Unmanned aerial vehicle system for radio frequency target simulation system and control method

Abstract

The invention discloses an unmanned aerial vehicle system for a radio frequency target simulation system and a control method, and aims to solve the problem that an existing radio frequency target simulator is difficult to simulate the motion characteristics of a target object by performing high-precision regular motion control on an unmanned aerial vehicle. The system is composed of an unmanned aerial vehicle, a radio frequency target simulator, an airborne task machine, a wireless link, an RTK assembly and data processing software, the flight interval of the unmanned aerial vehicle with a radar as the center is far smaller than the flight interval of an actual target object, and by means of a scaling test method, the unmanned aerial vehicle in short-distance low-dynamic flight can achieve high-speed flight. And high-dynamic motion characteristics of a long-distance actual target object can be simulated. A motion rule of the unmanned aerial vehicle is described by a time sequence of discrete track coordinate points, and a track line is discretized into a time sequence S2 with equal time intervals by data processing software. The sequence S2 is transmitted to an airborne task machine through a wireless link, and the airborne task machine converts the sequence S2 into a flight command which can be recognized by the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention relates to an unmanned aerial vehicle control system and a control method, and in particular to an unmanned aerial vehicle system and a control method for a radio frequency target simulation system. Background Art

[0002] As an important equipment in modern military and civilian fields, the stability and accuracy of radar performance are crucial. In the process of radar research and development, field tests are the traditional method to test the performance of radar systems. However, if real fighters, missiles, etc. are used as RF targets for radar tests, the costs of sites, equipment, personnel, etc. are extremely high. In contrast, RF target simulators can generate a variety of electromagnetic environments and various real and fake targets to comprehensively test the performance of radar systems. This method not only reduces testing costs, but also improves testing efficiency.

[0003] However, as a static device, although the RF target simulator can simulate the radar RF echo characteristics of targets such as fighters and missiles, it is difficult to simulate the motion characteristics of the target. By mounting the RF target simulator on a drone and then accurately controlling the drone according to certain rules, the drone carrying the RF target simulator can simulate the distance, speed, angle and other motion characteristics of the target, making the simulation process more comprehensive and realistic. However, there is little research on the method of combining the control of drones with RF target simulators on the market. Summary of the invention

[0004] The purpose of the present invention is to provide a UAV system and control method for a radio frequency target simulation system, aiming to solve the problem that existing radio frequency target simulators are difficult to simulate the motion characteristics of target objects by performing high-precision regular motion control on the UAV.

[0005] The present invention is achieved through the following technical solutions:

[0006] The system consists of a drone, a radio frequency target simulator, an airborne mission machine, a wireless link, an RTK component, and data processing software. The drone includes a flight controller. The RTK component is used to provide precise positioning for the drone, and the wireless link provides a data exchange channel between the drone and the ground-side data processing software.

[0007] The flight range of the UAV centered on the radar is much smaller than the flight range of the actual target. Through the scaled-down test method, the UAV flying at short distance and low dynamics can simulate the high dynamic motion characteristics of the actual target at a long distance.

[0008] The motion rules of the drone are described by the time series of discrete trajectory coordinate points, that is, the motion trajectory of the actual target object is recorded as L1, and the scaled drone motion trajectory is recorded as L2. The data processing software discretizes the trajectory line into a time series S2 with equal time intervals.

[0009] The sequence S2 is transmitted to the onboard mission machine via a wireless link, and the onboard mission machine converts it into a flight command recognizable by the UAV.

[0010] The control steps after the flight controller executes the received flight command include:

[0011] The first step is to establish a mathematical model of UAV dynamics and kinematics.

[0012] In the second step, the state vector of the UAV is predicted using the cached historical control quantities.

[0013] The third step is to calculate the optimal planning speed through trajectory commands and minimizing the cost function.

[0014] The fourth step is to derive and calculate the attitude and throttle output so that it can act on the drone to smoothly execute the expected trajectory action.

[0015] Furthermore, the first step comprises:

[0016] Establish the UAV attitude dynamics transfer function model:

[0017]

[0018] In the formula, K ai is the coefficient related to the aircraft body, s is the Laplace variable, U(s) is the Laplace transform of the motor throttle input, A tt (s) is the Laplace transform of the aircraft attitude. Due to the symmetry of the aircraft body, A tt (s) describes the dynamic characteristics of the roll angle and pitch angle. The dynamic characteristics of the yaw angle are the same in form, but only the values ​​are different.

[0019] Establish the UAV velocity kinematic transfer function model:

[0020]

[0021] In the formula, K vi is the coefficient related to the aircraft body, A cc (s) is the Laplace transform of the acceleration input, V(s) is the Laplace transform of the aircraft flight speed control, V(s) describes the motion characteristics of the east and north velocities, and the motion characteristics of the celestial velocity are the same in form, with only different numerical coefficients.

