Method for high-precision tracking of target by unmanned aerial vehicle based on position and attitude switching

Abstract

The invention provides a high-precision target tracking method of an unmanned aerial vehicle based on position and attitude switching. The basic principle of the method is that a three-section mode of switching high-precision position proportional filtering differential guidance to azimuth guidance and finally switching high-precision proportional filtering differential high-precision guidance is adopted, and the method has the advantages that the high-precision proportional filtering differential guidance adopted at the tail end has good precision; meanwhile, the angle guiding mode of switching back to the azimuth angle in the second-stage guiding can avoid the problem that the practical range and space are insufficient during single-position tracking guiding, proportional differential position guiding is divided into large signals and small signals, table look-up type attitude instruction generation is adopted, and the positioning accuracy is improved. The problem of instability caused by mismatching of guide parameters due to too large attitude instructions can be well solved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle control and guidance, and in particular to a method for high-precision tracking and guidance of a low-speed land target by an unmanned aerial vehicle. Background Art

[0002] With the rapid development of UAV technology, higher precision requirements are put forward for its tracking and guidance of moving targets in practical applications; especially when attacking small targets, such as targets less than 0.3 meters in size, many off-target situations will occur when traditional guidance methods are directly transferred.

[0003] Traditional guidance methods, including proportional guidance, tracking guidance, front-end guidance, and parallel approach guidance, are primarily used for high-speed missiles targeting high-speed targets on the sea or in the air. However, when used by lower-speed drones to strike small, slow-moving targets on the ground, misses of more than 5 meters can sometimes prevent the small targets from being destroyed.

[0004] It should be noted that the signals invented in the above background technology section are only used to enhance the understanding of the background of the present invention, and therefore may include signals that do not constitute prior art known to ordinary technicians in this field. Summary of the Invention

[0005] When using proportional guidance, azimuth guidance and other methods alone to move to low-speed UAVs to strike small targets, it cannot fundamentally solve the above-mentioned UAVs’ off-target problems when striking small targets. When the initial distance to the target is within a small range, the guidance method using position proportional differential has good accuracy, but its problem is that its applicability is only in a small range, so it is rarely used in the guidance of traditional high-speed moving aircraft. Therefore, in order to expand its applicability, the guidance is divided into three stages. The first stage adopts position guidance; the second stage introduces azimuth guidance; and the third stage switches to position tracking guidance. This is a good design idea. Based on the above background reasons, the present invention provides a three-stage guidance method based on position attitude switching, which can well approach the problem of high-precision tracking and guidance of UAVs for low-speed ground targets.

[0006] The purpose of the present invention is to provide a high-precision tracking method for a UAV based on position and attitude switching, thereby solving the problem of insufficient guidance accuracy of current UAVs for small moving targets on the ground to a certain extent.

[0007] An embodiment of the present invention provides a method for high-precision tracking of a target by a UAV based on position and attitude switching, the method comprising: Step S10: Install a radar device or a visual imaging device on the drone to calculate the target's altitude signal, longitudinal position signal, and lateral position signal; at the same time, calculate the drone's own altitude signal, lateral position signal, and longitudinal position signal based on the drone's own inertial navigation device, and at the same time calculate the distance between the drone and the target.

[0008] Set the start time of the second stage guidance, the start time of the third stage guidance, the altitude error threshold, and the lateral position error threshold; the start time of the second stage guidance is when the distance between the UAV and the target reaches the first threshold of the guidance switching distance, and the start time of the third stage guidance is when the distance between the UAV and the target reaches the second threshold of the guidance switching distance. The altitude error threshold, lateral position error threshold, first threshold of the guidance switching distance, and second threshold of the guidance switching distance are constant parameters and should be debugged and set according to the actual situation of the guidance.

[0009] Step S20: After the guidance starts, when the time is less than the second stage guidance start time, the first stage of the UAV guiding the target begins.

[0010] The first stage of guidance uses position tracking guidance. Its basic principle is to set the UAV's altitude command signal as the target's altitude signal, and set the UAV's lateral command signal as the target's lateral position signal.

[0011] The height error signal is calculated based on the target's height signal and the drone's own height signal; then, based on the height error, an inertia lag link is designed to obtain the height error inertia lag signal.

[0012] When the height error inertial hysteresis signal is greater than the height error threshold, the variable coefficient nonlinear height transformation method is used to solve the height error equivalent pitch angle expected signal.

[0013] When the height error inertial hysteresis signal is less than the height error threshold, an inertial filter differential proportional controller is designed to solve the height error equivalent pitch angle expected signal. Based on the altitude error equivalent pitch angle desired signal, the drone is tracked using a pitch angle stabilization tracking system. Since the design of these systems varies slightly for different drone models, and well-established design methods are readily available, and there's no need to specify a specific pitch angle stabilization tracking system, the methods we provide are generally applicable, so we won't elaborate on them here. Once the drone has tracked the altitude error equivalent pitch angle desired signal, it has achieved tracking and guidance of the target at the first stage of altitude.

