Rocket first sublevel descending section grid rudder control method

Abstract

The invention discloses a control method for a grid rudder of a first sublevel descending section of a rocket, particularly relates to the technical field of spaceflight control, and is used for solving the problem of poor speed reduction effect due to the fact that the descending section only depends on induced resistance. Firstly, a channel rudder deflection angle is calculated by combining an instruction attitude angle, an actual attitude angle and a rotation angular velocity; then, basic rudder deflection angles of the four grid rudders are obtained based on the control distribution relation; the method comprises the following steps: extracting deflection boundary margins of a use limit and a basic rudder deflection angle, and selecting a minimum value for optimizing and assigning to determine a conversion regulating variable; and finally, reconstructing a four-dimensional space control instruction by using a four-order orthogonal transformation matrix, and driving a control surface to deflect. The control surface is promoted to deflect to the limit, and scientific support is provided for efficient speed reduction of a rocket and propellant saving.

Description

Technical Field

[0001] This invention relates to the field of aerospace control technology, specifically to a grid fin control method for the descent phase of a rocket's first stage. Background Technology

[0002] With the development of space launch vehicle technology, the deceleration and recovery control of the first-stage rocket during the descent phase has become particularly critical. Typically, a grid fin is positioned at the top of the first-stage rocket. When the rocket is in the descent phase and the grid fin is deployed, the airflow over the fin generates control force and aerodynamic drag. Since the drag coefficient of the grid fin changes positively with the deflection angle, fully utilizing the aerodynamic drag of the grid fin for deceleration during descent is an important way to reduce the workload of the landing retro-rocket engine and save propellant consumption.

[0003] In existing first-stage descent recovery control methods, the system relies solely on the induced drag generated by the rocket body and grid fins under normal attitude control for deceleration. These methods do not actively control the grid fins for dedicated deceleration; the deflection angle allocated to the grid fins under normal attitude control is relatively small, resulting in a consistently low drag coefficient and poor aerodynamic drag. This forces the rocket to consume more propellant for retro-thrust deceleration during the recovery and landing phase, reducing the rocket's payload efficiency and economic benefits. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a grid fin control method for the descent phase of a rocket's first stage, thus solving the problems mentioned in the background.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a grid fin control method for the descent phase of a rocket first stage, comprising the following steps: S1, obtaining the pitch, yaw, and roll command attitude angles of the first-stage rocket based on the landing mission and guidance equations, obtaining the attitude angle deviations in conjunction with the actual attitude angles, and extracting the rotational angular velocities around the three axes of the rocket body, and calculating the channel fin deflection angles of the three channels based on the attitude angle deviations and rotational angular velocities; S2, performing control quantity allocation on the channel fin deflection angles according to the preset control allocation relationship from channel fins to individual fins, and obtaining the basic fin deflection angles of each of the four grid fins; S3, obtaining the preset grid fin usage restrictions, and extracting the grid fin usage restrictions and the basic fin deflection angles of the four grid fins. The deflection boundary margin between the rudder deflection angles, combined with the corresponding adjustment coefficient in the orthogonal transformation matrix, establishes the remaining usable rudder deflection margin for each of the four grid rudders. The minimum value among the rudder deflection margins of multiple grid rudders is selected as the magnitude of the adjustment amount, and the optimal value is assigned according to the sign direction of the base rudder deflection angle of the grid rudder that produces the minimum value, thus establishing the conversion adjustment amount; S4, construct a preset fourth-order orthogonal transformation matrix, and concatenate the channel rudder deflection angle and the conversion adjustment amount to construct a four-dimensional space control command. Through the fourth-order orthogonal transformation matrix, perform spatial mapping and feature decoupling on the four-dimensional space control command to reconstruct the final rudder deflection angle command for the four grid rudders. The final rudder deflection angle command is sent to the servo driver to drive the corresponding grid rudder deflection.

[0006] Furthermore, based on the landing mission and guidance equations, the commanded attitude angles of the first-stage rocket's pitch, yaw, and roll channels are obtained. The specific process of obtaining the attitude angle deviation by combining the actual attitude angles is as follows: The guidance equations are called to generate the commanded attitude angles of the pitch, yaw, and roll channels corresponding to the landing mission, and these are spliced ​​together to construct a three-dimensional commanded attitude angle vector; the actual attitude angles of the pitch, yaw, and roll channels fed back in real time by the onboard sensors are obtained, and these are spliced ​​together to construct a three-dimensional actual attitude angle vector; the node alignment deviation is calculated between the three-dimensional commanded attitude angle vector and the three-dimensional actual attitude angle vector to solve for the attitude angle deviation vector containing pitch deviation components, yaw deviation components, and roll deviation components.

[0007] Furthermore, the specific process of extracting the rotational angular velocities around the three axes of the rocket body and calculating the channel rudder deflection angles of the three channels based on the attitude angle deviation and rotational angular velocities is as follows: Obtain the pitch rotational angular velocities, yaw rotational angular velocities, and roll rotational angular velocities around the three axes of the rocket body from the feedback of the onboard measurement reference, and construct them as a rotational angular velocity vector; Construct a preset proportional control gain diagonal matrix and a differential control gain diagonal matrix; Perform feature mapping on the proportional control gain diagonal matrix and the attitude angle deviation vector to obtain the proportional adjustment component, and perform feature mapping on the differential control gain diagonal matrix and the rotational angular velocity vector to obtain the differential adjustment component; Perform linear state fusion on the proportional adjustment component and the differential adjustment component to obtain a channel rudder deflection angle vector containing the pitch channel rudder deflection angle, the yaw channel rudder deflection angle, and the roll channel rudder deflection angle.