[0022] Establish the UAV position kinematic transfer function model:

[0023]

[0024] In the formula, K pi is the coefficient related to the aircraft body, V(s) is the Laplace transform of the velocity input, P(s) is the Laplace transform of the aircraft position control, P(s) describes the position motion characteristics of the UAV in the longitude and latitude directions, and the motion characteristics of the altitude are the same in form, with only different numerical coefficients.

[0025] Furthermore, the second step includes:

[0026] The motor throttle vector calculated in the last iteration is cached and recorded as:

[0027]

[0028] In the above formula, t represents the current iteration period, and n is the length of the prediction interval.

[0029] When the flight control receives the next trajectory command, it uses the historical throttle input to act on the above-mentioned dynamics and kinematics models to make a forward prediction of the speed and position of the drone, and the predicted speed is recorded as:

[0030]

[0031] The predicted position is recorded as:

[0032]

[0033] The predicted speed and position are compared with the speed and position obtained by RTK and inertial sensors to obtain the model correction value. Used to improve the prediction accuracy of the next iteration.

[0034] Furthermore, the third step includes:

[0035] Introduce trajectory error tracking cost function:

[0036]

[0037] In the above formula, Q is the output error parameter matrix. Generally, this matrix is ​​a diagonal matrix and a positive definite matrix. is the predicted UAV position error vector, and the superscript T represents the transpose of the vector.

[0038] The first-order necessary condition By solving the minimization cost function J = min{E(t)}, the optimal planning speed V is obtained. ref :

[0039]

[0040] Where Kr is the input correlation coefficient matrix, Kx is the state error correlation coefficient matrix, and Ky is the control output correlation coefficient matrix.

[0041] Furthermore, the fourth step includes:

[0042] Given the planned flight speed V ref In the case of tt_ref , and expressed using the direction cosine matrix:

[0043]

[0044] Among them, a ij are the elements of the direction cosine matrix, which is an orthogonal matrix.

[0045] Since the commands received by the drone are discrete, in order to make the flight path smooth and improve the control accuracy, it is necessary to perform spherical interpolation on the planned flight attitude. The corrected transition direction cosine matrix A tt_fix for:

[0046]

[0047] In the formula, A tt_ref (0) is the flight attitude planned for this iteration, A tt_ref (1) is the flight attitude planned in the last iteration, k is the transition coefficient in the value range [0, 1], and the small increment of k can be set according to the desired smoothness.

[0048] Planning Posture A tt_ref The corresponding yaw angle is used to align the antenna of the RF target simulator with the ground radar; the planning attitude A tt_ref The corresponding roll and pitch angles are used for the acceleration control of the drone. The motor throttle output is given by the attitude closed-loop controller:

[0049] U(s)=[E att (s)·C att (s)-W(s)]·C w (s)

[0050] In the above formula, E att is the attitude loop error, C att is the attitude controller, W is the feedback angular velocity, C w It is the angular velocity loop controller.

[0051] It can be seen from the above formula that the output motor throttle U(s) can make the aircraft perform the expected flight attitude and accurately simulate the motion characteristics of the target object.

[0052] The present invention collects and processes the motion trajectory of the target object, and uses a scaled-down experimental method to enable a drone carrying a radio frequency target simulator to achieve a comprehensive simulation of the radio frequency characteristics and motion characteristics of the target object under high-precision motion control, and has the advantage of low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 Schematic diagram of the application scenario of the present invention

[0054] Figure 2 Schematic diagram of the functional modules of the present invention

[0055] Figure 3 The control step flow chart of the present invention is DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention more clear, the technical solution of the application is further elaborated in detail below in conjunction with the accompanying drawings.

[0057] The system consists of a drone 3, a radio frequency target simulator 2, an airborne mission machine, a wireless link, an RTK component, and data processing software. The drone 3 includes a flight control system. The RTK component is used to provide precise positioning for the drone, and the wireless link provides a data exchange channel between the drone and the data processing software on the ground.

[0058] The flight range of the UAV centered on the radar 1 is much smaller than the flight range of the actual target 4. Through the scaled-down test method, the UAV flying at short distance and low dynamics can simulate the high dynamic motion characteristics of the actual target at a long distance.

[0059] The motion rules of the drone are described by the time series of discrete trajectory coordinate points, that is, the motion trajectory of the actual target object is recorded as L1, and the scaled drone motion trajectory is recorded as L2. The data processing software discretizes the trajectory line into a time series S2 with equal time intervals.

[0060] The sequence S2 is transmitted to the onboard mission machine via a wireless link, and the onboard mission machine converts it into a flight command recognizable by the UAV.