[0014] The lateral position error signal is calculated based on the lateral position signal of the target and the lateral position signal of the UAV itself. Then, based on the lateral position error, an inertial lag link is designed to obtain the lateral position error inertial lag signal.

[0015] When the inertial hysteresis signal of the lateral position error is greater than the lateral position error threshold, the variable coefficient nonlinear lateral position transformation method is used to solve the expected signal of the equivalent yaw angle of the lateral position error.

[0016] When the inertial hysteresis signal of the lateral position error is less than the lateral position error threshold, an inertial filter differential PD controller is designed to solve the expected signal of the lateral position error equivalent yaw angle.

[0017] Based on the desired lateral position error equivalent yaw angle signal, the drone is tracked using a yaw angle stabilization tracking system. Since the design of yaw angle stabilization tracking systems for drones varies slightly for different models and has established design methods, and there's no need to specify a specific yaw angle stabilization tracking system, the methods we provide are adaptable and therefore will not be detailed here. Once the drone has tracked the desired lateral position error equivalent yaw angle signal, it has achieved the first stage of lateral tracking and guidance of the target.

[0018] Step S30: When the time is greater than the second stage guidance start time, the second stage of the UAV guiding the target begins.

[0019] The second stage of guidance uses angle tracking guidance. Its basic principle is to set the UAV's altitude command signal as the target's altitude signal, set the UAV's lateral command signal as the lateral azimuth angle formed between the UAV and the target to form a guidance composite signal, and then use the UAV's yaw angle stabilization tracking system to track it.

[0020] First, the lateral position error signal is calculated based on the target's lateral position signal and the drone's own lateral position signal. The longitudinal position error signal is then calculated based on the target's longitudinal position signal and the drone's own longitudinal position signal. The lateral azimuth angle signal is then calculated based on the lateral position error signal and the longitudinal position error signal. A first-order filter differentiator is then used to calculate the lateral azimuth angle filtered differential signal. Finally, the lateral azimuth angle signal and the lateral azimuth angle filtered differential signal are combined to produce the lateral angle guidance composite signal.

[0021] In the longitudinal direction, the height signal of the target set by the UAV is the height command signal of the UAV. Just like the first-stage guidance method, the equivalent pitch angle expected signal of the height error is solved, and then the UAV pitch angle stabilization tracking system completes the tracking, realizing the second-stage tracking and guidance of the UAV to the target height.

[0022] In the lateral direction, the UAV's yaw angle stabilization tracking system tracks the lateral angle guidance integrated signal to complete the second stage of the UAV's tracking and guidance of the target's lateral position.

[0023] Step S40: When the time is greater than the start time of the third stage guidance, the third stage of the UAV guidance of the target begins.

[0024] The third stage uses position tracking guidance. Its basic principles are the same as those of the first stage. However, building on the first stage, longitudinal and lateral nonlinear anti-saturation adaptive coefficients and nonlinear anti-saturation adaptive terms are introduced to generate the desired pitch angle equivalent to the altitude error and the desired yaw angle equivalent to the lateral position error as the final guidance signals. In this stage, the drone's altitude command signal is set to the target's altitude signal, and the drone's lateral command signal is set to the target's lateral position signal. This will not be repeated here.

[0025] The third-stage guidance uses position tracking to improve guidance accuracy for low-speed targets. To enhance terminal accuracy, a nonlinear anti-saturation fast adaptive compensation term was added during the design process. Details of this design are provided below in the Specific Implementation Implementation. The second-stage guidance uses azimuth angle guidance because it reduces the workload of position guidance. Otherwise, the attitude equivalent signal generated by position guidance could easily saturate the attitude system, reducing guidance accuracy.

