High-altitude cleaning unmanned aerial vehicle operation trajectory tracking control method

Abstract

The invention relates to the technical field of aircraft flight control, in particular to a high-altitude cleaning unmanned aerial vehicle operation trajectory tracking control method, which constructs a cooperative control system of a self-adaptive prediction time domain model prediction control framework and an event triggering execution mechanism. A controller can select an optimal prediction time domain parameter according to the speed information of the unmanned aerial vehicle, the calculation complexity is effectively reduced while the trajectory tracking precision is ensured, an event triggering mechanism is introduced to intelligently regulate and control the rolling optimization process of MPC, and by setting the dual triggering conditions of a speed threshold and a trajectory tracking error threshold, the trajectory tracking precision is improved. And rolling optimization solution is only executed at the key control node, so that the real-time performance of the system is remarkably improved, and an innovative control architecture is provided for technical breakthrough of the unmanned aerial vehicle in the field of special operation.

Description

Technical Field

[0001] This invention relates to the field of aircraft flight control technology, and more specifically to a method for tracking and controlling the operational trajectory of a high-altitude cleaning drone. Background Technology

[0002] In heavy industries such as steel, emissions generated during production easily adhere to the surfaces of factory buildings and chimneys, forming a layer of filth that is difficult to remove. This filth layer not only affects the appearance of the equipment but also accelerates its corrosion, thereby impacting production efficiency and safety. Therefore, regular cleaning is of paramount importance. However, traditional manual cleaning methods are not only labor-intensive and involve dangerous working environments but are also inefficient. With the advancement of technology, quadcopter drones have gradually gained widespread application in industries such as industry, agriculture, and forestry due to their advantages of simple operation, high reliability, and low maintenance costs. As a result, the development of a high-altitude cleaning system based on quadcopter drones has become a research direction that has attracted much attention.

[0003] Unmanned aerial vehicles (UAVs) are multi-input multi-output, strongly coupled, and underactuated systems. Their application in various complex environments requires solving the stability and accuracy problems of trajectory tracking control. This has prompted researchers to continuously explore and develop advanced control methods to meet the needs of UAVs in various application scenarios. For example, Chinese patents with announcement numbers CN120353246B, CN113961010B, and CN112241125B have all studied different trajectory tracking control methods for UAVs. Currently, quadcopter UAVs equipped with high-pressure spray booms require precise trajectory tracking control when performing cleaning operations. However, traditional controllers are difficult to guarantee the accuracy and real-time performance of the operation trajectory tracking control in such complex working environments, resulting in unsatisfactory cleaning results. Therefore, a high-altitude cleaning UAV operation trajectory tracking control method is proposed. Summary of the Invention

[0004] To address the technical problems existing in the prior art, the present invention provides a method for tracking and controlling the operational trajectory of a high-altitude cleaning drone.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for tracking and controlling the operational trajectory of a high-altitude cleaning drone, the control method comprising the following steps:

[0006] S1. A dynamic model of the high-altitude cleaning drone is established based on Euler's equations and Newton's laws.

[0007] S2 transforms the dynamic model of the high-altitude cleaning UAV into a state-space equation and discretizes it.

[0008] S3. Establish the position controller and attitude controller of the high-altitude cleaning drone. The position controller is used to control the drone's altitude and xy plane, and the attitude controller is used to control the drone's roll, pitch and yaw three attitude angles.

[0009] S4. Collect flight data of different prediction time-domain parameters corresponding to different speeds of the high-altitude cleaning drone, determine the correlation between the prediction time-domain parameters and the evaluation items through the grey relational method, select the optimal prediction time-domain parameters, and fit them through a cubic polynomial.

[0010] S5, when tracking the operation trajectory of the high-altitude cleaning drone, dynamically adjusts the prediction time-domain parameters of the MPC controller based on the drone's speed information. During adjustment, an event triggering mechanism is introduced to intelligently control the rolling optimization process of the MPC controller. By setting dual triggering conditions of speed threshold and trajectory tracking error threshold, rolling optimization is only performed at key control nodes.