[0008] Furthermore, based on the preset control allocation relationship from the channel rudder to the individual rudder, the control quantity allocation is performed on the channel rudder deflection angle to obtain the basic rudder deflection angle of each of the four grid rudders. The specific process is as follows: Based on the quadrant arrangement structure of the four grid rudders on the first-stage rocket body, a preset three-column four-row control quantity allocation matrix is ​​constructed; the control quantity allocation matrix and the channel rudder deflection angle vector are spatially upgraded and mapped, and the first basic rudder deflection angle, the second basic rudder deflection angle, the third basic rudder deflection angle and the fourth basic rudder deflection angle corresponding to the four grid rudders are decoupled and separated, and the basic rudder deflection angle is aggregated to construct the basic rudder deflection angle vector.

[0009] Furthermore, the specific process of obtaining the preset grid rudder usage limit, extracting the deflection boundary margin between the grid rudder usage limit and the basic rudder deflection angle of the four grid rudders, and combining the corresponding adjustment coefficients in the orthogonal transformation matrix to determine the remaining usable rudder deflection margin of each of the four grid rudders is as follows: extract the absolute value of the basic rudder deflection angle of each of the four grid rudders in the basic rudder deflection angle vector, perform boundary alignment calculation between the grid rudder usage limit and the absolute value to obtain the absolute deflection boundary margin of each grid rudder; extract the sign and direction parameters of the basic rudder deflection angle of each of the four grid rudders in the basic rudder deflection angle vector; retrieve the specific adjustment coefficient column vector in the preset fourth-order orthogonal transformation matrix; perform element-wise feature multiplication fusion on the absolute deflection boundary margin, sign and direction parameters, and corresponding row elements in the adjustment coefficient column vector to calculate the remaining usable first rudder deflection margin, second rudder deflection margin, third rudder deflection margin, and fourth rudder deflection margin of each of the four grid rudders.

[0010] Furthermore, the minimum value among the rudder deflection margins of multiple grid rudders is selected as the magnitude of the adjustment amount. The specific process for establishing the conversion adjustment amount is as follows: The absolute value parameters of the first, second, third, and fourth rudder deflection margins are extracted. Global boundary optimization is performed on these absolute value parameters to lock the target absolute value parameter that exhibits a minimum value. The target absolute value parameter is mapped back to the corresponding original grid rudder control node, and the sign direction parameter of the corresponding basic rudder deflection angle of the original grid rudder control node is extracted. The target absolute value parameter and the extracted sign direction parameter are fused and assigned to establish and output the conversion adjustment amount with spatial directional characteristics.

[0011] Furthermore, the specific process of constructing a preset fourth-order orthogonal transformation matrix and splicing the channel rudder deflection angle and the conversion adjustment amount to construct a four-dimensional spatial control command is as follows: Construct a fourth-order orthogonal transformation matrix whose internal element distribution satisfies that the product of itself and its transpose matrix is ​​equal to a constant multiple of the identity matrix; extract the pitch channel rudder deflection angle, yaw channel rudder deflection angle, and roll channel rudder deflection angle from the channel rudder deflection angle vector; arrange the pitch channel rudder deflection angle, yaw channel rudder deflection angle, roll channel rudder deflection angle, and conversion adjustment amount in the preset column dimension order, and splice them to construct a four-dimensional spatial control command column vector.

[0012] Furthermore, by performing spatial mapping and feature decoupling on the four-dimensional space control commands through a fourth-order orthogonal transformation matrix, the final deflection angle commands for the four grid fins are reconstructed. The specific process of sending the final deflection angle commands to the servo actuators to drive the corresponding grid fin deflection is as follows: A linear algebraic mapping is performed between the fourth-order orthogonal transformation matrix and the four-dimensional space control command column vector to reconstruct a four-dimensional output column vector containing the final deflection angle features of the four grid fins; Feature separation is performed on the four-dimensional output column vector to extract the first final deflection command corresponding to the first servo node, the second final deflection command corresponding to the second servo node, the third final deflection command corresponding to the third servo node, and the fourth final deflection command corresponding to the fourth servo node; The first, second, third, and fourth final deflection commands are sent to the servo actuator nodes in the corresponding physical quadrants of the first-stage rocket to drive the controlled object to complete the physical deflection of its posture.

[0013] The present invention has the following beneficial effects:

[0014] (1) A grid fin control method for the descent phase of a rocket first stage. This method obtains the three-channel command angles of the landing point mission and guidance equations, calculates the channel fin deflection angles by combining the actual attitude angles and rotational angular velocities, and then converts them into the basic swing angles of each grid fin using control allocation relationships. This process ensures that the attitude control requirements of rocket flight are accurately mapped to the physical actuators, guaranteeing stable control of the rocket's attitude. Accurately solving the basic attitude control commands and realizing the control quantity allocation of the physical actuators provides a stable and reliable attitude control benchmark for the descent phase of the rocket first stage, and provides solid data support for subsequent searches for the limit deflection margin.

[0015] (2) A grid fin control method for the descent phase of a rocket first stage. This method establishes an adjustment amount by extracting the deflection boundary margin between the grid fin usage limit and the basic swing angle. This adjustment amount, along with the channel fin deflection angle, is then substituted into an orthogonal transformation matrix for feature decoupling and reconstruction. This process, while fully satisfying the rocket attitude control constraints, fully utilizes the remaining deflection physical space of the grid fins, forcing them to deflect towards the maximum angle, thus significantly improving the system's drag coefficient and aerodynamic drag. By introducing an orthogonal transformation matrix and deflection adjustment amount, without interfering with the rocket's original flight attitude, the four grid fins are coordinated to deflect to the limit angle, significantly improving the overall aerodynamic drag coefficient of the spacecraft, significantly enhancing the deceleration effect, and effectively saving propellant consumption.

[0016] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0017] Figure 1 This is a flowchart of a grid fin control method for the descent phase of a rocket first stage according to the present invention.

