Method, device and equipment for predicting flight state of unmanned aerial vehicle, medium and product

Abstract

The invention provides an unmanned aerial vehicle flight state prediction method and device, equipment, a medium and a product. The method comprises the following steps: inputting prediction state data and trajectory data at the current moment into an external disturbance prediction model to obtain a target external disturbance type output by the external disturbance prediction model; and then determining an external disturbance compensation amount according to a target external disturbance type and sensor measurement data. And finally, according to the predicted state data at the current moment, the control data of the user for the unmanned aerial vehicle and the external disturbance compensation amount, determining predicted state data of the unmanned aerial vehicle at the next moment. The problem that the flight state prediction accuracy of the unmanned aerial vehicle is low is solved.

Description

Technical Field

[0001] The present application relates to the field of power inspection technology, and in particular to a method, device, equipment, medium and product for predicting the flight status of a drone. Background Art

[0002] In the field of power inspection, drones are widely used for inspections of transmission lines, substations, and other equipment due to their flexibility and efficiency, significantly improving inspection efficiency and quality. In complex inspection environments, drones are susceptible to wind speed fluctuations, electromagnetic interference, Global Positioning System (GPS) signal degradation, and sensor errors. These factors can lead to inaccurate state predictions, thus compromising autonomous navigation and control.

[0003] In the existing technology, there are three main solutions to this problem: (1) Extended Kalman Filter (EKF) method, which estimates the state information of the UAV in real time by combining nonlinear system linearization with state prediction and update; (2) Unscented Kalman Filter (UKF) method, which improves the accuracy of trajectory estimation by adopting unscented transformation; (3) Trajectory prediction method based on deep learning, which predicts the future flight status of the UAV based on historical trajectory data.

[0004] However, the existing technology has the problem of low accuracy in predicting the flight status of drones. Summary of the Invention

[0005] The embodiments of the present application provide a method, apparatus, device, medium, and product for predicting the flight status of a drone, so as to solve the problem of low accuracy in predicting the flight status of a drone.

[0006] In a first aspect, an embodiment of the present application provides a method for predicting the flight status of a drone, comprising:

[0007] Input the current predicted state data and trajectory data into the external disturbance prediction model to obtain the target external disturbance type output by the external disturbance prediction model. The external disturbance prediction model is trained based on the sample state data, sample trajectory data, and sample external disturbance types at historical moments.

[0008] Determine the external disturbance compensation amount based on the target external disturbance type and sensor measurement data;

[0009] The predicted state data of the drone at the next moment is determined based on the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

[0010] In a possible implementation, a weight matrix is ​​determined according to the target external disturbance type; and an external disturbance compensation amount is determined according to the weight matrix and sensor measurement data.

[0011] In one possible implementation, the objective function parameters are determined based on the target external disturbance type and the mapping relationship, and the mapping relationship is used to represent the correspondence between the external disturbance type and the function parameters; the weight matrix is ​​determined through the weight matrix function based on the objective function parameters, auxiliary variables, positive definite matrix and sensor measurement data.

[0012] In a possible implementation, the predicted state data at the next moment is updated according to the sensor measurement data and the Kalman gain to generate updated predicted state data at the next moment.

[0013] In a possible implementation, the trajectory data includes predicted trajectory sub-data and actual trajectory sub-data.

[0014] In a possible implementation, the external disturbance compensation amount is determined according to the positive definite matrix, the weight matrix, and sensor measurement data.

[0015] In a second aspect, an embodiment of the present application provides a device for predicting the flight status of a drone, comprising:

[0016] An acquisition module is used to input the predicted state data and trajectory data at the current moment into the external disturbance prediction model to obtain the target external disturbance type output by the external disturbance prediction model. The external disturbance prediction model is trained based on sample state data, sample trajectory data, and sample external disturbance types at historical moments.

[0017] A first determination module is used to determine an external disturbance compensation amount according to a target external disturbance type and sensor measurement data;

[0018] The second determination module is used to determine the predicted state data of the drone at the next moment based on the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

[0019] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory, a processor;

[0020] Memory stores computer-executable instructions;

[0021] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the first aspect above and / or various possible implementation methods of the first aspect.

[0023] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementation methods of the first aspect.

