and whether the parachute control unit has been unlocked. If yes, a locking command is sent to the parachute control unit; otherwise, control of the parachute control unit is terminated.

7. An emergency parachute control system for an unmanned aerial vehicle (UAV), characterized in that, The system includes a controller, which includes at least one processor and a memory for communicatively connecting to the at least one processor; the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the emergency parachute control method for a drone as described in any one of claims 1 to 6.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform the emergency parachute control method for a drone as described in any one of claims 1 to 6.

Images