[0026] Beneficial effects An embodiment of the present invention proposes a high-precision target tracking method for an unmanned aerial vehicle based on position and attitude switching; the basic principle is to use high-precision position proportional filter differential guidance to switch to azimuth guidance and finally switch to high-precision proportional filter differential high-precision guidance. The advantage is that the high-precision proportional filter differential guidance used at the end has good accuracy, which is better than traditional proportional guidance and azimuth guidance; traditional proportional guidance is suitable for guidance between high-speed moving bodies, but when targeting ground targets, especially small ground targets, the guidance sometimes has the problem of insufficient accuracy. The present invention switches back to azimuth angle guidance in the second stage, employing a combination of azimuth angle and azimuth rate. This approach combines the advantages of proportional guidance and azimuth guidance, while also avoiding the limited scope and space limitations of single-position tracking guidance. Specifically, when targets are present over a large area, the proportional filter differential of single-position guidance is prone to angle saturation. This means that the equivalent attitude angle is too large, rendering the UAV's attitude stabilization system unable to track, leading to guidance failure. The azimuth attitude guidance provided by the present invention is divided into large-signal and small-signal categories, and employs a lookup table-based attitude command generation approach, effectively resolving the instability caused by guidance parameter mismatches resulting from excessively large attitude commands. Furthermore, the present invention proposes a terminal high-precision guidance method that superimposes a nonlinear anti-saturation fast adaptive term on a position lag differential. This combination of nonlinear anti-saturation and adaptive design can, in principle, rapidly eliminate position static errors, effectively improving the UAV's tracking and guidance accuracy. In summary, the proposed three-stage guidance is highly valuable in engineering applications and is particularly suitable for UAV guidance, tracking, and strikes against small, low-speed ground targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The accompanying drawings are incorporated into and constitute a part of this specification, illustrate embodiments consistent with the present invention, and together with the description, serve to explain the principles of the present invention. Obviously, the drawings described below are only some embodiments of the present invention, and it is clear that those skilled in the art can derive other drawings based on these drawings without inventive effort.

[0028] Figure 1 It is a design flow chart of the method provided by the present invention; Figure 2 is a Simulink schematic block diagram of a local control structure of the method provided in an embodiment of the present invention; Figure 3 is a Simulink schematic diagram of a difference table of the method provided in an embodiment of the present invention; Figure 4 This is a graph showing the longitudinal positions of the drone and the target in Example 1 of the method provided by an embodiment of the present invention (unit: meter); Figure 5 This is a graph of the height positions of the drone and the target in Example 1 of the method provided by the embodiment of the present invention (unit: meter); Figure 6 This is a graph of the lateral positions of the drone and the target in Example 1 of the method provided by an embodiment of the present invention (unit: meter); Figure 7 This is a graph of the speed curves of the drone and the target in Example 1 of the method provided by an embodiment of the present invention (unit: meters per second); Figure 8 This is a horizontal relative motion curve diagram of the drone and the target in Example 1 of the method provided by the embodiment of the present invention (unit: meter); Figure 9 This is a distance curve diagram (unit: meter) between the drone and the target in Example 1 of the method provided by the embodiment of the present invention; Figure 10 This is a graph showing the end-point off-target amount amplification curve of Case 1 of the method provided in an embodiment of the present invention (unit: meter); Figure 11 This is a graph of the speed curves of the drone and the target in Example 2 of the method provided by the embodiment of the present invention (unit: meters per second); Figure 12 is a lateral position curve diagram of the drone and the target in Example 2 of the method provided by the embodiment of the present invention (unit: meter); Figure 13 This is a horizontal relative motion curve diagram of the drone and the target in Example 2 of the method provided by the embodiment of the present invention (unit: meter); Figure 14 is a distance curve diagram (unit: meter) between the drone and the target in Example 2 of the method provided by the embodiment of the present invention; Figure 15 This is a case two-terminal off-target amount amplification curve diagram (unit: meter) of the method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0029] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific implementations described herein are intended to explain the present invention rather than to limit the present invention. Figure 1 This is a flow chart of a method for high-precision tracking of a target by a UAV based on position and attitude switching provided by the present invention. Figure 1 Said method may specifically include the following steps: Step S10: Install a radar device or a visual imaging device on the drone to calculate the target's altitude signal, longitudinal position signal, and lateral position signal; at the same time, calculate the drone's own altitude signal, lateral position signal, and longitudinal position signal based on the drone's own inertial navigation device, and at the same time calculate the distance between the drone and the target.

[0030] The target's altitude signal is recorded as , the longitudinal position signal is recorded as and the lateral position signal ; The UAV's own altitude signal is recorded as , the lateral position signal is recorded as and the longitudinal position signal is denoted as The distance between the drone and the target The solution is as follows: ; Set the start time of the second stage guidance, the start time of the third stage guidance, the altitude error threshold, and the lateral position error threshold. The second stage guidance starts when the distance between the drone and the target reaches the first threshold of the guidance switching distance. The third stage guidance starts when the distance between the drone and the target reaches the second threshold of the guidance switching distance. The altitude error threshold, lateral position error threshold, first threshold of the guidance switching distance, and second threshold of the guidance switching distance are constant parameters. Adjust and set them according to the actual situation of the guidance.

[0031] In this case, the first threshold of the guidance switching distance is set , in meters, when the distance between the drone and the target equal The moment is recorded as ; Set the second threshold of the guidance switching distance , when the distance between the drone and the target equal The moment is recorded as . Set the height error threshold , lateral position error threshold .