[0018] Figure 2 This is a three-dimensional geometric image of the proportional and differential control gain functional in step S1.

[0019] Figure 3 This is the Bode plot of the pitch channel fine-tuned in the frequency domain during step S1. Detailed Implementation

[0020] This application provides a grid fin control method for the descent phase of a rocket's first stage, which solves the problems of poor deceleration and excessive propellant consumption caused by relying solely on conventional attitude control for induced drag in the prior art.

[0021] The overall concept of the solution in this application embodiment is as follows:

[0022] During the deceleration process in the first-stage descent phase, the effective information lost due to the transformation is defined as an independent adjustment quantity, and a multi-channel decoupling relationship is established using the constructed orthogonal transformation matrix. Under the constraint of ensuring that the basic attitude control requirements are not affected, the system automatically calculates and extracts the remaining usable minimum deflection margin of each grid rudder as the adjustment quantity input, and recalculates the final yaw angle command using the orthogonal transformation matrix, forcing the grid rudder to deflect to the physical limit angle as much as possible, thereby maximizing the drag coefficient, obtaining the extreme aerodynamic drag, and ultimately achieving the goal of efficient aerodynamic deceleration.

[0023] Please see Figure 1 This invention provides a technical solution: a grid fin control method for the descent phase of a rocket's first stage, comprising the following steps: S1, obtaining the pitch, yaw, and roll command attitude angles of the first-stage rocket based on the landing mission and guidance equations, obtaining the attitude angle deviations in combination with the actual attitude angles, and extracting the rotational angular velocities around the three axes of the rocket body, and calculating the channel fin deflection angles of the three channels based on the attitude angle deviations and rotational angular velocities; S2, performing control quantity allocation on the channel fin deflection angles according to the preset control allocation relationship from channel fins to individual fins, and obtaining the basic fin deflection angles of each of the four grid fins; S3, obtaining the preset grid fin usage restrictions, and extracting the grid fin usage restrictions and the basic fin deflection angles of the four grid fins. The remaining usable rudder deflection margin of each of the four grid rudders is determined by combining the deflection boundary margin between them with the corresponding adjustment coefficient in the orthogonal transformation matrix. The minimum value among the rudder deflection margins of multiple grid rudders is selected as the magnitude of the adjustment amount, and the optimal value is assigned according to the sign direction of the basic rudder deflection angle of the grid rudder that produces the minimum value, thus establishing the transformation adjustment amount; S4. A preset fourth-order orthogonal transformation matrix is ​​constructed, and the channel rudder deflection angle and the transformation adjustment amount are concatenated to construct a four-dimensional space control command. The four-dimensional space control command is spatially mapped and decoupled from its features through the fourth-order orthogonal transformation matrix to reconstruct the final rudder deflection angle command of the four grid rudders. The final rudder deflection angle command is sent to the servo driver to drive the corresponding grid rudder deflection.

[0024] In this implementation plan, step S1 is first executed to calculate the basic attitude control commands. After the onboard computer determines that the separation signal is valid and loads the recovery mission, the guidance equations calculate the current pitch, yaw, and roll three-channel command attitude angles. The channel rudders are abstract concepts for ease of control, corresponding to the rocket body's rotation around a three-dimensional right-handed coordinate system. Rotation around the Z-axis, Y-axis, and X-axis corresponds to the pitch, yaw, and roll channels, respectively. The system combines the actual attitude angles obtained from the sensors to derive the attitude angle deviations, and combines this with the rotational angular velocity around the rocket body to calculate the channel rudder deflection angles for these three channels. This step precisely transforms the overall landing guidance mission into specific rocket body attitude control requirements, providing benchmark input parameters for subsequent allocation of physical mechanism control quantities.

[0025] In step S2, the system performs the control quantity allocation from the channel fins to the physical actuators. Based on attitude control design practices and the arrangement of the grid fins, the system converts the previously calculated three-channel fin deflection angles into the basic fin deflection angles of the four grid fins. A grid fin is a small, grid-like aerodynamic control surface typically deployed in an X-shape on the top of a first-stage rocket, consisting of a pivot, a frame, and small fins arranged at ±45 degrees internally. When the grid fins are deployed, airflow passes through them, generating control force and drag due to the deflection of the fin surfaces. The drag coefficient is minimized when the deflection angle is zero, and increases with the deflection angle. This conversion calculation ensures that the theoretical attitude control quantities are reasonably mapped to the specific grid fins, and the resulting basic fin deflection angles provide the initial data support for subsequently finding the maximum deflection margin.

[0026] Step S3 aims to exploit the system's remaining deflection capability. The system has pre-defined limits on the use of grid fins. By calculating the difference between these limits and the base fin deflection angles obtained in step S2, the remaining usable fin deflection angles for each grid fin outside the control constraints are determined. Subsequently, the system selects the minimum value from these remaining usable angles as the adjustment amount, maintaining the same sign and direction as the original base fin deflection angle of the grid fin that generated the minimum value. The adjustment amount extracted here essentially represents the effective information lost during the conventional control transition. This process, while absolutely ensuring that the rocket's original attitude control remains undisturbed, accurately quantifies the deflection boundary margin of the grid fins from the physical or control limits, establishing a basis for maximizing angle compensation for subsequent increases in aerodynamic drag.