[0061] The control steps after the flight controller executes the received flight command include:

[0062] The first step is to establish a dynamic and kinematic mathematical model based on the B02 model UAV of Qingdao Hangpeng UAV Technology Co., Ltd.

[0063] In the second step, the state vector of the UAV is predicted using the cached historical control quantities.

[0064] The third step is to calculate the optimal planning speed through trajectory commands and minimizing the cost function.

[0065] The fourth step is to derive and calculate the attitude and throttle output so that it can act on the drone to smoothly execute the expected trajectory action.

[0066] Furthermore, the first step comprises:

[0067] Establish the UAV attitude dynamics transfer function model:

[0068]

[0069] Where s is the Laplace variable, U(s) is the Laplace transform of the motor throttle input, and A tt (s) is the Laplace transform of the aircraft attitude. Due to the symmetry of the aircraft body, A tt (s) describes the dynamic characteristics of the roll angle and pitch angle. The dynamic characteristics of the yaw angle are the same in form, but only the values ​​are different.

[0070] Establish the UAV velocity kinematic transfer function model:

[0071]

[0072] In the formula, A cc (s) is the Laplace transform of the acceleration input, V(s) is the Laplace transform of the aircraft flight speed control, V(s) describes the motion characteristics of the east and north velocities, and the motion characteristics of the celestial velocity are the same in form, with only different numerical coefficients.

[0073] Establish the UAV position kinematic transfer function model:

[0074]

[0075] Where V(s) is the Laplace transform of the velocity input, P(s) is the Laplace transform of the aircraft position control, and P(s) describes the position motion characteristics of the UAV in the longitude and latitude directions. The motion characteristics of the altitude are the same in form, with only different numerical coefficients.

[0076] Furthermore, the second step includes:

[0077] The motor throttle vector calculated in the last iteration is cached and recorded as:

[0078]

[0079] In the above formula, t represents the current iteration period, and n is the length of the prediction interval.

[0080] When the flight control receives the next trajectory command, it uses the historical throttle input to act on the above-mentioned dynamics and kinematics models to make a forward prediction of the speed and position of the drone, and the predicted speed is recorded as:

[0081]

[0082] The predicted position is recorded as:

[0083]

[0084] The predicted speed and position are compared with the speed and position obtained by RTK and inertial sensors to obtain the model correction value. Used to improve the prediction accuracy of the next iteration.

[0085] Furthermore, the third step includes:

[0086] Introduce trajectory error tracking cost function:

[0087]

[0088] In the above formula, Q is the output error parameter matrix. Generally, this matrix is ​​a diagonal matrix and a positive definite matrix. is the predicted UAV position error vector, and the superscript T represents the transpose of the vector.

[0089] The first-order necessary condition By solving the minimization cost function J = min{E(t)}, the optimal planning speed V is obtained. ref :

[0090]

[0091] Where Kr is the input correlation coefficient matrix, Kx is the state error correlation coefficient matrix, and Ky is the control output correlation coefficient matrix.

[0092] Furthermore, the fourth step includes:

[0093] Given the planned flight speed V ref In the case of tt_ref , and expressed using the direction cosine matrix:

[0094]

[0095] Among them, a ij are the elements of the direction cosine matrix, which is an orthogonal matrix.

[0096] Since the commands received by the drone are discrete, in order to make the flight path smooth and improve the control accuracy, it is necessary to perform spherical interpolation on the planned flight attitude. The corrected transition direction cosine matrix A tt_fix for:

[0097]

[0098] In the formula, A tt_ref (0) is the flight attitude planned for this iteration, A tt_ref (1) is the flight attitude planned in the last iteration, k is the transition coefficient in the value range [0, 1], and a small increment of k can be set according to the desired smoothness.

[0099] Planning Posture A tt_ref The corresponding yaw angle in is used to align the antenna of the RF target simulator with the ground radar; the planning attitude A tt_ref The corresponding roll and pitch angles are used for the acceleration control of the drone. The motor throttle output is given by the attitude closed-loop controller:

[0100] U(s)=[E att (s)·C att (s)-W(s)]·C w (s)

[0101] In the above formula, E att is the attitude loop error, C att is the attitude controller, W is the feedback angular velocity, C w It is the angular velocity loop controller.

[0102] It can be seen from the above formula that the output motor throttle U(s) can make the aircraft perform the expected flight attitude and accurately simulate the motion characteristics of the target object.