[0024] The embodiments of the present application provide a method, device, equipment, medium and product for predicting the flight status of a drone. By inputting the predicted state data and trajectory data at the current moment into an external disturbance prediction model, the target external disturbance type output by the external disturbance prediction model is obtained. The external disturbance prediction model is trained based on sample state data, sample trajectory data and sample external disturbance types at historical moments. Then, the external disturbance compensation amount is determined based on the target external disturbance type and sensor measurement data. Finally, based on the predicted state data at the current moment, the user's control data for the drone and the external disturbance compensation amount, the predicted state data of the drone at the next moment is determined, thereby solving the problem of low accuracy in predicting the flight status of the drone. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0026] Figure 1 A flowchart of a method for predicting the flight status of a drone provided in an embodiment of the present application Figure 1 ;

[0027] Figure 2 A flowchart of a method for predicting the flight status of a drone provided in an embodiment of the present application Figure 2 ;

[0028] Figure 3 A schematic diagram of the results of a method for predicting the flight status of a drone provided in an embodiment of the present application;

[0029] Figure 4 A schematic diagram of the structure of a device for predicting the flight status of a drone provided in an embodiment of the present application;

[0030] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0031] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0032] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0033] First, let’s explain the terms involved in this application:

[0034] Extended Kalman Filter: refers to a Kalman filter method suitable for nonlinear systems. It achieves local linearization by performing a first-order Taylor expansion on the system model, and is then applied to state estimation.

[0035] Unscented Kalman filter: refers to a nonlinear filtering method based on unscented transformation. It approximates the state distribution by selecting a set of deterministic sample points, avoiding the explicit linearization of the nonlinear model.

[0036] Finite-time neural network disturbance observer: This is an observer designed using a neural network structure that can quickly and accurately estimate the external disturbance to the system within a finite time. It is often used to improve the robustness and response speed of control systems to disturbances.

[0037] Adaptive weight adjustment strategy: refers to a method that dynamically adjusts the weights of various parameters or input signals in the model according to the system operating status and environmental changes to optimize performance, improve accuracy and robustness.

[0038] In the field of power inspection technology, drones are widely used for inspections of transmission lines, substations, and other equipment due to their flexibility and efficiency, significantly improving inspection efficiency and quality. However, in complex inspection environments, drones are susceptible to external disturbances such as wind speed fluctuations, electromagnetic interference, GPS signal weakening, and sensor errors. These factors severely reduce the accuracy of drone trajectory tracking and may even lead to flight deviations or loss of control.

[0039] In order to solve the impact of external disturbances on the autonomous navigation and control of UAVs, existing technologies generally use technical solutions based on three directions: (1) EKF method, which estimates the position, velocity and attitude information of UAVs in real time by combining the linearization of nonlinear systems with state prediction and update steps. However, the performance of this method is significantly limited in highly nonlinear environments or severe interference conditions; (2) UKF method, which improves the accuracy of trajectory estimation by adopting untraceable transformation, avoids the linearization error of EKF, and can more accurately approximate the nonlinear state distribution, showing better performance in high dynamic environments and scenes with severe electromagnetic interference. However, UKF has high computational complexity and needs to balance the relationship between real-time performance and accuracy; (3) Trajectory prediction method based on deep learning, which uses recurrent neural network models such as long short-term memory networks to predict the future flight status of UAVs based on historical trajectory data, avoiding excessive reliance on physical dynamic models. However, this type of method has strict requirements on the quantity and quality of training data, and is prone to prediction errors when the inspection environment varies greatly or data is insufficient.

[0040] In summary, the existing technology still has the problem of low accuracy in predicting the flight status of drones.

[0041] Based on the above technical problems, the invention concept of this application is as follows: Since different types of external disturbances have different effects on the prediction process of the drone's flight state, the external disturbance compensation amount can be generated according to the drone's current external disturbance type, and then the predicted state data for the next moment can be generated according to the external disturbance compensation amount, so as to make a more accurate prediction of the drone's state and solve the problem of low accuracy in predicting the drone's flight state.

[0042] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0043] Figure 1 A flowchart of a method for predicting the flight status of a drone provided in an embodiment of the present application Figure 1 ,like Figure 1 As shown, the method includes:

[0044] S101: Input the predicted state data and trajectory data at the current moment into an external disturbance prediction model to obtain a target external disturbance type output by the external disturbance prediction model.

[0045] Among them, predicted state data refers to the predicted state information of the drone, including position, speed, acceleration, etc.