[0032] Step S20: After the guidance starts, when the time is less than the second stage guidance start time, the first stage of the UAV guiding the target begins.

[0033] The first stage of guidance uses position tracking guidance. Its basic principle is to set the UAV's altitude command signal as the target's altitude signal, and set the UAV's lateral command signal as the target's lateral position signal.

[0034] For ease of understanding, the following additional explanation is provided: the altitude command signal of the UAV is from the perspective of the UAV's altitude control system. The same is true for the UAV's lateral command signal. That is, the altitude command signal is the input signal of the UAV's altitude loop from the perspective of the altitude control system; similarly, the target's lateral position signal is used as the input signal of the UAV's lateral loop.

[0035] The height error signal is calculated based on the target's height signal and the drone's own height signal; then, based on the height error, an inertia lag link is designed to obtain the height error inertia lag signal.

[0036] Specifically, the height error signal is recorded as , which is calculated as follows: The height error inertial hysteresis signal is recorded as , which is calculated as follows: ; in is the lag time constant, is the differential operator of the transfer function. In this case, This parameter cannot be set too large. Generally, Because the lag time here is too large, it will affect the performance of the entire system. This lag design can play a filtering role and eliminate clutter.

[0037] When the height error inertial hysteresis signal is greater than the height error threshold, the variable coefficient nonlinear height transformation method is used to solve the height error equivalent pitch angle expected signal; Specifically, the expected signal of the equivalent pitch angle of the height error is recorded as , which is calculated as follows: when hour, ; in , , This parameter is usually fixed and mainly plays the role of converting degrees to radians. The calculation method uses the following Matlab table form for interpolation and extrapolation, that is, when hour, Select in sequence ; and when When between nodes or outside nodes, basic linear interpolation or linear extrapolation methods are used. For details, please refer to the table difference method of Matlab, which will not be repeated here. The schematic diagram of its local control structure Siulink is as follows Figure 2 As shown, the difference table Simulink schematic diagram is as follows Figure 3 shown.

[0038] When the height error inertial hysteresis signal is less than the height error threshold, an inertial filter differential proportional controller is designed to solve the height error equivalent pitch angle expected signal. when When the inertia differential signal of the height error is obtained, the inertia differential signal of the height error is obtained by designing the inertia filter differential link. as follows: ; in , is the differential operator of the transfer function.

[0039] Secondly, the desired unsaturated limiting signal of the pitch angle is obtained by combining the height error inertial hysteresis signal with the height error inertial differential signal as follows: ; in ; is the expected unsaturated clipped signal for the pitch angle.

[0040] Finally, the expected unsaturated pitch angle limiting signal is limited to obtain the expected pitch angle signal equivalent to the height error as follows: ; in , which is a constant parameter used for saturation limiting, is the differential operator of the transfer function.

[0041] Based on the altitude error equivalent pitch angle desired signal, the drone is tracked using a pitch angle stabilization tracking system. While the design of these systems varies slightly for different drone models, there are standardized, proven design methods. Furthermore, there's no need to specify a specific pitch angle stabilization tracking system; the methods we provide are generally applicable and therefore will not be elaborated upon. Once the drone has tracked the altitude error equivalent pitch angle desired signal, it has achieved tracking and guidance of the target at the first stage of altitude.

[0042] The above is a description of the longitudinal design, and the following describes the lateral design.

[0043] The lateral position error signal is calculated based on the target's lateral position signal and the drone's own lateral position signal. Then, based on the lateral position error, an inertia hysteresis link is designed to obtain a lateral position error inertia hysteresis signal. Specifically, the lateral position error signal is recorded as , which is calculated as follows: The lateral position error inertia lag signal is recorded as , which is calculated as follows: ; in is the lag time constant, is the differential operator of the transfer function. In this case, .

[0044] When the lateral position error inertia hysteresis signal is greater than the lateral position error threshold, the variable coefficient nonlinear lateral position transformation method is used to solve the lateral position error equivalent yaw angle expected signal; Specifically, the lateral position error equivalent yaw angle expected signal is recorded as , which is calculated as follows: when hour, ; in , , , The calculation method is in the following table form, that is, when hour, Select in sequence ; and when When between nodes or outside nodes, linear interpolation or linear extrapolation methods are used. For details, please refer to Matlab's table difference method, which will not be repeated here.

[0045] When the lateral position error inertial hysteresis signal is less than the lateral position error threshold, an inertial filter differential PID controller is designed to solve the expected signal of the lateral position error equivalent yaw angle. when When the lateral position error inertia lag signal is used, the inertia filter differential link is designed to obtain the lateral position error inertia differential signal. as follows: ; in , is the differential operator of the transfer function.

