Data-based preset performance attitude control method for spatial non-cooperative target assembly

Abstract

The invention belongs to the technical field of spacecraft performance attitude control, and discloses a data-based preset performance attitude control method for a spatial non-cooperative target assembly, which comprises the steps of nonlinear attitude data linearization, preset performance function design, error conversion and sliding mode control design. The scheme does not depend on an accurate system model, control is carried out completely based on the operation data of the spacecraft, and the method adapts to uncertainty in actual operation. An index term in a traditional control method is replaced by a tangent error function, the index explosion problem is effectively avoided, and the stability and reliability of the system are improved. Even under the condition that system parameters are uncertain, quick response and stable control can still be ensured, and the method is suitable for attitude control after non-cooperative target capture.

Description

Technical Field

[0001] This invention belongs to the field of spacecraft performance and attitude control technology, specifically relating to a data-based method for pre-setting performance and attitude control of space non-cooperative target assemblies. Background Technology

[0002] On-orbit servicing and maintenance of spacecraft is a crucial future development direction in the aerospace field, with numerous research programs underway both domestically and internationally. A key target group for this technology includes non-cooperative targets such as failed spacecraft, runaway spacecraft, and space debris. Before conducting on-orbit servicing and maintenance missions, the capture of these non-cooperative targets is essential. After capture, the mass and center of mass of the servicing spacecraft undergo significant changes, leading to alterations in its attitude angles. To ensure the safety and reliability of the capture process and prevent the servicing spacecraft from being dragged by the non-cooperative target, rapid and reliable stable control of the captured assembly is necessary. However, due to the significant non-cooperative nature of the captured targets, inertial information such as the mass and torque of the captured assembly is difficult to obtain online in real time, posing a significant challenge to the design of the corresponding attitude control system. Furthermore, considering the servicing spacecraft's own fuel consumption and the complex characteristics of the space environment, obtaining precise dynamic information about the spacecraft is difficult, further complicating the robust design of the attitude control system. Therefore, achieving safe and reliable attitude control of spacecraft under unknown and uncertain inertial conditions during on-orbit servicing is of significant practical importance.

[0003] Currently, the following methods are commonly used: 1. Space robot on-orbit non-cooperative target capture control method CN118348902A: Based on the spatial position of the on-orbit non-cooperative target, a capture system model is established. The capture system model is used to describe the numerical relationship between the end-effector attitude angle and the intermediate joint rotation angle. Based on the capture system model, a motion decomposition controller is used to solve for the end-effector attitude angle. Based on the end-effector attitude angle and the intermediate joint rotation angle, the control parameters of the space robot are determined. Based on the control parameters, the space robot is controlled to capture the target.

[0004] 2. A Data-Predictive-Based Spatial Non-Cooperative Target Attitude Takeover Disturbance Resistance Control Method (CN118192655A): This method addresses the problems of poor modeling accuracy and poor immunity to large offsets and periodic disturbances in existing takeover control methods. It includes obtaining the combined body's state equations from the kinematic and dynamic equations of the combined body's attitude, followed by a discrete linearized state equation; further deformation to obtain a deformed dynamic equation incorporating unknown dynamic parameters; establishing relationships between unknown dynamic parameters using historical data, identifying these parameters, and predicting changes in disturbance torque to obtain a combined body state prediction model; updating the discrete linearized state equations; and establishing a cost function considering the predicted disturbance torque and a service satellite flywheel control torque constraint model to solve for the service satellite flywheel control torque for non-cooperative target attitude control.

[0005] 3. Spacecraft Attitude Preset Performance Control Method (CN120178913A): This method establishes a spacecraft error attitude dynamics and kinematic model considering unknown bounded input delays; constructs a preset performance function to impose preset performance constraints on the spacecraft's attitude errors; introduces an error transformation function to convert constrained attitude errors into unconstrained errors; and, based on the preset performance function, preset performance constraints, and error transformation function, performs state transformations on the spacecraft error attitude dynamics and kinematic model to obtain a spacecraft error attitude dynamics and kinematic model with respect to state variables. Disadvantages of existing technology: 1. Existing attitude control methods are mostly based on precise system models, such as sliding mode control: by constructing a sliding surface, the system's state trajectory is restricted to a set area. Although this can provide robust stability guarantees, its control performance is often limited when facing complex conditions such as unknown target characteristics and time-varying connection stiffness. The unknown inertial variables caused by mass changes will often limit its control performance.