[0046] Trajectory data refers to the motion path of a UAV, typically consisting of a series of position points. In one possible implementation, trajectory data includes predicted trajectory sub-data and actual trajectory sub-data.

[0047] Among them, the predicted trajectory sub-data refers to the predicted flight path of the UAV in the future, which is used for the UAV's path planning and control decision-making.

[0048] The actual trajectory sub-data refers to the flight path of the UAV during the actual flight process, which is usually directly measured and recorded by sensors, reflecting the position information of the UAV at each time point.

[0049] The external disturbance prediction model is obtained by training based on sample state data, sample trajectory data and sample external disturbance types at historical moments.

[0050] Among them, historical moments can be flexibly selected according to needs. For example, the drone flight data within three days before the current moment can be selected, and each moment within these three days can be used as a historical moment.

[0051] The sample state data refers to the state information recorded by the drone’s sensors at a historical moment, including position, speed, acceleration, etc. The sample trajectory data refers to the trajectory information formed during the drone’s flight process at a historical moment.

[0052] The sample external disturbance type refers to the type of disturbance that the UAV is subjected to during flight at a historical moment, including wind, electromagnetic interference, and terrain influence.

[0053] The target external disturbance type refers to the external disturbance type determined based on the predicted state data and trajectory data at the current moment, and is output based on the pre-trained external disturbance prediction model.

[0054] In practical applications, sample status data can be acquired in real time through the drone’s multi-sensor fusion technology, such as the Inertial Measurement Unit (IMU), GPS, LiDAR, etc.

[0055] S102: Determine an external disturbance compensation amount according to the target external disturbance type and sensor measurement data.

[0056] Among them, sensor measurement data refers to the data measured by the sensor in real time during the flight, including position, speed, acceleration, etc.

[0057] In practical applications, the data measured by the sensor needs to be denoised and filtered to ensure the reliability of the data. Denoising and filtering are conventional technical means and are not limited in this application.

[0058] The external disturbance compensation refers to the value that needs to be corrected for the state prediction of the UAV for the target external disturbance type, so that the UAV has stronger environmental adaptability.

[0059] In practical applications, a neural network can be combined to determine the external disturbance compensation amount corresponding to the target external disturbance type. In one possible implementation, a weight matrix is ​​first determined based on the target external disturbance type, and then the external disturbance compensation amount is determined based on the weight matrix and sensor measurement data.

[0060] The weight matrix reflects how different types of external disturbances affect the drone's various state information. For example, for wind disturbances, the weight matrix describes how the wind changes the drone's position and velocity; for electromagnetic interference, the weight matrix describes how electromagnetic interference affects sensor readings and estimation errors.

[0061] It should be understood that in order to optimize the trajectory tracking performance of the UAV, an adaptive weight adjustment strategy is adopted to dynamically adjust the weight matrix according to the sensor data of the UAV at different flight phases to further improve the reliability of trajectory estimation.

[0062] Therefore, in one possible implementation method, the weight matrix can be determined according to the target external disturbance type and the mapping relationship, after determining the objective function parameters, and then the weight matrix can be determined through the weight matrix function according to the objective function parameters, auxiliary variables, positive definite matrix and sensor measurement data.

[0063] The mapping relationship is used to represent the correspondence between the external disturbance type and the function parameter. Different external disturbance types correspond to different function parameters.

[0064] Function parameters refer to the parameters in the neural network's function for calculating the weight matrix.

[0065] The objective function is the function for calculating the weight matrix, which can be expressed by the formula: Calculate the weight matrix, where Used to represent the updated weight matrix, Used to represent the weight matrix before updating, Used to express the convergence gain, It is used to represent positive definite matrices, Used to represent auxiliary variables, and Used to represent function parameters, Used to represent sensor measurement data.

[0066] In practical applications, after obtaining the target external disturbance type, the corresponding external disturbance type can be selected. 、 , thereby determining the weight matrix , and according to the real-time measurement data of the sensor, the weight matrix After updating, you get .

[0067] In a possible implementation, the external disturbance compensation amount is determined based on a positive definite matrix, a weight matrix, and sensor measurement data.

[0068] In practical applications, the external disturbance compensation can be determined based on the finite time neural network disturbance observer. The positive definite matrix is ​​used to control the dynamic behavior of the observer. ,in, It is used to represent the external disturbance compensation amount. It can be seen that by determining the updated weight matrix and combining it with the positive definite matrix and sensor measurement data, the external disturbance compensation amount can be updated.