[0027] Step S4 maximizes aerodynamic deceleration through matrix reconstruction. The system constructs a complete fourth-order orthogonal transformation matrix. The control command, composed of the previous three-channel rudder deflection angles and the newly established adjustment values, is substituted into this matrix for recalculation, yielding the final tilt angle command for the four grid rudders. This command is then passed to the servo drive mechanism for deflection. The characteristics of this orthogonal transformation matrix ensure that the newly introduced adjustment values ​​remain orthogonal to the original pitch, yaw, and roll channel rudders, without interfering with each other. This step uses the identified adjustment margin to perform extreme corrections on the basic commands, forcing the grid rudders to deflect to the maximum angle possible. Based on fluid dynamics principles, when parameters such as characteristic area and relative velocity are relatively fixed, the drag coefficient is positively correlated with the deflection angle. This maximizes the drag coefficient, thereby generating extreme aerodynamic drag, achieving the ultimate goal of significantly improving aerodynamic deceleration and saving propellant consumption during the first-stage descent phase.

[0028] Specifically, the process of obtaining the pitch, yaw, and roll command attitude angles of the first-stage rocket based on the landing mission and guidance equations, and combining them with the actual attitude angles to obtain the attitude angle deviation is as follows: The guidance equations are called to generate the pitch, yaw, and roll command attitude angles corresponding to the landing mission, and these are then concatenated to construct a three-dimensional command attitude angle vector. The actual pitch, yaw, and roll attitude angles fed back in real-time by the onboard sensors are obtained and concatenated to construct a three-dimensional actual attitude angle vector. The node alignment deviation is calculated between the three-dimensional command attitude angle vector and the three-dimensional actual attitude angle vector to obtain the attitude angle deviation vector containing pitch deviation components, yaw deviation components, and roll deviation components.

[0029] In this implementation plan, this step aims to precisely translate the rocket's overall flight landing point mission into specific attitude control requirements. First, the onboard computer, by invoking built-in guidance equations, calculates the landing point mission into a desired flight attitude reference. This reference is then compared with the current attitude data collected in real-time by sensors, calculating the difference between the two along the three spatial axes. This process is achieved through deviation calculation, the specific formula of which is as follows: In the formula, Pitch deviation component; : Yaw deviation component; : Roll deviation component; Pitch channel command attitude angle; : Yaw channel command attitude angle; : Roll channel command attitude angle; : Actual attitude angle of the pitch channel; : Actual attitude angle of the yaw channel; : Actual attitude angle of the roll channel. Through the above formula, the system clarifies the specific degree of deviation of the current attitude from the target, providing a basic input for subsequent attitude correction. The descent guidance equation is divided into two parts: one part, before engine landing ignition, mainly relies on the grid fins; the other part, after engine landing ignition, mainly relies on the engine, with the grid fins as an auxiliary. This invention mainly discusses the period before engine landing ignition. Before engine landing ignition, the main task of the grid fin control is to control the descent trajectory and guide the rocket body to the predetermined ignition point. The formula for calculating the overload command through proportional guidance is as follows: ; In the above formula: , , , , , The current position and velocity information output by the navigation system is parallel to the launch coordinate system of the landing point, with the origin at the landing point launch point. The position information is equivalent to the longitude, latitude, and altitude of the satellite positioning and has a fixed transformation relationship. , , This refers to the target point information, i.e., the location information of the landing ignition point. The reference coordinate system is parallel to the landing point launch coordinate system, with the origin at the landing ignition point. , , , , , This is the deviation between position and velocity. r is the straight-line distance between the current rocket body and the target point (landing ignition point). When r < 200, r = 200. In the above formula: , , This represents the line-of-sight angular velocity component in the inertial frame. This represents the rate of change of relative distance. ; In the above formula: , , This represents the component of gravitational acceleration within the projectile system. , , This represents the component of the line-of-sight angular velocity within the projectile system. This is the transformation matrix from the landing ignition point reference coordinate system to the rocket body coordinate system. Therefore, the rocket body overload command can be obtained as follows: In the above formula: , This is an overload command. The coefficient 3 is an empirical value, which varies depending on the model, and is usually set to 3~5. The formula for calculating the command attitude angle using the overload command is as follows: In the above formula: The pitch channel commands the attitude angle. Yaw channel command attitude angle, This is the attitude for the roll channel command. And... ; The inclination angle is the ballistic tilt angle, and the coefficients 3.5 and 1.5 are gain parameters, which are selected based on experience.

[0030] Please see Figure 2 and Figure 3Specifically, the process of extracting the rotational angular velocities of the three axes around the rocket body and calculating the channel rudder deflection angles of the three channels based on the attitude angle deviation and rotational angular velocities is as follows: The pitch, yaw, and roll rotational angular velocities of the three axes around the rocket body, fed back from the onboard measurement reference, are obtained and concatenated to form a rotational angular velocity vector; a preset proportional control gain diagonal matrix and differential control gain diagonal matrix are constructed; the proportional control gain diagonal matrix and attitude angle deviation vector are subjected to feature mapping to obtain the proportional adjustment component, and the differential control gain diagonal matrix and rotational angular velocity vector are subjected to feature mapping to obtain the differential adjustment component; the proportional adjustment component and the differential adjustment component are subjected to linear state fusion to obtain a channel rudder deflection angle vector containing the pitch, yaw, and roll channel rudder deflection angles.