[0006] 2. While adaptive control algorithms can automatically adjust control parameters based on changes in system parameters, they still rely to some extent on the structure of the model. They typically require prior knowledge of the system and have high computational complexity. The performance of adaptive control may be affected by rapid changes in system parameters or the presence of large external disturbances.

[0007] 3. Current attitude control schemes mostly only consider the steady-state performance of the system. The high transient performance requirements after non-cooperative target acquisition are often overlooked. In the initial stage after acquisition, the system response often exhibits significant overshoot, affecting mission reliability.

[0008] In traditional pre-designed performance functions, the exponential term in the error transformation function will manifest in the subsequently designed controller. Improper parameter design can lead to exponential explosion, which complicates subsequent theoretical proofs and experimental verification. Summary of the Invention

[0009] To overcome the above-mentioned technical problems, the present invention provides a preset performance attitude control method for a data-based space non-cooperative target assembly, which includes dynamic linearization of nonlinear spacecraft attitude control data, construction of an adjustable preset performance function, and design of a data-driven control algorithm for the spacecraft.

[0010] The present invention adopts the following technical solution: First, based on all the data from the spacecraft assembly's operational data, the determined input and output data (control torque and angular velocity) are represented using a nonlinear autoregressive moving average model to express their dynamic characteristics, and then dynamically linearized. Finally, an estimation algorithm for the pseudo-Jacobi matrix in the dynamically linearized model is derived using a criterion function.

[0011] Second, a preset performance function is constructed to impose preset performance constraints on the spacecraft's attitude angular velocity error. A tangent-type error transformation function is proposed to convert the constrained spacecraft angular velocity error into an equivalent unconstrained error.

[0012] Third, a sliding mode surface is established based on the unconstrained error, and a data-driven, pre-defined performance attitude controller is designed in conjunction with the designed reaching law, ensuring that the angular velocity error converges within a predefined region. The designed controller updates in real time and does not require consideration of unknown dynamic model accuracy or inertial variables of the combined body, nor the complexity of the space environment. Furthermore, it achieves fewer parameter adjustments, enhanced system robustness, and rapid spacecraft response after capture. This enables the spacecraft to achieve rapid and stable operation after capture by a non-cooperative target, meeting practical application requirements.

[0013] Compared with the prior art, the beneficial effects of the present invention are: The preset performance attitude control method designed in this invention, when applied to the attitude control system of a spacecraft after capturing a non-cooperative target, only utilizes the input and output data during operation; the proposed control method does not require any precise model information, including unknown inertial variables, thus avoiding the construction of complex dynamic models; In the data-driven framework, the existing preset performance schemes mainly rely on error transformation functions that contain exponential terms. Improper parameter design can lead to exponential explosion. This invention introduces a tangent error function and replaces the exponential term in the designed control with an arctangent function, effectively avoiding the related problems caused by the exponential term. Without relying on spacecraft dynamics model information, a data-driven sliding mode control algorithm with preset performance was constructed based on process data. In terms of attitude control of non-cooperative targets, it has both the robust performance of sliding mode and can constrain the spacecraft's attitude angular velocity error within the preset function envelope, the boundary of which is an adjustable boundary. Attached Figure Description

[0014] Figure 1 This is a block diagram of the spacecraft attitude and angular velocity control. Detailed Implementation

[0015] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. Unless otherwise specified, the raw materials and equipment used are commercially available or commonly used in the art. The methods in the embodiments, unless otherwise specified, are conventional methods in the art. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0016] A data-based pre-defined performance attitude control method for a spatial non-cooperative target assembly includes: I. Based on all the data from the spacecraft assembly's operational data, the dynamic characteristics of the determined input and output data (control torque and angular velocity) are represented using a nonlinear autoregressive moving average model, and then dynamically linearized. An estimation algorithm for the pseudo-Jacobi matrix in the dynamically linearized model is then derived using a criterion function.

[0017] Second, a preset performance function is constructed to impose preset performance constraints on the spacecraft's attitude angular velocity error. A tangent-type error transformation function is proposed to convert the constrained spacecraft angular velocity error into an equivalent unconstrained error.

[0018] Third, a sliding mode surface is established based on the unconstrained error. Combined with the designed reaching law, a data-driven, pre-defined performance attitude controller is designed, ensuring that the angular velocity error converges within a predefined region. The designed controller updates in real time and does not require consideration of unknown dynamic model accuracy or inertial variables of the combined object, nor the complexity of the space environment. Furthermore, it achieves fewer parameter adjustments, enhanced system robustness, and rapid spacecraft response after capture. This enables the spacecraft to achieve rapid and stable operation after capture by a non-cooperative target, meeting practical application requirements.