[0011] Preferably, the dynamic model of the high-altitude cleaning UAV is expressed as follows:

[0012] ;

[0013] in, Let x, y, and z be the mass and gravitational acceleration of the drone, respectively, and let x, y, and z be the position coordinates of the drone. These are the roll angle, pitch angle, and yaw angle of the drone, respectively. The quadcopter drones along axis, axis, Moment of inertia of the shaft These are the three components of the angular velocity along the body axis. The length between the center of the drone and the rotor. The total thrust generated by the propeller. The torque generated on the propeller These represent the accelerations of the drone in three directions, respectively. This represents the angular acceleration of the drone in three directions.

[0014] Preferably, the The specific expression is as follows:

[0015] ;

[0016] in, Let be the angular velocity of the drone rotor. The lift coefficient, This is the torque coefficient.

[0017] Preferably, step S2 includes the following steps:

[0018] S21, the state-space equation expression of the high-altitude cleaning UAV position system is:

[0019] ;

[0020] in, and The specific expression is: ;

[0021] The discrete model expression for the UAV's position is:

[0022] ;

[0023] in, This is the position error vector. To control the input error vector;

[0024] The state-space equation expression for the attitude system of the S22 high-altitude cleaning UAV is as follows:

[0025] ;

[0026] The discrete model expression for the attitude component of the UAV is:

[0027] ;

[0028] in, Let be the state vector of the attitude system. To control the input vector, The state matrix, The input matrix;

[0029] S23, by constructing a virtual reference drone with the same dynamic characteristics as the high-altitude cleaning drone on the same trajectory, the offline reference trajectory of the high-altitude cleaning drone is obtained. Assuming that the virtual reference drone is free from external interference, its dynamic equations are:

[0030] ;

[0031] in, and These are the reference state vector and the reference input control, respectively.

[0032] Preferably, step S3 includes the following steps:

[0033] S31, Establish the position controller: Construct an MPC strategy based on linear error, and control the position of the quadrotor through the MPC strategy. The resulting position error model is divided into an altitude controller and a plane controller, which are controlled by input... Controlling the altitude of drones By input To control the translation of the UAV in the xy plane, the position error model consists of an altitude error model and a plane error model;

[0034] S32, Establish the attitude controller: Obtain the control torque input of the quadcopter UAV by solving the attitude controller. and through , Together, they work on the quadcopter drone to achieve trajectory tracking and control of the high-altitude cleaning drone.

[0035] Preferably, step S31 includes the following steps:

[0036] S311, the expression for the altitude error model of the altitude controller is:

[0037] ;

[0038] The control input is obtained by predicting the control law using the altitude error model, and the calculation formula is:

[0039] ;

[0040] in, All are diagonal matrices. The altitude state vector is the direct object of the optimization problem of predicting the dynamics model of a high-altitude cleaning drone; For height control vector, The height state error reference vector is the reference target for the objective function optimization; The reference vector for highly controlled errors is used to reduce optimization load and improve tracking performance; the calculation formulas are as follows:

[0041] ;

[0042] ;

[0043] ;

[0044] ;

[0045] in, To control input The reference value, k is a certain moment, For the prediction time domain, it represents the range of the position controller's predictions for the future system state; To control the time domain, this represents the position controller's method for minimizing the cost function at each execution. The optimization step size is obtained by solving the problem of minimizing the cost function of the height subsystem, and thus obtaining the control input sequence.

[0046] However, the high-altitude cleaning drone system only implements the first control input in the input sequence. Then, the high-altitude cleaning drone system enters the next time step and re-executes the optimization process based on the current state, thus forming a closed-loop control strategy.

[0047] Therefore, the control input for high-altitude cleaning drones is:

[0048] ;

[0049] S312, similar to the height error model calculation method, the expression for the plane error model of the plane controller is:

[0050] ;

[0051] Following the same calculation process as the altitude error model, a cost function is set. To minimize the tracking error in the xy direction, a quadratic problem is formed:

[0052] ;

[0053] in, All are diagonal matrices. For planar state vectors, For planar control vectors, This is the plane state error reference vector. This is the plane control error reference vector;

[0054] By minimizing the cost function Obtain control input ,in:

[0055] ;

[0056] in, Let be the first control input of the optimal control input sequence obtained at time k. , They are vectors Components in the x-axis and y-axis directions, For reference control input, This is the total control input vector actually applied to the controlled high-altitude cleaning UAV system at time k;

[0057] Right now:

[0058] ;

[0059] By setting Calculate the expected values ​​of the roll and pitch angles of the high-altitude cleaning drone. and .