Claims

1. A UAV system for a radio frequency target simulation system, characterized in that The system consists of a drone, a radio frequency target simulator, an airborne mission machine, a wireless link, an RTK component, and data processing software. The drone includes a flight control system. The RTK component is used to provide precise positioning for the drone. The wireless link provides a data interaction channel between the drone and the data processing software on the ground. The flight range of the UAV centered on the radar is much smaller than the flight range of the actual target. Through the scaled-down test method, the UAV flying at short distance and low dynamics can simulate the high dynamic motion characteristics of the actual target at a long distance. The motion rules of the drone are described by the time series of discrete trajectory coordinate points, that is, the actual target motion trajectory is recorded as L1, and the scaled drone motion trajectory is recorded as L2. The data processing software discretizes the trajectory line into a time series S2 with equal time intervals. The sequence S2 is transmitted to the onboard mission machine via a wireless link, and the onboard mission machine converts it into a flight command recognizable by the UAV.

2. A UAV control method for a radio frequency target simulation system, characterized in that The control steps after the flight controller executes the received flight command include: The first step is to establish the mathematical model of UAV dynamics and kinematics; The second step is to predict the state vector of the UAV through the cached historical control quantity; The third step is to calculate the optimal planning speed through trajectory command and minimization cost function; The fourth step is to derive and calculate the attitude and throttle output so that it can act on the drone to smoothly execute the expected trajectory action.

3. The method for controlling a UAV for a radio frequency target simulation system according to claim 2, characterized in that: The first step of the control step comprises: Establish the UAV attitude dynamics transfer function model: In the formula, K ai is the coefficient related to the aircraft body, s is the Laplace variable, U(s) is the Laplace transform of the motor throttle input, A tt (s) is the Laplace transform of the aircraft attitude. Due to the symmetry of the aircraft body, A tt (s) describes the dynamic characteristics of the roll angle and pitch angle. The dynamic characteristics of the yaw angle are the same in form, but only the values ​​are different; Establish the UAV velocity kinematic transfer function model: In the formula, K vi is the coefficient related to the aircraft body, A cc (s) is the Laplace transform of the acceleration input, V(s) is the Laplace transform of the aircraft flight speed control, V(s) describes the motion characteristics of the east and north speeds, and the motion characteristics of the celestial speed are the same, only the numerical coefficients are different; Establish the UAV position kinematic transfer function model: In the formula, K pi is the coefficient related to the aircraft body, V(s) is the Laplace transform of the velocity input, P(s) is the Laplace transform of the aircraft position control, P(s) describes the position motion characteristics of the UAV in the longitude and latitude directions, and the motion characteristics of the altitude are the same in form, with only different numerical coefficients.

4. The method for controlling a UAV for a radio frequency target simulation system according to claim 2, characterized in that: The second step of the control step comprises: The motor throttle vector calculated in the last iteration is cached and recorded as: In the above formula, t represents the current iteration period, and n is the length of the prediction interval; When the flight control receives the next trajectory command, it uses the historical throttle input to act on the above-mentioned dynamics and kinematics models to make a forward prediction of the speed and position of the drone, and the predicted speed is recorded as: The predicted position is recorded as: The predicted speed and position are compared with the speed and position obtained by RTK and inertial sensors to obtain the model correction value. Used to improve the prediction accuracy of the next iteration.

5. The method for controlling a UAV for a radio frequency target simulation system according to claim 2, characterized in that: The third step of the control step comprises: Introduce trajectory error tracking cost function: In the above formula, Q is the output error parameter matrix. Generally, the matrix is ​​a diagonal matrix and a positive definite matrix. is the predicted UAV position error vector, and the superscript T represents the transpose of the vector; The first-order necessary condition By solving the minimization cost function J = min{E(t)}, the optimal planning speed V is obtained. ref : Where Kr is the input correlation coefficient matrix, Kx is the state error correlation coefficient matrix, and Ky is the control output correlation coefficient matrix.

6. The method for controlling a UAV for a radio frequency target simulation system according to claim 2, characterized in that: The fourth step of the control step comprises: Given the planned flight speed V ref In the case of tt_ref , and expressed using the direction cosine matrix: Among them, a ij are the elements of the direction cosine matrix, which is an orthogonal matrix; Since the commands received by the drone are discrete, in order to make the flight path smooth and improve the control accuracy, it is necessary to perform spherical interpolation on the planned flight attitude. The corrected transition direction cosine matrix A tt_fix for: In the formula, A tt_ref (0) is the flight attitude planned for this iteration, A tt_ref (1) is the flight attitude planned in the last iteration, k is the transition coefficient in the range [0, 1], and the small increment of k can be set according to the desired smoothness; Planning Posture A tt_ref The corresponding yaw angle in is used to align the antenna of the RF target simulator with the ground radar; the planning attitude A tt_ref The corresponding roll and pitch angles are used for the acceleration control of the drone; the motor throttle output is given by the attitude closed-loop controller: U(s)=[E att (s)·C att (s)-W(s)]·C w (s) In the above formula, E att is the attitude loop error, C att is the attitude controller, W is the feedback angular velocity, C w It is the angular velocity loop controller.

Images