[0031] In this implementation scheme, the core of this step lies in introducing a proportional-derivative control law model to smoothly convert a single angle deviation signal into the deflection control command required by the physical execution channel. Controlling solely based on angle deviation can easily lead to system oscillations; therefore, the system further extracts the angular velocity information of the orbiting rocket body as a damping term. A preset proportional control gain matrix is ​​used to linearly amplify the attitude deviation to eliminate steady-state error, while a differential control gain matrix is ​​used to perform characteristic mapping on the rotational angular velocity to suppress dynamic overshoot. The fusion of these two states constitutes a complete control law. The specific linear state fusion calculation formula is as follows: In the formula, Pitch channel rudder deflection; Yaw channel rudder deflection; Roll channel rudder deflection angle; Pitch channel proportional control gain; : Yaw channel proportional control gain; Roll channel proportional control gain; Pitch channel differential control gain; : Yaw channel differential control gain; Roll channel differential control gain; Pitch and rotation angular velocity; : Yaw rotational angular velocity; : Roll rotational angular velocity. Through this fusion calculation, a stable and reliable three-dimensional channel rudder deflection angle is obtained. This example presents a two-step coefficient determination method based on feature points (divided into several discrete points according to different dynamic pressure values), where the first step provides two different implementation paths. Gain parameters are obtained using the above method at different feature points, and then used by looking up tables based on dynamic pressure and time. The specific determination method for each coefficient in the proportional-derivative control gain matrix is ​​as follows: coefficients are designed separately for the three channels "pitch," "yaw," and "roll." This example uses the "pitch" channel as an example; the design methods for the "yaw" and "roll" channels are the same as for the pitch channel. The channel rudder deflection command is filtered before being output to the servo driver. An approximate range is determined using an image method, and the constructor is... ; It is about , The generalized function, whose value can be used to evaluate the quality of the control system. This represents the deviation between the pitch angle and the pitch command angle. The optimal solution is determined using Newton's iteration method, with initial guess values ​​set. =4.0、 =1.4. Gradient: Hessian matrix: Iteration change: In the above formula: for abbreviation, For Hessian matrix, The iteration step size is the rate of change of parameters in the next iteration. Iteration begins with the above formula, converging to near-optimal parameter values. Based on modal test results, an elastic correction network is designed to ensure amplitude stability of the elastic mode shape. Parameters are fine-tuned in the frequency domain, and Bode plots are drawn based on the dynamic coefficients to observe the stability margin of the pitch channel. Parameters are fine-tuned to ensure amplitude Gm ≥ 6 dB and phase margin Pm ≥ 45 degrees. The final parameter values ​​are related to the entire system and are not universal; specific values ​​are omitted here only to describe the method. The channel rudder deflection angle calculated using the method shown in the frequency domain parameter fine-tuning is adjusted by the elastic correction network and finally fed into the servo driver to drive the grid rudder deflection. The correction network design consists of two parts: an inertial filter and a notch filter. The inertial filter is a first-order filter, with the following form: In the above formula: Let be the transfer function of a first-order inertial filter. T is the time constant, and s is the Laplace operator. The notch filter uses a high-frequency notch filter network, as follows: In the above formula: This is the transfer function for the notch filter network. Let be the center angular frequency of the i-th notch filter. The molecular damping ratio, The denominator is the damping ratio. s is the Laplace operator. The design of the correction network is related to the rocket body modes and is not a universal parameter; only the method is described here, omitting the specific values.

[0032] Specifically, based on the preset control allocation relationship between the channel rudder and the individual rudder, the control quantity allocation is performed on the channel rudder deflection angle to obtain the basic rudder deflection angle of each of the four grid rudders. The specific process is as follows: Based on the quadrant arrangement structure of the four grid rudders on the first-stage rocket body, a preset three-column four-row control quantity allocation matrix is ​​constructed; the control quantity allocation matrix and the channel rudder deflection angle vector are spatially upgraded and mapped, and the first basic rudder deflection angle, the second basic rudder deflection angle, the third basic rudder deflection angle and the fourth basic rudder deflection angle corresponding to the four grid rudders are decoupled and separated, and the basic rudder deflection angle is aggregated to construct the basic rudder deflection angle vector.

[0033] In this implementation plan, this step serves to complete the spatial mapping transformation from the theoretical control channel to the actual physical execution surface nodes. Since a first-stage rocket typically employs four grid fins arranged in a cross quadrant, and the previously calculated channel fin deflection angles only encompass three spatial dimensions, it is necessary to construct the allocation relationship based on the physical installation positions of the grid fins. This decouples and coordinates the three-dimensional control quantities to the four independent grid fin mechanisms. The control quantity allocation formula is expressed as: In the formula, First basic rudder deflection angle; Second basic rudder deflection angle; Third basic rudder deflection angle; The fourth basic rudder deflection angle. Through this matrix mapping, the system successfully transforms the abstract channel attitude adjustment requirements into basic rudder deflection angles of each grid that can be directly identified by the servo drive, laying the foundational physical data benchmark for subsequent exploration of its ultimate deflection margin.

[0034] Specifically, the process of obtaining the preset grid rudder usage limit, extracting the deflection boundary margin between the grid rudder usage limit and the basic rudder deflection angle of the four grid rudders, and combining the corresponding adjustment coefficients in the orthogonal transformation matrix to determine the remaining usable rudder deflection margin of each of the four grid rudders is as follows: Extract the absolute value of the basic rudder deflection angle of each of the four grid rudders in the basic rudder deflection angle vector, perform boundary alignment calculation between the grid rudder usage limit and the absolute value to obtain the absolute deflection boundary margin of each grid rudder; extract the sign and direction parameters of the basic rudder deflection angle of each of the four grid rudders in the basic rudder deflection angle vector; retrieve the specific adjustment coefficient column vector in the preset fourth-order orthogonal transformation matrix; perform element-wise feature multiplication fusion on the absolute deflection boundary margin, sign and direction parameters, and corresponding row elements in the adjustment coefficient column vector to calculate the remaining usable first, second, third, and fourth rudder deflection margins of each of the four grid rudders.