[0060] Preferably, in step S32, similar to the position controller, the control law calculation formula predicted by the attitude controller model is:

[0061] ;

[0062] in, , All are diagonal matrices. , , , The calculation method is similar to that of the position control scheme, by minimizing Obtain control input , That is The control torque input of the high-altitude cleaning drone can be obtained in one step. .

[0063] Preferably, step S4 specifically includes the following steps:

[0064] S41, when collecting flight data, simulation experiments were conducted on the controller with different prediction time-domain parameters set in the speed range of 0~6m / s, using a speed interval of 0.5m / s, and the control time domain was selected. The input variables are made to be velocity time-domain parameters and prediction time-domain parameters, and reference experimental data for trajectory tracking of high-altitude cleaning UAVs in the simulation environment are obtained in the simulation.

[0065] S42, Select the prediction time-domain parameters that have relatively high stability and tracking accuracy in the tracking experiment, and use the grey relational analysis method to select the prediction time-domain parameters. Using the term as the parent sequence, the maximum position error and maximum attitude angle error are used as evaluation terms to analyze the correlation between the evaluation terms and the parent sequence. The maximum position error is... The mean of the maximum tracking error, the maximum attitude angle error is The mean of the maximum tracking error is obtained by comparing and processing the flight data of the high-altitude cleaning drone. The maximum position error is used to reflect the tracking accuracy of the drone, and the maximum attitude angle error is used to reflect the tracking stability of the high-altitude cleaning drone.

[0066] S43, using the minimum value of the maximum position error of the UAV as a reference, the optimal data set for the predicted time-domain parameters at different speeds is selected. A cubic polynomial method is then used to fit the optimal data set for the predicted time-domain parameters and the speed parameters, yielding an approximate relationship between the speed time-domain parameters and the predicted time-domain parameters:

[0067] .

[0068] Preferably, in step S5, the x-axis error during trajectory tracking is considered. y-axis error z-axis error and the speed of drones Set the event trigger condition when the drone's speed... When the value exceeds the threshold set based on the trajectory tracking accuracy requirements, the event triggering mechanism is executed;

[0069] When a drone travels at a high speed and covers a long distance within the same time period, the MPC is optimized at each sampling time to ensure the drone can track the target trajectory. Simultaneously, the x-axis error during drone trajectory tracking is also addressed. y-axis error and z-axis error When the value exceeds the threshold, the MPC is also optimized. That is, an optimal control problem in a finite time domain is solved in each control cycle. Only the first control variable of the optimal solution is used as the control input to ensure the trajectory tracking accuracy and stability of the UAV in long-term motion.

[0070] When the triggering condition is met, MPC will re-optimize the solution to obtain new control inputs. If the triggering condition is not met, the control input of the previous cycle will be used.

[0071] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0072] 1. Compared with the problem of constant computational load caused by the fixed prediction time domain of traditional MPC, the present invention uses a speed-adaptive prediction time domain dynamic adjustment mechanism to ensure tracking accuracy while intelligently reducing the problem scale according to the flight status, which greatly reduces the real-time computational complexity.

[0073] 2. Compared with the traditional periodic execution MPC controller, this invention introduces an event triggering mechanism based on dual threshold conditions, which only starts optimization and solving when the speed or tracking error exceeds the set range, avoiding a large amount of redundant calculations and allowing computing resources to be concentrated on key control nodes, significantly improving the real-time performance of the system.

[0074] 3. Traditional MPCs are rigid in parameter adjustment in complex operation scenarios, while this invention enables the controller to autonomously adapt to the dynamic processes of UAV acceleration and deceleration by predicting the time domain and flight speed online adaptive matching, and can maintain the optimal balance between control performance and computational load in both smooth trajectory segments and rapid change segments.