[0046] Secondly, the expected unsaturated limiting signal of the yaw angle is obtained by combining the lateral position error inertial hysteresis signal with the lateral position error inertial differential signal as follows: ; in ; The expected unsaturated clipped signal for the yaw angle.

[0047] Finally, the expected unsaturated limited signal of the yaw angle is limited to obtain the expected signal of the lateral position error equivalent to the yaw angle as follows: ; in , is the differential operator of the transfer function.

[0048] Based on the desired lateral position error equivalent yaw angle signal, the drone is tracked using a yaw angle stabilization tracking system. While the design of yaw angle stabilization tracking systems for drones varies slightly for different models, there are standardized, proven design methods. Furthermore, there's no need to specify a specific yaw angle stabilization tracking system; all methods we provide are applicable, so we won't elaborate on them here. Once the drone has tracked the desired lateral position error equivalent yaw angle signal, it has achieved the first stage of lateral tracking and guidance of the target.

[0049] Step S30: When the time is greater than the second stage guidance start time, the second stage of the UAV guiding the target begins.

[0050] The second stage of guidance uses angle tracking guidance. Its basic principle is to set the UAV's altitude command signal as the target's altitude signal, set the UAV's lateral command signal as the lateral azimuth angle formed between the UAV and the target to form a guidance composite signal, and then use the UAV's yaw angle stabilization tracking system to track it.

[0051] First, the lateral position error signal is calculated based on the target's lateral position signal and the drone's own lateral position signal. The longitudinal position error signal is then calculated based on the target's longitudinal position signal and the drone's own longitudinal position signal. The lateral azimuth angle signal is then calculated based on the lateral position error signal and the longitudinal position error signal. A first-order filter differentiator is then used to calculate the lateral azimuth angle filtered differential signal. Finally, the lateral azimuth angle signal and the lateral azimuth angle filtered differential signal are combined to produce the lateral angle guidance composite signal.

[0052] Specifically, the lateral position error signal and the longitudinal position error signal are first calculated as follows: , , Secondly, the lateral azimuth angle signal is solved as follows: ; Then, a first-order filter differentiator is used to solve the lateral azimuth filtered differential signal as follows: ; in , , .

[0053] Finally, the lateral azimuth angle signal and the lateral azimuth angle filtered differential signal are combined to obtain the lateral angle guidance integrated signal as follows: ; in ; It is a comprehensive signal for lateral angle guidance.

[0054] In the longitudinal direction, the height signal of the target set by the UAV is the height command signal of the UAV. Just like the first-stage guidance method, the equivalent pitch angle expected signal of the height error is solved, and then the UAV pitch angle stabilization tracking system completes the tracking, realizing the second-stage tracking and guidance of the UAV to the target height.

[0055] In the lateral direction, the UAV's yaw angle stabilization tracking system tracks the lateral angle guidance integrated signal to complete the second stage of the UAV's tracking and guidance of the target's lateral position.

[0056] Step S40: When the time is greater than the start time of the third stage guidance, the third stage of the UAV guidance of the target begins.

[0057] The third stage of guidance uses position tracking guidance. Its basic principle is the same as the first stage, but in order to improve the accuracy of the terminal, a nonlinear anti-saturation fast adaptive compensation term is added during the design process. The core of this design is to increase the rapidity of the terminal guidance and eliminate position deviations, especially static position errors, through fast adaptation, thereby improving the accuracy of the terminal guidance. Specifically, it is divided into longitudinal and lateral directions. The longitudinal design is as follows: In the first step, the altitude command signal of the UAV is set as the altitude signal of the target. According to the altitude signal of the target and the altitude signal of the UAV itself, the altitude error signal is calculated. The altitude error signal is recorded as , which is calculated as follows: .

[0058] The second step is to solve the height error inertial lag signal and record it as , which is calculated as follows: ; in is the lag time constant, is the differential operator of the transfer function. In this case, .

[0059] When the altitude error inertial hysteresis signal is greater than the altitude error threshold, the variable coefficient nonlinear altitude transformation method is used to solve the altitude error equivalent pitch angle expected signal.

[0060] The third step is to solve the expected signal of the equivalent pitch angle of the height error, which is recorded as , which is calculated as follows: when hour: ; in , , , The calculation method is in the following table form, that is, when when hour, Select in sequence ; and when When between nodes or outside nodes, basic linear interpolation or linear extrapolation methods are used. For details, see Matlab's table difference method. It is the same as before and will not be repeated here.

[0061] When the height error inertial hysteresis signal is less than the height error threshold, an inertial filter differential proportional controller is designed to solve the expected signal of the height error equivalent pitch angle as follows.