[0035] In this implementation plan, the core objective of this step is to accurately quantify the spatial margin of each grid rudder from the physical or control deflection limit, under the premise of fully satisfying the predetermined attitude control, i.e., to calculate how much deflection potential the system still has to exploit. Since the grid rudders are limited by the mechanical dead zone or aerodynamic heating limit of the servo mechanism in actual flight, there exists a maximum permissible deflection boundary. The system extracts the absolute value of the basic rudder deflection angle of the four grid rudders, aligns it with this usage limit, and calculates the difference. Then, combining the initial motion direction of each rudder with the transformation matrix coefficients, it calculates the deflection margin with a signed characteristic. The specific feature multiplicative fusion calculation formula is expressed as follows: In the formula, First rudder margin; Second rudder margin; Third rudder margin; Fourth rudder offset margin; to The fourth column of the fourth-order orthogonal transformation matrix contains specific adjustment coefficient column vector elements. Symbol extraction function; First basic rudder deflection angle; Second basic rudder deflection angle; Third basic rudder deflection angle; Fourth basic rudder deflection angle; Grid rudder usage limitations. This calculation clarifies the maximum additional deflection each execution node can add under unsaturated conditions. The specific rules for establishing the grid rudder usage limitation angle are as follows: mechanical limit of approximately 32°, software limit of 30°. Ground calculations before flight testing provide the maximum hinge torque under the worst operating conditions (maximum dynamic pressure, maximum rudder deflection angle). If the maximum hinge torque does not exceed structural and servo limitations, the limit angle is 30° throughout the flight. Overload calculations are not required on the rocket. If ground calculations indicate overload or insufficient margin in some operating conditions, a three-input, eight-output interpolation table is provided by the ground, showing the maximum and minimum rudder deflection angles that each grid rudder can use at a specified angle of attack, sideslip angle, and Mach number. The grid rudder usage limitation angle must simultaneously satisfy the table lookup result and the ≤30° limit. Due to differences between different models and individual units, the 30° mentioned in this example only applies to this model.

[0036] Specifically, the minimum value among the rudder deflection margins of multiple grid rudders is selected as the magnitude of the adjustment amount. The process of optimizing and assigning values ​​based on the sign direction of the base rudder deflection angle of the grid rudder that produces the minimum value is as follows: The absolute value parameters of the first, second, third, and fourth rudder deflection margins are extracted. Global boundary optimization is performed on these absolute value parameters to lock the target absolute value parameter that exhibits a minimum value. The target absolute value parameter is mapped back to the corresponding original grid rudder control node, and the sign direction parameter of the base rudder deflection angle corresponding to the original grid rudder control node is extracted. The target absolute value parameter and the extracted sign direction parameter are fused and assigned to establish and output the conversion adjustment amount with spatial directional characteristics.

[0037] In this implementation plan, this step aims to identify the global bottleneck node from the multiple deflection margins obtained in the previous step, and establish the final safe and usable conversion adjustment amount. Because the four grid rudders are strongly coupled across multiple channels in attitude control and aerodynamic deceleration tasks, to ensure that no rudder exceeds physical limits and causes servo saturation and runaway after applying deceleration deflection, the system must perform global boundary optimization to find the rudder with the smallest remaining space as the weakest link. The system extracts this minimum value and assigns it the sign and direction of its original action, transforming it into an independent control quantity with both magnitude and direction. The specific optimization and assignment logic can be expressed by the following formula: ; In the formula, : Target absolute value parameter; to First to fourth rudder offset margins; : Switch adjustment amount; Symbol extraction function; The base rudder deflection angle corresponding to the grid rudder that produces the minimum value. By establishing this adjustment amount, the system successfully extracts effective redundant information that is hidden or lost in the conventional control mapping, providing an absolutely safe numerical boundary for subsequent extreme increases in aerodynamic drag.

[0038] Specifically, the process of constructing a preset fourth-order orthogonal transformation matrix and concatenating the channel rudder deflection angle and the conversion adjustment amount to construct a four-dimensional spatial control command is as follows: Construct a fourth-order orthogonal transformation matrix whose internal element distribution satisfies that the product of itself and its transpose matrix is ​​equal to a constant multiple of the identity matrix; extract the pitch channel rudder deflection angle, yaw channel rudder deflection angle, and roll channel rudder deflection angle from the channel rudder deflection angle vector; arrange the pitch channel rudder deflection angle, yaw channel rudder deflection angle, roll channel rudder deflection angle, and conversion adjustment amount in the preset column dimension order and concatenate them to construct a four-dimensional spatial control command column vector.

[0039] In this implementation scheme, this step serves to establish a full-rank decoupling framework from abstract attitude commands to physical execution actions. Based on the original three-dimensional attitude control signals, the system actively introduces pre-calculated transformation adjustment quantities, concatenating and completing them into four-dimensional spatial control commands. Simultaneously, the system incorporates a special fourth-order orthogonal transformation matrix, where each row vector is mutually orthogonal and satisfies the property that its product with its transpose equals a constant multiple of the identity matrix. The specific spatial vector construction process is expressed as follows: ; In the formula, : Four-dimensional control instruction column vector; Pitch channel rudder deflection; Yaw channel rudder deflection; Roll channel rudder deflection angle; : Switch adjustment amount; The fourth-order orthogonal transformation matrix. This process places deceleration and attitude requirements on an equal control dimension at the data structure level, and mathematically ensures through the orthogonality of the matrix that the subsequent introduction of deceleration deflection will not cause any cross-interference with the attitude control torques of pitch, yaw, and roll. The transformation relationship from channel rudders to single-piece rudders: "X"-shaped grid rudders, each rudder forming a 45° angle with the Y and Z axes of the rocket body. During deflection, the aerodynamic force generated by the rudder surfaces has components on both the Y and Z axes of the rocket body, and these components are respectively... and The geometric relationship is 0.707 for all three channels. The roll channel is generated by differential deflection. Therefore, in the "X" configuration, each rudder contributes to all three channels simultaneously. Since the attitude control professionals incorporate this correspondence into the design of the control parameters, the allocation matrix follows the simplest expression. This is the origin of the following formula. The conversion relationship from a single-piece rudder to a channel rudder is given, and the output control allocation matrix in the conversion relationship from a channel rudder to a single-piece rudder is... According to the formula for calculating the full column rank of a pseudo-inverse matrix Easy to obtain Obviously, ; ;final, Calculation of the fourth-order orthogonal transformation matrix, given... ; Assuming the existence of an orthogonal matrix C, then the expression for C should satisfy... At the same time, according to , The following formula should exist: And exist Obviously , , And because ; ;so, ;get .