[0075] 4. Compared with the MPC architecture that performs continuous calculations throughout the entire cycle, this invention reduces the risk of cumulative errors and processor load fluctuations caused by frequent controller execution through an event-triggered mechanism. Combined with the collaborative control formed by adaptive prediction time domain, it provides a more stable and reliable control foundation for UAVs in long-term, high-precision special operations. Attached Figure Description

[0076] Figure 1 This is a schematic diagram of the unmanned aerial vehicle (UAV) operation trajectory tracking and control method of the present invention;

[0077] Figure 2 This is a simulation diagram comparing the control method of the present invention with the traditional control method for three-dimensional trajectory tracking.

[0078] Figure 3 This is a simulation diagram comparing the position tracking response curves of the UAV using the control method of the present invention with those of the traditional MPC control method;

[0079] Figure 4 This is a simulation diagram comparing the position tracking error curves of a UAV with those of the control method of the present invention and the traditional MPC control method. Detailed Implementation

[0080] The present invention will be further described below with reference to the accompanying drawings and embodiments, which illustrate the above and other technical features and advantages of the present invention. However, the following embodiments are merely preferred embodiments of the present invention and are not exhaustive.

[0081] Example 1:

[0082] like Figure 1-4 As shown, the present invention provides a method for tracking and controlling the operational trajectory of a high-altitude cleaning drone, the control method comprising the following steps:

[0083] S1. A dynamic model of the high-altitude cleaning drone is established based on Euler's equations and Newton's laws.

[0084] S2 transforms the dynamic model of the high-altitude cleaning UAV into a state-space equation and discretizes it.

[0085] S3. Establish the position controller and attitude controller of the high-altitude cleaning drone. The position controller is used to control the drone's altitude and xy plane, and the attitude controller is used to control the drone's roll, pitch and yaw three attitude angles.

[0086] S4. Collect flight data of different prediction time-domain parameters corresponding to different speeds of the high-altitude cleaning drone, determine the correlation between the prediction time-domain parameters and the evaluation items through the grey relational method, select the optimal prediction time-domain parameters, and fit them through a cubic polynomial.

[0087] S5, when tracking the operation trajectory of the high-altitude cleaning drone, dynamically adjusts the prediction time-domain parameters of the MPC controller based on the drone's speed information. During adjustment, an event triggering mechanism is introduced to intelligently control the rolling optimization process of the MPC controller. By setting dual triggering conditions of speed threshold and trajectory tracking error threshold, rolling optimization is only performed at key control nodes.

[0088] In this embodiment, the dynamic model of the high-altitude cleaning drone is represented as follows:

[0089] ;

[0090] in, Let x, y, and z be the mass and gravitational acceleration of the drone, respectively, and let x, y, and z be the position coordinates of the drone. These are the roll angle, pitch angle, and yaw angle of the drone, respectively. The quadcopter drones along axis, axis, Moment of inertia of the shaft These are the three components of the angular velocity along the body axis. The length between the center of the drone and the rotor. The total thrust generated by the propeller. The torque generated on the propeller These represent the accelerations of the drone in three directions, respectively. This represents the angular acceleration of the drone in three directions.

[0091] In this embodiment, The specific expression is as follows:

[0092] ;

[0093] in, Let be the angular velocity of the drone rotor. The lift coefficient, This is the torque coefficient.

[0094] In this embodiment, step S2 includes the following steps:

[0095] S21, the state-space equation expression of the high-altitude cleaning UAV position system is:

[0096] ;

[0097] in, and The specific expression is: ;

[0098] The discrete model expression for the UAV's position is:

[0099] ;

[0100] in, This is the position error vector. To control the input error vector;

[0101] The state-space equation expression for the attitude system of the S22 high-altitude cleaning UAV is as follows:

[0102] ;

[0103] The discrete model expression for the attitude component of the UAV is:

[0104] ;

[0105] in, Let be the state vector of the attitude system. To control the input vector, The state matrix, The input matrix;

[0106] S23, by constructing a virtual reference drone with the same dynamic characteristics as the high-altitude cleaning drone on the same trajectory, the offline reference trajectory of the high-altitude cleaning drone is obtained. Assuming that the virtual reference drone is free from external interference, its dynamic equations are:

[0107] ;

[0108] in, and These are the reference state vector and the reference input control, respectively.

