82 results about "Space vehicle control" patented technology

The invention discloses a satellite formation autonomous control method based on relative motion fitting, belongs to the technical field of spacecraft control, and aims to solve the problem that a traditional formation control method cannot meet formation control requirements in an environment with navigation signal deficiency. The method comprises the following steps: step 1, when a main satellite passes through an ascending node, performing relative motion fitting by utilizing inter-satellite spacing and direction of a plurality of discretely distributed sub-satellites relative to the main satellite observed in a previous orbital period so as to obtain coefficients of relative motion functions of the sub-satellites; step 2, obtaining orbit parameter deviation of the child satellites relative to the primary satellite according to the coefficient of the relative motion function of the child satellites; step 3, acquiring a control vector and a control duration required by the formation adjustment of the child satellites according to the orbit parameter deviation; and step 4, each sub-satellite performs autonomous control according to the control vector and the control duration obtained in the step 3. The method is used for satellite formation autonomous control.

The invention discloses a large spacecraft lightweight control system based on knowledge distillation, which comprises a spacecraft control knowledge distillation training system and a spacecraft control knowledge graph, and realizes lightweight of a control algorithm by optimizing a knowledge distillation loss function through a dynamic knowledge migration mechanism. According to the method, a lightweight method optimization model based on knowledge distillation is adopted, a dynamic training data set is constructed, and computing resource occupation and power consumption are reduced; the knowledge association precision is enhanced and optimized through the atlas; and meanwhile, the accuracy loss is controlled by adopting attention weighted fusion cross-task knowledge, so that efficient, real-time and low-power-consumption operation and response are realized.

The invention discloses an interplanetary orbit transfer method based on a hidden state and reinforcement learning, and relates to the technical field of spacecraft control. According to the method, by establishing the sequence hidden variable model, representation learning can be explicitly carried out on the uncertain environment, and hidden information hidden under observation of the uncertain environment is extracted. And a reinforcement learning controller is formed through the actor network and the commentator network. A sequence hidden variable model and a reinforcement learning controller are adopted to construct a reinforcement learning algorithm framework, so that the training of an intelligent agent in an uncertain environment is accelerated, and the processing capability of the intelligent agent on uncertainty is improved. Besides, according to the embodiment of the invention, the next state is predicted based on the current observation and expected operation, and the quality of the predicted next state is brought into the reward structure, so that the real-time reward can capture the validity of the current and subsequent strategies at the same time, and the learning efficiency of the algorithm is further improved.

The invention discloses a six-degree-of-freedom multi-spacecraft control experiment platform and method based on semi-physics-virtuality, belongs to the technical field of spacecraft system and control system simulation, and constructs a multi-spacecraft system composed of semi-physics spacecrafts and pure virtual spacecrafts. The attitude of the semi-physical spacecraft is simulated by a simulator mounted on a three-dimensional attitude air floating table, is acquired by a sensor and is converted into a rotation matrix form, and a translation part of the semi-physical spacecraft is virtually calculated through a simulation integrator; the pose of the virtual spacecraft is virtually simulated by an SE (3) integrator with high-guarantee geometric characteristics; the system takes the ROS as a communication framework, and realizes state and instruction release by adopting a custom message; and ensuring that the time system of each spacecraft is consistent with the real time through a time synchronization mechanism. The platform supports cooperative operation of a semi-physical spacecraft and a pure virtual spacecraft, and has the characteristics of flexible structure, high simulation precision and good expandability.

The invention discloses a method for predicting the remaining service life of a control moment gyroscope based on parameter fusion, and relates to the technical field of health management of spacecraft attitude control systems. According to the method, optimal fusion of measurement parameters and implicit parameters is realized through adversarial learning feature extraction and a channel attention mechanism, and the method is particularly suitable for high-precision residual life prediction of the spacecraft control moment gyroscope under variable working conditions. The method comprises four main steps of data preparation and multi-source input, adversarial learning feature extraction, SE-Attention feature fusion and three-stage training optimization. In the data preparation step, original time sequences of voltage and current of a rotor motor are collected through a built-in electrical sensor of a CMG, and key physical parameters of a full-film lubrication friction coefficient, lubricating oil viscosity, a bearing clearance and contact stiffness are inverted based on an electromechanical coupling model; in the adversarial learning step, a game framework of a feature extractor and a working condition discriminator is constructed, and working condition-independent robust feature extraction is realized through a gradient inversion layer; in the SE-Attention feature fusion step, importance weights are adaptively distributed to different feature channels through an extrusion-excitation mechanism; and the three-stage training optimization step adopts a preheating-confrontation-fine tuning strategy to ensure the network convergence. The method solves the problem of domain adaptation, has the technical advantages of strong robustness, high precision, strong generalization, interpretability and self-adaptation, and can realize high-precision RUL prediction under variable working conditions.