[0040] Specifically, the process of performing spatial mapping and feature decoupling on the four-dimensional space control commands through a fourth-order orthogonal transformation matrix to reconstruct the final deflection angle commands for the four grid fins, and sending the final deflection angle commands to the servo actuators to drive the corresponding grid fin deflection is as follows: A linear algebraic mapping is performed between the fourth-order orthogonal transformation matrix and the four-dimensional space control command column vector to reconstruct a four-dimensional output column vector containing the final deflection angle features of the four grid fins; feature separation is performed on the four-dimensional output column vector to extract the first final deflection command corresponding to servo node 1, the second final deflection command corresponding to servo node 2, the third final deflection command corresponding to servo node 3, and the fourth final deflection command corresponding to servo node 4; the first, second, third, and fourth final deflection commands are sent to the servo actuator nodes in the corresponding physical quadrants of the first-stage rocket to drive the controlled object to complete the physical deflection of its posture.

[0041] In this implementation scheme, this step is the final execution stage of the control method, responsible for generating the final physical instructions to drive the hardware actions. The system uses a fourth-order orthogonal transformation matrix to perform linear algebraic mapping on the assembled four-dimensional space control instructions, completely decoupling the relationships between channels and reconstructing the final swing angle instruction containing extreme deflection characteristics. The specific reconstruction mapping formula is expressed as follows: In the formula, First final rudder deflection command; Second final rudder deflection command; Third final rudder deflection command; Fourth final rudder deflection command; : Fourth-order orthogonal transformation matrix; Pitch channel rudder deflection; Yaw channel rudder deflection; Roll channel rudder deflection angle; : Conversion adjustment. Since the drag coefficient of the grid fins is directly proportional to the deflection angle in hydrodynamics, this step sends this set of margin-limit compensated commands to the servo drives in each physical quadrant. This forces the four grid fins to deflect collaboratively to the maximum physical angle within a safe range while maintaining the rocket's original descent attitude balance. This significantly increases the wind resistance characteristics, fully utilizes the deceleration potential of existing aerodynamic components, and thus significantly reduces the rocket's descent speed and saves valuable propellant required for landing retro-rockets.

[0042] In summary, this application has at least the following effects:

[0043] A novel grid fin control method for the descent phase of a rocket's first stage overcomes the limitations of conventional first-stage rockets that rely solely on induced drag under normal control by the rocket body and grid fins for deceleration during the descent phase. This method defines the effective information lost during the transition process as independent adjustment variables and combines this with a constructed orthogonal transformation matrix to reconstruct control commands. This allows the system to automatically extract and fully utilize the remaining usable fin deflection angles beyond the control constraints of each grid fin, while fully satisfying the original flight attitude control constraints. This method precisely forces the grid fins to deflect at their maximum angle, significantly increasing their drag coefficient and maximizing overall aerodynamic drag. This not only significantly enhances the aerodynamic deceleration effect during descent but also effectively conserves propellant required for landing retro-propulsion.

[0044] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] This invention is described with reference to flowchart illustrations and / or block diagrams of systems, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0046] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0047] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0048] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0049] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A grid fin control method for the descent phase of a rocket's first stage, characterized in that, Includes the following steps: S1. Obtain the pitch, yaw, and roll command attitude angles of the first-stage rocket based on the landing mission and guidance equations. Combine the actual attitude angles to obtain the attitude angle deviations and extract the rotational angular velocities around the three axes of the rocket body. Calculate the channel rudder deflection angles of the three channels based on the attitude angle deviations and rotational angular velocities. S2. Based on the preset control allocation relationship between the channel rudder and the single rudder, perform control allocation on the channel rudder deflection angle to obtain the basic rudder deflection angle of each of the four grid rudders. S3. Obtain the preset grid rudder usage limit, extract the deflection boundary margin between the grid rudder usage limit and the basic rudder deflection angle of the four grid rudders, combine the corresponding adjustment coefficient in the orthogonal transformation matrix, establish the remaining usable rudder deflection margin of each of the four grid rudders, select the minimum value among the rudder deflection margins of multiple grid rudders as the magnitude of the adjustment amount, and perform optimization assignment according to the sign direction of the basic rudder deflection angle of the grid rudder that produces the minimum value, and establish the transformation adjustment amount. S4. Construct a preset fourth-order orthogonal transformation matrix, and concatenate the channel rudder deflection angle and the transformation adjustment amount to construct a four-dimensional spatial control command. Perform spatial mapping and feature decoupling on the four-dimensional spatial control command through the fourth-order orthogonal transformation matrix to reconstruct the final rudder deflection angle command of the four grid rudders. Send the final rudder deflection angle command to the servo driver to drive the corresponding grid rudder deflection.

2. The grid fin control method for the descent phase of a rocket first stage according to claim 1, characterized in that: the specific process of obtaining the pitch, yaw, and roll three-channel command attitude angles of the first stage rocket based on the landing point mission and guidance equations, and combining the actual attitude angles to obtain the attitude angle deviation, is as follows: The guidance equations are called to generate the pitch channel command attitude angle, yaw channel command attitude angle and roll channel command attitude angle corresponding to the landing point mission, and then spliced ​​to construct a three-dimensional command attitude angle vector. The actual attitude angles of the pitch channel, yaw channel, and roll channel, which are fed back in real time by the onboard sensors, are obtained and spliced ​​together to construct a three-dimensional actual attitude angle vector. The alignment deviation between the three-dimensional commanded attitude angle vector and the three-dimensional actual attitude angle vector is calculated at the execution node, and the attitude angle deviation vector containing pitch deviation component, yaw deviation component and roll deviation component is calculated.