[0109] In this embodiment, step S3 includes the following steps:

[0110] S31, Establish the position controller: Construct an MPC strategy based on linear error, and control the position of the quadrotor through the MPC strategy. The resulting position error model is divided into an altitude controller and a plane controller, which are controlled by input... Controlling the altitude of drones By input To control the translation of the UAV in the xy plane, the position error model consists of an altitude error model and a plane error model;

[0111] S32, Establish the attitude controller: Obtain the control torque input of the quadcopter UAV by solving the attitude controller. and through , Together, they work on the quadcopter drone to achieve trajectory tracking and control of the high-altitude cleaning drone.

[0112] In this embodiment, step S31 includes the following steps:

[0113] S311, the expression for the altitude error model of the altitude controller is:

[0114] ;

[0115] The control input is obtained by predicting the control law using the altitude error model, and the calculation formula is:

[0116] ;

[0117] in, All are diagonal matrices. The altitude state vector is the direct object of the optimization problem of predicting the dynamics model of a high-altitude cleaning drone; For height control vector, The height state error reference vector is the reference target for the objective function optimization; The reference vector for highly controlled errors is used to reduce optimization load and improve tracking performance; the calculation formulas are as follows:

[0118] ;

[0119] ;

[0120] ;

[0121] ;

[0122] in, To control input The reference value, k is a certain moment, For the prediction time domain, it represents the range of the position controller's predictions for the future system state; To control the time domain, this represents the position controller's method for minimizing the cost function at each execution. The optimization step size is obtained by solving the problem of minimizing the cost function of the height subsystem, and thus obtaining the control input sequence.

[0123] However, the high-altitude cleaning drone system only implements the first control input in the input sequence. Then, the high-altitude cleaning drone system enters the next time step and re-executes the optimization process based on the current state, thus forming a closed-loop control strategy.

[0124] Therefore, the control input for high-altitude cleaning drones is:

[0125] ;

[0126] S312, similar to the height error model calculation method, the expression for the plane error model of the plane controller is:

[0127] ;

[0128] Following the same calculation process as the altitude error model, a cost function is set. To minimize the tracking error in the xy direction, a quadratic problem is formed:

[0129] ;

[0130] in, All are diagonal matrices. For planar state vectors, For planar control vectors, This is the plane state error reference vector. This is the plane control error reference vector;

[0131] By minimizing the cost function Obtain control input ,in:

[0132] ;

[0133] in, Let be the first control input of the optimal control input sequence obtained at time k. , They are vectors Components in the x-axis and y-axis directions, For reference control input, This is the total control input vector actually applied to the controlled high-altitude cleaning UAV system at time k;

[0134] Right now:

[0135] ;

[0136] By setting Calculate the expected values ​​of the roll and pitch angles of the high-altitude cleaning drone. and .

[0137] In this embodiment, in step S32, similar to the position controller, the control law calculation formula predicted by the attitude controller model is:

[0138] ;

[0139] in, , All are diagonal matrices. , , , The calculation method is similar to that of the position control scheme, by minimizing Obtain control input , That is The control torque input of the high-altitude cleaning drone can be obtained in one step. .

[0140] In this embodiment, step S4 specifically includes the following steps:

[0141] S41, when collecting flight data, simulation experiments were conducted on the controller with different prediction time-domain parameters set in the speed range of 0~6m / s, using a speed interval of 0.5m / s, and the control time domain was selected. The input variables are made to be velocity time-domain parameters and prediction time-domain parameters, and reference experimental data for trajectory tracking of high-altitude cleaning UAVs in the simulation environment are obtained in the simulation.

[0142] S42, Select the prediction time-domain parameters that have relatively high stability and tracking accuracy in the tracking experiment, and use the grey relational analysis method to select the prediction time-domain parameters. Using the term as the parent sequence, the maximum position error and maximum attitude angle error are used as evaluation terms to analyze the correlation between the evaluation terms and the parent sequence. The maximum position error is... The mean of the maximum tracking error, the maximum attitude angle error is The mean of the maximum tracking error is obtained by comparing and processing the flight data of the high-altitude cleaning drone. The maximum position error is used to reflect the tracking accuracy of the drone, and the maximum attitude angle error is used to reflect the tracking stability of the high-altitude cleaning drone.