The present invention provides a pure three-axis magnetic control preset performance attitude control method for an on-orbit spacecraft, comprising the following steps: Step 1: providing a spacecraft control torque calculation formula when a single magnetic torquer is used as an actuator; Step 2: providing a mathematical model of an attitude tracking system; Step 3: designing an attitude controller with preset performance to complete the attitude control with preset performance. The beneficial effects of the present invention are as follows: the pure three-axis magnetic control preset performance attitude control method for an on-orbit spacecraft of the present invention only uses a magnetic torquer as an actuator, and considering the requirements for static and dynamic performance required by the spacecraft in an actual working environment, a preset performance controller is designed to achieve the control of the dynamic steady-state performance of the attitude response of the spacecraft, thereby improving the safety and performance reliability of the magnetic control spacecraft system, and having broad application prospects in the fields of spacecraft docking and separation, on-orbit assembly, formation flying, rescue of faulty spacecraft, and de-orbiting of abandoned spacecraft.

The invention discloses a flexible constraint multi-agent formation coordination optimal control method based on hierarchical training reinforcement learning, and belongs to the field of flexible constraint multi-agent control. The method comprises the following steps: constructing a dynamic model of the flexible constraint multi-agent formation system; decomposing the control problem of the flexible constraint multi-agent formation system into a tether control layer and a node spacecraft control layer; training a tether control strategy in a tether control layer environment; freezing the trained tether control strategy, taking the tether control strategy as a part of a node spacecraft layer environment, and providing a real-time tether state; training a node spacecraft control strategy in a node spacecraft control layer environment; and deploying the training convergence tether control strategy and the node spacecraft control strategy to an actual system or a simulation environment to realize coordinated optimal control of the flexible constraint multi-agent formation. The method autonomously learns and realizes an approximate optimal cooperative control strategy coping with a strong coupling dynamics system under the multi-target constraint.

The invention discloses a space target approaching method based on multi-target optimization and a particle swarm optimization algorithm, and belongs to the technical field of spacecraft control, and the method comprises the steps: building a multi-target optimization model according to the demands of a space target approaching task; defining constraint conditions; designing a fitness function based on constraint conditions and task requirements; initializing a particle swarm: setting the scale of the particle swarm, and initializing the speed and position of particles; calculating a fitness function; updating a particle swarm: updating the speed and the position of each particle through an iteration process of a particle swarm optimization algorithm; evaluating the objective function; processing constraint conditions; iterating until a termination condition is met; and outputting an optimization result: after multiple iterations, outputting an optimal orbital transfer strategy corresponding to the global optimal position. The space target approaching method based on the multi-target optimization and the particle swarm optimization can improve the calculation efficiency, and is high in adaptability and high in precision.

A method for composite attitude control of a spacecraft with nonlinear sloshing and large flexible appendages disclosed by the present invention belongs to the technical field of spacecraft control. The implementation method of the present invention is as follows: various non-conservative disturbance torques acting on the spacecraft during on-orbit operation are characterized by Te; the attitude motion of the spacecraft body at any angle is described by Euler quaternions; the elastic vibration of 1 to m flexible appendages is described by the assumed mode discretization method; the motion description of the equivalent spherical pendulum model of liquid sloshing is improved by using the split variable technique; a large-range motion dynamics model of a liquid-filled spacecraft with large flexible appendages is established; the spacecraft attitude dynamics model is linearized; an output feedback attitude control law for a liquid-filled flexible spacecraft designed according to the Lyaponuv stability theorem; an input shaping-output feedback composite attitude controller for a liquid-filled flexible spacecraft designed based on the output feedback attitude control law of the liquid-filled flexible spacecraft and the Euler quaternion input command, thereby realizing active vibration suppression and improving the accuracy of spacecraft composite attitude control.

This invention relates to a test device and verification method for the space-ground consistency of a flexible spacecraft control system based on uncertainty quantification, belonging to the fields of intelligent manufacturing equipment industry and space-ground consistency verification technology. To address the problem of space-ground consistency verification and evaluation for flexible spacecraft, this invention includes an air-floating robot and a flexible component. The air-floating robot simulates the central rigid body of the flexible spacecraft, and the flexible component simulates the deployed solar panels. One end of the flexible component is connected to the upper support column of the air-floating robot, while the other end is free and equipped with a weight. Fiber optic grating sensors are distributed on the flexible component to measure flexible vibrations. A piezoelectric ceramic plate is installed at the end of the flexible component connected to the air-floating robot for driving and controlling the vibration of the flexible component. This invention evaluates the space-ground consistency of the designed flexible spacecraft control test system and, based on uncertainty quantification analysis, provides the impact level of uncertainty sources on space-ground consistency.