[0062] The fourth step is when When the inertia differential signal of the height error is obtained, the inertia differential signal of the height error is obtained by designing the inertia filter differential link. as follows: ; in , is the differential operator of the transfer function.

[0063] The fifth step is to solve the longitudinal nonlinear adaptive factor based on the height error signal and the height error inertial differential signal as follows: ; in It is the longitudinal nonlinear adaptive factor, which is mainly composed of the first item of height error, the second item of inertia differential of height error, and the third item of hinge product of height error and its inertia differential. The design principle of the third item is that as long as one of the height error and its inertia differential is not stable to 0, then the value of this item is not 0, which will start the dynamic process of rapid adaptation. If only the third item is designed, then when one of them is 0, the adaptive process will stop. Therefore, if one of the height error or its differential is 0, the nonlinear adaptive factor will still work because one of the first two items must work. 、 、 are constant parameters, which are used to adjust the weight factors of the first, second and third items respectively. 、 、 .

[0064] In the sixth step, the longitudinal anti-saturation nonlinear adaptive term is solved according to the longitudinal nonlinear adaptive factor, and the longitudinal nonlinear anti-saturation adaptive coefficient is obtained by cumulative operation as follows: ; ; in is the longitudinal anti-saturation nonlinear adaptive term obtained by anti-saturation transformation of the longitudinal nonlinear adaptive factor. Its design principle is , has anti-saturation function; and is a constant positive parameter, which can be selected as , other cases can be selected around 2 according to actual conditions. is the longitudinal nonlinear anti-saturation adaptive coefficient; and Respectively No. Data and , Indicates the time period of its data interval, which is selected as .

[0065] In the seventh step, a linear combination of the altitude error inertial lag signal, the altitude error inertial differential signal, the longitudinal nonlinear anti-saturation adaptive coefficient, and the longitudinal anti-saturation nonlinear adaptive term is performed to obtain the desired unsaturated limiting signal of the pitch angle as follows: ; is a constant parameter used to adjust the convergence rate of the entire nonlinear anti-saturation adaptation. In this case, it is set to ;in ; is the expected unsaturated clipped signal for the pitch angle.

[0066] In the eighth step, the pitch angle expected unsaturated limited signal is limited to obtain the height error equivalent pitch angle expected signal as follows: ; in , is the differential operator of the transfer function.

[0067] The above is the longitudinal design process; in the lateral design, the lateral command signal of the UAV is set as the lateral position signal of the target. First, the lateral position error signal is solved and recorded as , which is calculated as follows: The lateral position error inertia lag signal is recorded as , which is calculated as follows: ; when When , the expected signal of the equivalent yaw angle of the lateral position error is explained as follows: ; in , , , The calculation method is in the following table form, that is, when hour, Select in sequence ; and when When between nodes or outside nodes, linear interpolation or linear extrapolation methods are used, which are the same as before and will not be repeated here.

[0068] When the lateral position error inertial hysteresis signal is less than the lateral position error threshold, an inertial filter differential PID controller is designed to solve the expected signal of the lateral position error equivalent yaw angle. when When the lateral position error inertia lag signal is used, the inertia filter differential link is designed to obtain the lateral position error inertia differential signal. as follows: ; in , is the differential operator of the transfer function.

[0069] Secondly, similarly, in the longitudinal manner, the lateral position error inertial hysteresis signal and the lateral position error inertial differential signal, the lateral nonlinear anti-saturation adaptive coefficient and the lateral anti-saturation nonlinear adaptive term are combined to obtain the expected unsaturated limiting signal of the yaw angle as follows: ; ; ; ; in It is the lateral nonlinear adaptive factor, which is mainly composed of the first term lateral position error, the second term lateral position error inertia differential, and the third term hinge product of the lateral position error and its inertia differential. The design principle of the third term is that as long as one of the lateral position error and its inertia differential is not stable to 0, the value of this term is not 0, which will start the dynamic process of rapid adaptation. If only the third term is designed, then when one of them is 0, the adaptive process will stop. Therefore, if one of the lateral position error or its differential is 0, the nonlinear adaptive factor will still work because one of the first two terms must work. 、 、 are constant parameters, which are used to adjust the weight factors of the first, second and third lateral terms. 、 、 . is the lateral anti-saturation nonlinear adaptive term obtained by the anti-saturation transformation of the lateral nonlinear adaptive factor. Its design principle is the same as above and it has the anti-saturation function. is a constant positive parameter, which can be selected as , other cases can be selected around 2 according to actual conditions. is the lateral nonlinear anti-saturation adaptive coefficient; and Respectively No. Data and data. is a constant parameter used to adjust the convergence rate of the entire lateral nonlinear anti-saturation adaptation. In this case, it is set to ; ; The expected unsaturated clipped signal for the yaw angle.