[0069] It should be understood that the calculated external disturbance compensation amount can be introduced into the state prediction equation to reduce the impact of external disturbance on the system.

[0070] S103: Determine the predicted state data of the drone at the next moment based on the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

[0071] Among them, the user's control data for the drone refers to the thrust, direction angle and other data required to control the flight of the drone.

[0072] After determining the types of external disturbances that the UAV may encounter and the amount of external disturbance compensation that needs to be compensated, the external disturbance compensation amount can be used to optimize the state equation of the UAV.

[0073] In practical applications, the formula: = + , determine the predicted state data of the drone at the next moment.

[0074] in, Used to represent the current moment's forecast status data, Used to represent the predicted state data for the next moment, Used to represent the user's control data for the drone. It is used to indicate the external disturbance compensation amount. The specific form of is flexibly set according to the needs. For example, if the dynamic model of the UAV is based on Newton's laws of motion, then It should include the relationship between position, velocity and acceleration.

[0075] In practical applications, it is also necessary to correct the predicted state data through the update step of the extended Kalman filter to improve the stability and accuracy of the trajectory estimation. Therefore, in one possible implementation method, the predicted state data at the next moment is updated according to the sensor measurement data and the Kalman gain to generate the updated predicted state data at the next moment.

[0076] Among them, combined with the prediction model of the extended Kalman filter, through the formula: = , calculate the updated predicted state data for the next moment.

[0077] in, Used to represent the updated forecast state data for the next moment. is used to represent the Kalman gain, The specific form of is flexibly set according to the needs. It is a measurement model used to convert the real state value of the drone into a measurement value.

[0078] In practical applications, the measured value of the sensor needs to be expressed by the formula Perform the conversion, where Used to represent the real status data of the drone, Used to represent measurement noise.

[0079] And for the Kalman gain , through the formula: Calculated. Among them, is used to represent the forecast covariance, Used to represent the observation matrix, Used to represent the measurement noise covariance matrix.

[0080] The predicted covariance , through the formula: Calculated, where Used to represent the state transition matrix, Also used to represent the measurement noise covariance matrix.

[0081] It can be seen that this application introduces the external disturbance compensation amount output by the neural network in real time into the EKF prediction model, effectively reducing the model error in complex inspection environments and significantly improving the accuracy and stability of EKF's state prediction and update.

[0082] An embodiment of the present application provides a method for predicting the flight status of a drone. An external disturbance prediction model is pre-trained using sample status data, sample trajectory data, and sample external disturbance types at historical moments. The predicted status data and trajectory data at the current moment can be input into the external disturbance prediction model to determine the type of external disturbance to which the drone may be subjected. The external disturbance compensation amount can be determined based on the external disturbance type and sensor measurement data, so that the predicted status data at the current moment and the user's control data for the drone can be combined to determine the predicted status data of the drone at the next moment, thereby solving the problem of low accuracy in predicting the flight status of the drone.

[0083] In practical applications, the real state data of the UAV is affected by many factors, such as external wind field, sensor noise and system control input, so the state equation of the UAV can be established as , to describe the real state of the drone, where Used to represent the real state data of the drone at the next moment. Used to represent the real status data of the drone at the current moment. Used to represent the user's control data for the drone. Used to represent external disturbance data, Used to represent process noise.

[0084] It should be understood that at this time, the control equation for the real state of the drone , which can be used to predict the state data of the drone at the next moment.

[0085] In practical applications, the formula: , to determine the external disturbance data, where Used to represent the random error of the disturbance, The dependency term used to represent the disturbance can be determined according to the requirements. It can be seen that the external disturbance data at this time is determined based on the true state value and the random error of the disturbance.

[0086] In practical applications, external disturbance prediction models can be combined with FTNNDO's real-time disturbance estimation method to accurately model and estimate external disturbances (such as wind speed changes and electromagnetic interference) through neural network training. This leverages the universal approximation properties of neural networks to model external disturbances that may occur during flight in real time. This neural network, employing a multi-layer long short-term memory (LSTM) or attention model (Transformer) structure, can learn sample state and trajectory data from historical moments, thereby dynamically adjusting the accuracy of disturbance estimation and significantly reducing the nonlinear error introduced by external disturbances.