3. The method for grid fin control during the descent phase of a rocket first stage according to claim 2, characterized in that: the specific process of extracting the rotational angular velocities around the three axes of the rocket body and calculating the deflection angles of the three-channel fins based on the attitude angle deviation and rotational angular velocities is as follows: The pitch, yaw, and roll angular velocities around the rocket body are obtained from the onboard measurement reference feedback and then spliced ​​together to form a rotational angular velocity vector. Construct a pre-defined proportional control gain diagonal matrix and a derivative control gain diagonal matrix; The proportional control gain diagonal matrix and the attitude angle deviation vector are subjected to eigenmap to obtain the proportional adjustment component, and the differential control gain diagonal matrix and the rotational angular velocity vector are subjected to eigenmap to obtain the differential adjustment component. The proportional control component and the derivative control component are linearly fused to obtain a channel rudder deflection vector that includes the pitch channel rudder deflection, yaw channel rudder deflection, and roll channel rudder deflection.

4. The method for controlling the grid fins during the descent phase of a rocket first stage according to claim 1, characterized in that: based on the preset control allocation relationship between the channel fins and individual fins, the specific process of allocating control quantities to the channel fin deflection angle and obtaining the basic fin deflection angles of each of the four grid fins is as follows: Based on the quadrant arrangement structure of the four grid fins on the first-stage rocket body, a pre-defined three-column, four-row control quantity allocation matrix is ​​constructed. The control quantity allocation matrix and the channel rudder deflection angle vector are subjected to spatial dimension-up mapping, and the first, second, third and fourth basic rudder deflection angles corresponding to the four grid rudders are decoupled and separated, and then aggregated to construct the basic rudder deflection angle vector.

5. The method for controlling the grid fins during the descent phase of a rocket first stage according to claim 1, characterized in that: the specific process of obtaining the preset grid fin usage limit, extracting the deflection boundary margin between the grid fin usage limit and the basic fin deflection angle of the four grid fins, and determining the remaining usable fin deflection margin of each of the four grid fins by combining the corresponding adjustment coefficients in the orthogonal transformation matrix is ​​as follows: Extract the absolute values ​​of the basic deflection angles of the four grid rudders in the basic rudder deflection angle vector, and perform boundary alignment calculations between the grid rudder usage limits and absolute values ​​to obtain the absolute deflection boundary margin of each grid rudder. Extract the sign and direction parameters of the basic rudder deflection angle of each of the four grid rudders from the basic rudder deflection angle vector; Retrieve a specific adjustment coefficient column vector from a preset fourth-order orthogonal transformation matrix; The absolute deflection boundary margin, sign direction parameter, and corresponding row elements in the adjustment coefficient column vector are fused using element-wise feature multiplication to calculate the remaining usable first rudder deflection margin, second rudder deflection margin, third rudder deflection margin, and fourth rudder deflection margin for each of the four grid rudders.

6. A method for controlling the grid fins during the descent phase of a rocket first stage according to claim 5, characterized in that: the minimum value among the fin deflection margins of multiple grid fins is selected as the magnitude of the adjustment amount, and the optimal value is assigned according to the sign direction of the base fin deflection angle of the grid fin that produces the minimum value. The specific process for establishing the conversion adjustment amount is as follows: Extract the absolute value parameters of the first rudder offset margin, the second rudder offset margin, the third rudder offset margin and the fourth rudder offset margin, perform global boundary optimization in the absolute value parameters, and lock the target absolute value parameter whose value is in a minimum state. Map the target absolute value parameter back to the corresponding original grid rudder control node, and extract the sign direction parameter of the basic rudder deflection angle corresponding to the original grid rudder control node; The target absolute value parameter and the extracted symbol direction parameter are fused and assigned to establish and output the conversion adjustment amount with spatial direction characteristics.

7. The method for grid fin control during the descent phase of a rocket first stage according to claim 1, characterized in that: the specific process of constructing a preset fourth-order orthogonal transformation matrix and splicing the channel fin deflection angle and the transformation adjustment amount to construct a four-dimensional spatial control command is as follows: Construct a fourth-order orthogonal transformation matrix whose internal element distribution satisfies that the product of itself and its transpose is equal to a constant multiple of the identity matrix; Extract the pitch channel rudder deflection, yaw channel rudder deflection, and roll channel rudder deflection from the channel rudder deflection vector; The pitch channel rudder deflection angle, yaw channel rudder deflection angle, roll channel rudder deflection angle, and conversion adjustment amount are arranged sequentially according to the preset column dimension order and spliced ​​together to construct a four-dimensional spatial control command column vector.

8. A method for controlling the grid fins during the descent phase of a rocket first stage according to claim 7, characterized in that: the four-dimensional space control command is spatially mapped and decoupled using a fourth-order orthogonal transformation matrix to reconstruct the final deflection angle command for the four grid fins, and the final deflection angle command is sent to the servo driver to drive the corresponding grid fin deflection. The specific process is as follows: Perform a linear algebraic mapping between the fourth-order orthogonal transformation matrix and the four-dimensional spatial control command column vector to reconstruct and generate a four-dimensional output column vector containing the final tilt angle characteristics of the four grid rudders; Feature separation is performed on the four-dimensional output column vector to extract the first final rudder deflection command corresponding to the first servo node, the second final rudder deflection command corresponding to the second servo node, the third final rudder deflection command corresponding to the third servo node, and the fourth final rudder deflection command corresponding to the fourth servo node. The first, second, third, and fourth final rudder deflection commands are sent to the servo drive nodes in the corresponding physical quadrants of the first-stage rocket, driving the controlled object to complete the physical deflection of its position.

Images