[0070] Finally, the expected unsaturated limited signal of the yaw angle is limited to obtain the expected signal of the lateral position error equivalent to the yaw angle as follows: ; in , is the differential operator of the transfer function.

[0071] The drone's yaw angle stabilization tracking system tracks the desired lateral position error equivalent yaw angle signal, completing high-resolution tracking guidance of the target in the third stage. Position tracking guidance in the third stage is intended to improve guidance accuracy for low-speed targets. Azimuth guidance is used in the second stage because it reduces the workload of position guidance. Otherwise, if position guidance were used in the second stage, the attitude equivalent signal generated by the position guidance would easily saturate the attitude system, resulting in reduced guidance accuracy.

[0072] Case Implementation 1 In order to verify the feasibility and effectiveness of the above method, the following case simulation analysis of a UAV attacking a low-speed ground target is carried out.

[0073] The initial position of the target is 15,000 meters vertically, 0 meters in height, and 0 meters in the lateral direction. After the start, the target moves laterally to 50 meters. After 50 seconds, the target moves at a constant speed in a direction exceeding 30 degrees laterally. The initial speed of the target movement is 0 meters per second, and then it moves at a constant speed of 20 meters per second.

[0074] Before the drone begins guidance, its initial speed is 100 meters per second. After takeoff, it accelerates to a constant cruising speed of 200 meters per second. Its initial position is 0 meters vertically, 1000 meters high, and 0 meters laterally. After takeoff and before entering guidance, the drone cruises at an altitude of 500 meters and a lateral distance of -350 meters. After approximately 30 seconds, it enters the first stage of guidance.

[0075] According to the three-stage guidance provided by the present invention, the final guidance result is as follows, where the longitudinal position of the UAV and the target is as follows: Figure 4 As shown; the altitude position of the UAV and the target is as shown Figure 5 The lateral position of the UAV and the target is shown in Figure 6 As shown; the speed curves of the UAV and the target are as follows Figure 7 As shown, the relative motion between the drone and the target is as follows: Figure 8 As shown, the distance curve between the UAV and the target is as follows Figure 9 As shown, the terminal off-target amount amplification curve is as follows Figure 10 As shown in the figure, the final miss distance is less than 2 meters, and the three-stage guidance method can hit the target.

[0076] Case Implementation 2 The initial position of the target is 12,000 meters vertically, 0 meters in height, and 0 meters in the lateral direction. After the start, the target moves laterally to 50 meters. After 50 seconds, the target moves at a constant speed in a direction exceeding 20 degrees laterally. The initial speed of the target is 0 meters per second, and then it moves at a constant speed of 15 meters per second.

[0077] Before the drone begins guidance, its initial speed is 100 meters per second. After takeoff, it accelerates to a constant cruising speed of 150 meters per second. Its initial position is 0 meters vertically, 1000 meters high, and 0 meters laterally. After takeoff and before entering guidance, the drone cruises at an altitude of 300 meters and a lateral distance of 150 meters. After approximately 30 seconds, it enters the first stage of guidance.

[0078] According to the three-stage guidance provided by the present invention, the final guidance result is as follows: Figure 12 As shown; the speed curves of the UAV and the target are as follows Figure 13 As shown, the relative motion between the drone and the target is as follows: Figure 14 As shown, the distance curve between the UAV and the target is as follows Figure 12 As shown, the terminal off-target amount amplification curve is as follows Figure 15 As shown in the figure, the final miss distance is less than 0.4 meters, and the three-stage guidance method is able to hit the target. This shows that the three-stage UAV guidance of the target using position and attitude switching has excellent hit accuracy, which is significantly improved compared to traditional guidance methods. This is also the reason why this invention has high engineering value.

[0079] Please note that the above are only embodiments of the present invention and the technical principles used therein, but the present invention is not limited to the above embodiments. Without departing from the concept of the present invention, it can also include more equivalent embodiments. The scope of the present invention is determined by the scope of the attached claims.