[0143] S43, using the minimum value of the maximum position error of the UAV as a reference, the optimal data set for the predicted time-domain parameters at different speeds is selected. A cubic polynomial method is then used to fit the optimal data set for the predicted time-domain parameters and the speed parameters, yielding an approximate relationship between the speed time-domain parameters and the predicted time-domain parameters:

[0144] .

[0145] In this embodiment, in step S5, the x-axis error during trajectory tracking is used as a basis. y-axis error z-axis error and the speed of drones Set the event trigger condition when the drone's speed... When the value exceeds the threshold set based on the trajectory tracking accuracy requirements, the event triggering mechanism is executed;

[0146] When a drone travels at a high speed and covers a long distance within the same time period, the MPC is optimized at each sampling time to ensure the drone can track the target trajectory. Simultaneously, the x-axis error during drone trajectory tracking is also addressed. y-axis error and z-axis error When the value exceeds the threshold, the MPC is also optimized. That is, an optimal control problem in a finite time domain is solved in each control cycle. Only the first control variable of the optimal solution is used as the control input to ensure the trajectory tracking accuracy and stability of the UAV in long-term motion.

[0147] When the triggering condition is met, MPC will re-optimize the solution to obtain new control inputs. If the triggering condition is not met, the control input of the previous cycle will be used.

[0148] Example 2:

[0149] To verify the effectiveness of the high-altitude cleaning drone operation trajectory tracking and control method provided in this embodiment, a simulation verification experiment was conducted on the high-altitude cleaning drone operation trajectory tracking and control method provided in this embodiment in the MATLAB 2021 environment.

[0150] A drone trajectory tracking simulation experiment was conducted in MATLAB. The desired trajectory in the simulation was a circular trajectory. The drone's speed was 4.8 m / s from 0-2s, 1.2 m / s from 2-15s, 2.3 m / s from 15-25s, and 0.4 m / s from 25-40s. The speed trigger threshold for the event triggering mechanism was set to 2 m / s, and the drone trajectory tracking error threshold was set to 0.1 m.

[0151] Depend on Figure 2 It can be seen that the adaptive predictive time-domain MPC controller can track the target trajectory faster than the traditional MPC controller;

[0152] like Figure 3 As shown in the figure, (a), (b) and (c) are the tracking response curves of the traditional MPC controller and the adaptive predictive time domain MPC controller on the x, y and z axes respectively, when tracking the trajectory and the reference trajectory during operation. It can be seen that the adaptive predictive time domain MPC controller has higher trajectory tracking accuracy than the traditional MPC controller.

[0153] like Figure 4 As shown in the figure, (a), (b) and (c) are comparisons of the tracking error curves of the traditional MPC controller and the adaptive predictive time-domain MPC controller on the x, y and z axes, respectively, for the operation trajectory tracking and reference trajectory tracking.

[0154] The above description is merely a preferred embodiment of the present invention and is illustrative rather than restrictive. Those skilled in the art will understand that many changes, modifications, and even equivalents can be made within the spirit and scope defined by the claims of the present invention, all of which will fall within the protection scope of the present invention.

Claims

1. A method for tracking and controlling the operational trajectory of a high-altitude cleaning drone, characterized in that, The control method includes the following steps: S1. A dynamic model of the high-altitude cleaning drone is established based on Euler's equations and Newton's laws. S2 transforms the dynamic model of the high-altitude cleaning UAV into a state-space equation and discretizes it. S3, establish the position controller and attitude controller for the high-altitude cleaning drone; S4. Collect flight data of different prediction time-domain parameters corresponding to different speeds of the high-altitude cleaning drone, determine the correlation between the prediction time-domain parameters and the evaluation items through the grey relational method, select the optimal prediction time-domain parameters, and fit them through a cubic polynomial. S5, when tracking the operation trajectory of the high-altitude cleaning drone, dynamically adjusts the prediction time-domain parameters of the MPC controller based on the drone's speed information. During adjustment, an event triggering mechanism is introduced to intelligently control the rolling optimization process of the MPC controller. By setting dual triggering conditions of speed threshold and trajectory tracking error threshold, rolling optimization is only performed at key control nodes.