[0087] Figure 2 A flowchart of a method for predicting the flight status of a drone provided in an embodiment of the present application Figure 2 ,like Figure 2 As shown, this embodiment Figure 2 Based on the embodiment, a method for predicting the flight status of a drone is described in detail. The method includes:

[0088] S201. Obtain the status data of the UAV in real time through sensors. The status data includes position, speed, acceleration, etc., and then input the status data into the neural network interference observer to estimate the impact of external disturbances on the UAV flight trajectory.

[0089] S202. Based on a finite-time neural network interference observer, dynamically model and estimate the external interference, and output the external interference compensation amount in real time.

[0090] S203. Introduce the external interference compensation into the extended Kalman filter, modify the state equation, and complete the preliminary prediction of the UAV trajectory.

[0091] S204. In combination with the sensor measurement data, the state prediction value is corrected by extending the Kalman filter to obtain an updated state prediction value, thereby determining a more accurate UAV trajectory estimate.

[0092] The present invention provides a method for predicting the flight state of a drone. Sensors are used to obtain the drone's position, velocity, and other state data. A neural network interference observer is then used to estimate external disturbances in real time and output compensation values. The compensation values ​​are then introduced into an EKF (Extended Kernel Function) to modify the state equation and perform a preliminary trajectory prediction. Finally, the predicted values ​​are corrected using an EKF (Extended Kernel Function) based on the sensor measurement data to obtain a more accurate drone trajectory estimate. This method addresses the issue of low accuracy in drone flight state prediction and enables efficient and accurate flight trajectory tracking.

[0093] Figure 3 This is a schematic diagram of the results of a method for predicting the flight status of a drone provided in an embodiment of the present application, as shown in FIG. Figure 3 As shown in Figure 3, the EKF predicted trajectory with a state observer is closer to the actual trajectory of the UAV than the EKF predicted trajectory without a state observer because it introduces external disturbance compensation into the state equation.

[0094] The method for predicting the flight status of a UAV in the present application can effectively reduce the impact of sudden changes in wind speed and electromagnetic interference on the estimation of the UAV trajectory, and improve the accuracy and stability of the UAV on the inspection route. At the same time, the present application optimizes the EKF calculation process, reduces dependence on GPS data, improves the anti-interference ability of the system, and can maintain high-precision positioning even in an environment with damaged signals. In addition, by utilizing the learning ability of neural networks, UAVs can autonomously optimize external disturbance prediction models during long-term flight, enhance their adaptability to different inspection environments, and ensure that the system has a higher level of intelligence. This method not only improves the accuracy, real-time nature, and computational efficiency of trajectory estimation, but also has strong scalability. It can be widely used in scenarios such as UAV inspection, border monitoring, forest fire prevention, and agricultural remote sensing, providing more stable, more accurate, and more intelligent technical support for the development of power inspection and the UAV industry.

[0095] Figure 4 A schematic diagram of a device for predicting the flight status of a drone provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the drone flight state prediction device 40 provided in this embodiment includes:

[0096] An acquisition module 401 is used to input the predicted state data and trajectory data at the current moment into an external disturbance prediction model to obtain the target external disturbance type output by the external disturbance prediction model. The external disturbance prediction model is trained based on sample state data, sample trajectory data, and sample external disturbance types at historical moments.

[0097] A first determination module 402 is configured to determine an external disturbance compensation amount based on the target external disturbance type and sensor measurement data;

[0098] The second determination module 403 is used to determine the predicted state data of the drone at the next moment based on the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

[0099] In a possible implementation, the first determination module 402 is further configured to determine a weight matrix according to the target external disturbance type; and determine an external disturbance compensation amount according to the weight matrix and sensor measurement data.

[0100] In one possible implementation, the first determination module 402 is also used to determine the objective function parameters based on the target external disturbance type and the mapping relationship, and the mapping relationship is used to represent the correspondence between the external disturbance type and the function parameters; and determine the weight matrix through the weight matrix function based on the objective function parameters, auxiliary variables, positive definite matrix and sensor measurement data.

[0101] In a possible implementation, the second determining module 403 is further configured to update the predicted state data at the next moment according to the sensor measurement data and the Kalman gain, and generate updated predicted state data at the next moment.

[0102] In a possible implementation, the first determination module 402 is further configured to determine an external disturbance compensation amount according to the positive definite matrix, the weight matrix, and sensor measurement data.