Claims

1. A method for high-precision tracking of a target by a UAV based on position and attitude switching, comprising the following steps: Step 1: Install radar equipment or visual imaging equipment on the drone to calculate the target's altitude signal, longitudinal position signal and lateral position signal; at the same time, use the drone's own inertial navigation equipment to calculate the drone's own altitude signal, lateral position signal and longitudinal position signal, and at the same time calculate the distance between the drone and the target Step 2: When the time is less than the start time of the second-stage guidance, the first-stage UAV guidance of the target is started, the UAV's altitude command signal is set to the target's altitude signal, and the UAV's lateral command signal is set to the target's lateral position signal; the altitude error signal is calculated according to the target's altitude signal and the UAV's own altitude signal; then, according to the altitude error signal, an inertia lag link is designed to obtain an altitude error inertia lag signal; when the altitude error inertia lag signal is greater than the altitude error threshold, a variable coefficient nonlinear altitude transformation method is used to solve the altitude error equivalent pitch angle expected signal; when the altitude error inertia lag signal is less than the altitude error threshold, an inertial filter differential proportional controller is designed to solve the altitude error equivalent pitch angle expected signal; according to the altitude error equivalent pitch angle expected signal, the UAV pitch angle stabilization tracking system is used to track it to achieve the first-stage altitude tracking and guidance of the target; the lateral first-stage guidance design adopts the same method as the altitude tracking and guidance; Step 3: When the time is greater than the start time of the second-stage guidance, the second-stage UAV guidance of the target begins; the second-stage guidance adopts angle tracking guidance, the basic principle of which is to set the UAV's altitude command signal as the target's altitude signal, solve the lateral azimuth angle formed between the UAV and the target, use a first-order filter differentiator to solve the lateral azimuth angle filter differential signal, and finally combine the lateral azimuth angle signal with the lateral azimuth angle filter differential signal to obtain the lateral angle guidance integrated signal, and then use the UAV yaw angle stabilization tracking system to track it laterally; Step 4: When the time is greater than the start time of the third stage guidance, the third stage of the UAV's guidance of the target begins. The third stage guidance adopts position tracking guidance. Its basic principle is basically the same as the first stage, but in the design process, a nonlinear anti-saturation term and a fast adaptive compensation term are added to generate the expected signal of the equivalent pitch angle of the height error and the expected signal of the equivalent yaw angle of the lateral position error as the guidance signal of the guidance segment, thereby increasing the terminal guidance accuracy.

2. The method according to claim 1, characterized in that The altitude error signal is calculated based on the target altitude signal and the UAV's own altitude signal. Then, based on the altitude error, an inertial hysteresis link is designed to obtain an altitude error inertial hysteresis signal. When the altitude error inertial hysteresis signal is greater than the altitude error threshold, a variable coefficient nonlinear altitude transformation method is used to solve the altitude error equivalent pitch angle desired signal. When the altitude error inertial hysteresis signal is less than the altitude error threshold, an inertial filter differential proportional controller is designed, and a nonlinear anti-saturation term and a fast adaptive compensation term are added to generate the altitude error equivalent pitch angle desired signal as follows: ； ； in is the height error signal, is the height error inertial hysteresis signal; is the lag time constant, is the differential operator of the transfer function; is the target height signal, It is the altitude signal of the UAV itself; when hour, ; ; in is the height error threshold, which is a preset constant parameter. is the expected signal of the equivalent pitch angle of height error, 、 is a constant parameter, where , this parameter is fixed. The calculation method uses Matlab table form to perform interpolation and extrapolation, that is, when hour, Select in sequence ; and when When between nodes or outside nodes, the commonly used linear interpolation or linear extrapolation method is used to solve ; when When designing ； ； ； ； ； ； in is the height error inertial differential signal, 、 is a constant parameter, is the differential operator of the transfer function; It is the longitudinal nonlinear adaptive factor, which is mainly composed of the first height error, the second height error inertia differential, and the third height error and its inertia differential hinge product; 、 、 are constant parameters, which are used to adjust the weight factors of the first, second and third items respectively; is the longitudinal anti-saturation nonlinear adaptive term obtained by anti-saturation transformation of the longitudinal nonlinear adaptive factor; and is a constant positive parameter, is the longitudinal nonlinear anti-saturation adaptive coefficient; and Respectively No. Data and , Indicates the time period of its data interval; is a constant parameter used to adjust the convergence rate of the entire nonlinear anti-saturation adaptation. is a constant control parameter used for saturation limiting. is the expected unsaturated clipped signal for the pitch angle; is the expected signal of the equivalent pitch angle of the height error.

3. The method according to claim 1, characterized in that Solve the lateral azimuth angle formed between the UAV and the target, use a first-order filter differentiator to solve the lateral azimuth angle filter differential signal, and finally combine the lateral azimuth angle signal with the lateral azimuth angle filter differential signal to obtain the lateral angle guidance integrated signal. Then, the UAV yaw angle stabilization tracking system is used to track it laterally, including: ， ， ； ； ； in is the target longitudinal position signal, is the target lateral position signal; is the longitudinal position signal of the UAV, is the lateral position signal of the UAV, is the lateral position error signal, is the longitudinal position error signal, is the lateral azimuth angle signal; is the lateral azimuth filtered differential signal, 、 、 、 It is a constant parameter and is adjusted according to the actual project conditions; It is a comprehensive signal for lateral angle guidance.

Images