2. The high-altitude cleaning drone operation trajectory tracking and control method as described in claim 1, characterized in that, The dynamic model of the high-altitude cleaning UAV is expressed as follows: ； in, Let x, y, and z be the mass and gravitational acceleration of the drone, respectively, and let x, y, and z be the position coordinates of the drone. These are the roll angle, pitch angle, and yaw angle of the drone, respectively. The quadcopter drones along axis, axis, Moment of inertia of the shaft These are the three components of the angular velocity along the body axis. The length between the center of the drone and the rotor. The total thrust generated by the propeller. The torque generated on the propeller These represent the accelerations of the drone in three directions, respectively. This represents the angular acceleration of the drone in three directions.

3. The method for tracking and controlling the operational trajectory of a high-altitude cleaning drone as described in claim 1, characterized in that, The The specific expression is as follows: ； in, Let be the angular velocity of the drone rotor. The lift coefficient, This is the torque coefficient.

4. The method for tracking and controlling the operational trajectory of a high-altitude cleaning drone as described in claim 1, characterized in that, Step S3 includes the following steps: S31, Establish the position controller: Construct an MPC strategy based on linear error, control the position of the quadcopter through the MPC strategy, and obtain a position controller consisting of a height controller and a plane controller. This is achieved by inputting... Controlling the altitude of drones By input To control the translation of the UAV in the xy plane, the position error model consists of an altitude error model and a plane error model; S32, Establish the attitude controller: Obtain the control torque input of the quadcopter UAV by solving the attitude controller. and through , Together, they work on the quadcopter drone to achieve trajectory tracking and control of the high-altitude cleaning drone.

5. The method for tracking and controlling the operational trajectory of a high-altitude cleaning drone as described in claim 1, characterized in that, Step S4 specifically includes the following steps: S41, when collecting flight data, simulation experiments were conducted on the controller with different prediction time-domain parameters set in the speed range of 0~6m / s, using a speed interval of 0.5m / s, and the control time domain was selected. The input variables are made to be velocity time-domain parameters and prediction time-domain parameters, and reference experimental data for trajectory tracking of high-altitude cleaning UAVs in the simulation environment are obtained in the simulation. S42, Select the prediction time-domain parameters that have relatively high stability and tracking accuracy in the tracking experiment, and use the grey relational analysis method to select the prediction time-domain parameters. Using the term as the parent sequence, the maximum position error and maximum attitude angle error are used as evaluation terms to analyze the correlation between the evaluation terms and the parent sequence. The maximum position error is... The mean of the maximum tracking error, the maximum attitude angle error is The mean of the maximum tracking error is obtained by comparing and processing the flight data of the high-altitude cleaning drone. The maximum position error is used to reflect the tracking accuracy of the drone, and the maximum attitude angle error is used to reflect the tracking stability of the high-altitude cleaning drone. S43, using the minimum value of the maximum position error of the UAV as a reference, the optimal data set for the predicted time-domain parameters at different speeds is selected. A cubic polynomial method is then used to fit the optimal data set for the predicted time-domain parameters and the speed parameters, yielding an approximate relationship between the speed time-domain parameters and the predicted time-domain parameters: 。 6. The method for tracking and controlling the operational trajectory of a high-altitude cleaning drone as described in claim 1, characterized in that, In step S5, the x-axis error during trajectory tracking is used as a basis. y-axis error z-axis error and the speed of drones Set the event trigger condition when the drone's speed... When the value exceeds the threshold set based on the trajectory tracking accuracy requirements, the event triggering mechanism is executed; When a drone travels at a high speed and covers a long distance within the same time period, the MPC is optimized at each sampling time to ensure the drone can track the target trajectory. Simultaneously, the x-axis error during drone trajectory tracking is also addressed. y-axis error and z-axis error When the value exceeds the threshold, the MPC is also optimized. That is, an optimal control problem in a finite time domain is solved in each control cycle. Only the first control variable of the optimal solution is used as the control input to ensure the trajectory tracking accuracy and stability of the UAV in long-term motion. When the triggering condition is met, MPC will re-optimize the solution to obtain new control inputs. If the triggering condition is not met, the control input of the previous cycle will be used.

Images