[0103] The device for predicting the flight status of a drone provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effects are similar, and are not described in detail in this embodiment.

[0104] Figure 5 This is a schematic diagram of the structure of the electronic device provided in the embodiment of the present application. Figure 5 As shown, the electronic device 50 provided in this embodiment includes: at least one processor 501 and a memory 502. Optionally, the electronic device 50 further includes a communication component 503. The processor 501, the memory 502 and the communication component 503 are connected via a bus 504.

[0105] In a specific implementation process, at least one processor 501 executes the computer-executable instructions stored in the memory 502, so that the at least one processor 501 performs the above method.

[0106] The specific implementation process of the processor 501 can be found in the above method embodiment. Its implementation principle and technical effects are similar and will not be repeated here in this embodiment.

[0107] In the above embodiments, it should be understood that the processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASICs), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly executed by a hardware processor or by a combination of hardware and software modules within the processor.

[0108] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage.

[0109] A bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Buses can be categorized as address buses, data buses, and control buses. For ease of illustration, the buses in the drawings of this application are not limited to just one bus or just one type of bus.

[0110] The present application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0111] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above method is implemented.

[0112] The readable storage medium may be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium may be any available medium that can be accessed by a general-purpose or special-purpose computer.

[0113] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be an integral part of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0114] The division of units is merely a logical functional division; actual implementations may employ alternative divisions, such as combining or integrating multiple units or components into another system, or omitting or disabling certain features. Furthermore, any direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units, either through an interface, electrical, mechanical, or other means.

[0115] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0116] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0117] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0118] Those skilled in the art will appreciate that all or part of the steps in the above-described method embodiments can be implemented using hardware associated with program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0119] Finally, it should be noted that those skilled in the art will readily identify other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art not disclosed herein. The present invention is not limited to the precise structure described above and illustrated in the accompanying drawings, and various modifications and variations may be made without departing from the scope thereof. The scope of the present invention is limited solely by the appended claims.

Claims

1. A method for predicting the flight status of a drone, characterized in that: include: Inputting the predicted state data and trajectory data at the current moment into an external disturbance prediction model to obtain a target external disturbance type output by the external disturbance prediction model, wherein the external disturbance prediction model is trained based on sample state data, sample trajectory data, and sample external disturbance types at historical moments; Determining an external disturbance compensation amount according to the target external disturbance type and sensor measurement data; The predicted state data of the drone at the next moment is determined according to the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

2. The method according to claim 1, characterized in that The determining of the external disturbance compensation amount according to the target external disturbance type and sensor measurement data includes: Determining a weight matrix according to the target external disturbance type; The external disturbance compensation amount is determined according to the weight matrix and the sensor measurement data.

3. The method according to claim 2, characterized in that Determining a weight matrix according to the target external disturbance type includes: Determining objective function parameters according to the target external disturbance type and a mapping relationship, wherein the mapping relationship is used to represent a correspondence between the external disturbance type and the function parameter; The weight matrix is ​​determined by a weight matrix function according to the objective function parameters, auxiliary variables, a positive definite matrix and the sensor measurement data.

4. The method according to any one of claims 1 to 3, characterized in that The method further comprises: The predicted state data at the next moment is updated according to the sensor measurement data and the Kalman gain to generate updated predicted state data at the next moment.

5. The method according to any one of claims 1 to 3, characterized in that The trajectory data includes predicted trajectory sub-data and actual trajectory sub-data.

6. The method according to claim 2 or 3, characterized in that The determining the external disturbance compensation amount according to the weight matrix and the sensor measurement data includes: The external disturbance compensation amount is determined according to the positive definite matrix, the weight matrix and the sensor measurement data.

7. A device for predicting the flight status of a drone, characterized in that: include: An acquisition module is used to input the predicted state data and trajectory data at the current moment into an external disturbance prediction model to obtain the target external disturbance type output by the external disturbance prediction model, where the external disturbance prediction model is trained based on sample state data, sample trajectory data, and sample external disturbance types at historical moments; A first determination module is configured to determine an external disturbance compensation amount according to the target external disturbance type and sensor measurement data; The second determination module is used to determine the predicted state data of the drone at the next moment according to the predicted state data at the current moment, the user's control data for the drone, and the external disturbance compensation amount.

8. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 6 when executed by a processor.

10. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 6 when the computer program is executed by a processor.

Images