[0019] The specific technical solution is as follows: 1. Linearization of nonlinear attitude data in spacecraft after non-cooperative target acquisition 1.1: Obtaining the control torque in the spacecraft's operational control system and actual attitude angular velocity The data information, using a nonlinear autoregressive moving average model, represents the dynamic characteristics of the system as follows: (1) in, It concerns the control input torque in spacecraft data. and control output Nonlinear functions of data right The partial derivatives are continuous. Let ω be the angular velocity vector at the next moment. This is the angular velocity vector at the current moment. Its vector form is: , representing the angular velocities of the X, Y, and Z axes of the aerospace system, respectively. The control torque of the spacecraft at the current moment is a bounded signal. Its vector form is: , representing the torque along the X, Y, and Z axes. k represents the time interval. and These are two unknown parameters.

[0020] 1.2: The nonlinear attitude control system model of the spacecraft described above is transformed into a dynamic linear model using the compact scheme dynamic linearization method in model-free adaptive control. (2) in, , . The pseudo-Jacobi matrix, used to control the input and output differential signals, is dynamically changing. And for any time K, we have , , , , .in, 1.3: In actual spacecraft attitude control systems, without considering the system model, The value of cannot be obtained directly, therefore it is designed according to the criterion function. Prediction method: (3) in, for From the estimated value, a corresponding dynamic model for prediction can be derived. , , The purpose of setting it as a constant is to prevent the denominator from being equal to 0 in the formula.

[0021] To enhance the estimation algorithm's ability to track time-varying parameters and better adapt it to combined body control problems, a reset algorithm was added: when ,or ,but ; when ,or ,but ;, .

[0022] 2. Design of Preset Performance Functions 2.1: Based on the actual attitude angular velocity of the spacecraft and target angular velocity To define the error in angular velocity: (4) in, , This refers to the angular velocity required for the spacecraft to operate. .

[0023] 2.2: Set angular velocity error as the constraint target and design a preset performance function. : (5) (6) (7) in, Specify the initial value of the boundary. satisfy It is not difficult to see that The convergence rate is The function is a monotonically decreasing function, and the constraint objective is to keep the tracking error within a pre-specified interval. Inside.

[0024] 2.3: Converting the original tracking error into unconstrained error ,definition .

[0025] To concretize using the arctangent function ,Right now: (8) Here, It is a strictly monotonically increasing function and has the following properties: For , All satisfied and , .

[0026] Therefore, conversion error The expression can be described as: (9) 3. Modify constraint boundaries 3.1: To make the design scheme more flexible and adjustable, that is, to extend the symmetric constraint boundary (5) to an asymmetric version, the following two boundary adjustment functions are introduced. and : (10) (11) Both have the following properties , ; , .in and Represent and initial value, ; .

[0027] 3.2: The original constraint boundary (5) and error transformation function (9) are redefined as follows: (12) (13) parameter , and These are key factors in adjusting a given performance function. Parameters This represents the maximum constraint value in the initial stage. Parameter This determines the maximum allowable boundary of the steady-state error. Parameters It affects the convergence speed of the constraint boundary. The larger the value, the faster the convergence speed.

[0028] 4. Attitude Control Algorithm Design 4.1: Transformation error constructed based on the spacecraft's angular velocity error Define the sliding surface: (14) 4.2: The designed sliding mode controller consists of two parts: equivalent control and switching control, namely... First, through the reaching law The equivalent control portion is obtained: (15) in, .

[0029] 4.3: Provide the switching controller : (16) 4.4: The final master controller is represented as follows: (17) in, , The purpose is to prevent the denominator from being equal to 0.

[0030] Although embodiments of the present invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to the above embodiments without departing from the principles and spirit of the present invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A data-based method for pre-setting performance attitude control of a spatial non-cooperative target assembly, characterized in that, include: The first step is to use a nonlinear autoregressive moving average model to represent the dynamic characteristics of the determined input and output data based on all the data in the spacecraft assembly's operational data, and then to dynamically linearize the data. And an algorithm for estimating the pseudo-Jacobi matrix in the dynamic linearization model is derived using the criterion function; The second step is to construct a preset performance function and impose preset performance constraints on the spacecraft's attitude angular velocity error through the preset performance function; and to propose a tangent error transformation function to convert the constrained spacecraft angular velocity error into an equivalent unconstrained error. The third step involves establishing a sliding mode surface based on the unconstrained error and designing a data-driven, preset performance attitude controller based on the designed reaching law. This ensures that the angular velocity error converges within a predefined region. The designed controller is updated in real time and does not require consideration of the accuracy of the dynamic model of the combined body, unknown inertial variables, or complex space environment. Furthermore, it achieves fewer parameter adjustments, enhanced system robustness, and rapid response of the spacecraft after capture. This enables the spacecraft to achieve rapid and stable operation after capture by a non-cooperative target.

2. The preset performance attitude control method for a data-based spatial non-cooperative target assembly according to claim 1, characterized in that, Linearization of nonlinear attitude data in spacecraft after non-cooperative target acquisition: Step 1: Obtain the control torque in the spacecraft's operational control system and actual attitude angular velocity The data information, using a nonlinear autoregressive moving average model, represents the dynamic characteristics of the system as follows: (1) in, It concerns the control input torque in spacecraft data. and control output Nonlinear functions of data right The partial derivatives are continuous; Let ω be the angular velocity vector at the next moment. This is the angular velocity vector at the current moment; its vector form is: , representing the angular velocities of the X, Y, and Z axes of the aerospace system, respectively; The control torque of the spacecraft at the current moment is a bounded signal; its vector form is: , representing the torque along the X, Y, and Z axes; k represents the time interval. and These are two unknown parameters; Step 2: Transform the above-mentioned nonlinear attitude control system model of the spacecraft into a dynamic linear model using the compact scheme dynamic linearization method in model-free adaptive control: (2) in, , ; The pseudo-Jacobi matrix, used to control the input and output differential signals, is dynamically changing. And for any time K, we have , , , , ;in, Step 3: In the actual spacecraft attitude control system, without considering the system model, The value of cannot be obtained directly, therefore it is designed according to the criterion function. Preliminary estimation method: (3) in, for From the estimated value, a corresponding dynamic model for prediction can be derived. , , The purpose of setting it as a constant is to prevent the denominator from being equal to 0 in the formula; To enhance the estimation algorithm's ability to track time-varying parameters and better adapt it to combined body control problems, a reset algorithm was added: when ,or ,but ; when ,or ,but ;, .

3. The preset performance attitude control method for a data-based spatial non-cooperative target assembly according to claim 1, characterized in that, Preset performance function design: Step 1: Based on the actual attitude angular velocity of the spacecraft and target angular velocity To define the error in angular velocity: (4) in, , The angular velocity required for the spacecraft to operate; ; Step 2: Set angular velocity error as the constraint target and design a preset performance function. : (5) (6) (7) in, ; Specifies the initial value of the boundary satisfy It is not difficult to see that The convergence rate is The function is a monotonically decreasing function, and the constraint objective is to keep the tracking error within a pre-specified interval. Inside; Step 3: Convert the original tracking error into unconstrained error ,definition ; To concretize using the arctangent function ,Right now: (8) Here, It is a strictly monotonically increasing function and has the following properties: For , All satisfied and , ; Therefore, conversion error The expression can be described as: (9) 4. The preset performance attitude control method for a data-based spatial non-cooperative target assembly according to claim 1, characterized in that, Modify constraint boundaries: Step 1: To make the design more flexible and adjustable, that is, to extend the symmetric constraint boundary (5) to an asymmetric version; for this purpose, the following two boundary adjustment functions are introduced. and : (10) (11) Both have the following properties , ; , ;in and Represent and initial value, ; ; Step 2: The original constraint boundary (5) and error transformation function (9) are redefined as follows: (12) (13) parameter , and These are key factors in adjusting a given performance function; parameters Indicates the maximum constraint value in the initial stage; parameter The parameters determine the maximum allowable boundary of steady-state error. It affects the convergence speed of the constraint boundary. The larger the value, the faster the convergence speed.

5. The preset performance attitude control method for a data-based spatial non-cooperative target assembly according to claim 1, characterized in that, Attitude control algorithm design: Step 1: Constructing the transformation error based on the spacecraft's angular velocity error Define the sliding surface: (14) Step 2: The designed sliding mode controller consists of two parts: equivalent control and switching control, namely... First, through the law of convergence The equivalent control portion is obtained: (15) in, ; Step 3: Provide the switching controller : (16) Step 4: The final master controller is represented as follows: (17) in, , The purpose is to prevent the denominator from being equal to 0.

Images