The present invention discloses an online mid-course interception method for high-speed maneuvering targets based on convex programming and capture space, belonging to the field of spacecraft control technology, including Step 1. transforming the mid-course guidance problem of the mid-course ballistic missile into an interceptor mid-course guidance model; Step 2. transforming the mid-course guidance problem into a discrete convex programming problem; Step 3. using the convex programming method to solve the offline interception trajectory of the mid-course guidance of high-speed maneuvering targets and establishing an online guidance framework; this method relaxes the affine control variables through the weighted relaxation method with terminal constraints, solving the problem of no feasible solution in dynamics due to strong equality constraints; proposes to solve the problem of errors caused by the solution of guidance commands through the online guidance framework; by using the capture space as the criterion for the success of online ballistic correction in mid-course guidance, it provides quantitative support for the success of mid-course guidance, getting rid of the narrow criterion problem of traditional zero-control interception; it can effectively correct the interceptor trajectory online and complete the online mid-course guidance interception of high-speed maneuvering targets.

The invention discloses a double-shaft solar wing control method and device giving consideration to flexibility suppression, and belongs to the technical field of spacecraft control. The method comprises the following steps: respectively calculating nominal angles of an A axis and a B axis according to a unit vector of the sun in a satellite orbit system; wherein the nominal angle of the B axis is determined according to a flexible vibration suppression interval after discretization processing; the flexible vibration suppression interval is an angle avoiding interval set for avoiding the resonant frequency; calculating control errors of the A axis and the B axis according to the nominal angles and the rotation angle measurement values of the A axis and the B axis; and according to the satellite imaging state and the state of the double-axis solar wing, the rotation of the A axis and the rotation of the B axis are controlled by using the control error so as to eliminate the sun alignment error of the double-axis solar wing. According to the invention, the resonance interval between the driving frequency and the axial vibration frequency can be avoided, and the resonance risk is reduced.

The invention discloses a control distribution method for a spacecraft with an eight-thruster structure, and belongs to the technical field of spacecraft control. The problem that after the number of thrusters of an existing spacecraft is reduced, decoupling control requirements of force / torque in all directions cannot be met is solved. Comprising the steps that eight thrusters are arranged on a spacecraft, so that the eight thrusters are distributed in an axial symmetry mode in a structure installation coordinate system, nozzles face an evasion limiting area, and the axes of the nozzles evade the centroid deviation range of a whole-satellite structure center area; for the combined control of the expected single-axis force and the expected three-axis torque, establishing a relational expression by taking the longest total ignition duration of the four thrusters as a target, and calculating the ignition duration of each thruster; and for combined control of the expected triaxial force and the expected triaxial torque, six thrusters are selected from the eight thrusters to establish a relational expression between the ignition duration and the expected triaxial force and the expected triaxial torque, and the ignition duration of the six thrusters is calculated. The method is used for ignition control of the spacecraft.

The present invention discloses a method for bidirectional association between graphical engineering and code of a spacecraft control system, which includes screening and encapsulating the source code of the simulation verification program, storing the extracted information; using the extracted information to associate with the graphical design project, identifying change items, including addition, deletion, and modification, and automatically updating the graphical design project. The present invention can, based on the existing graphical design project of the spacecraft control system, import the modified simulation verification code, scan and parse it, and can compare the new and old codes, achieving the maximum degree of automatic update of the graphical design project, avoiding repeated modification of the simulation verification program and the graphical design project, and greatly improving the iteration efficiency of the design verification process.

The invention relates to an intelligent synthesis-oriented spacecraft component equipment description method, and belongs to the technical field of architecture design. Abstract modeling and unified description are carried out on management operation of common component equipment in a spacecraft control subsystem, an application example is given, and the development efficiency and quality of spaceflight embedded software intelligent composite codes can be improved. According to the method, firstly, component equipment composition of an existing spacecraft control subsystem is combed, then general and difference demand analysis is carried out on component equipment management functions of the spacecraft control subsystem, then a software unified equipment model oriented to control system components is established, and finally a specific example is given. The method can quickly adapt to software codes formed by different parts of the control system, the reuse capability of the control software is improved, and then intelligent synthesis of the software is supported.

The present invention discloses a method for spacecraft attitude control and RCS optimal allocation, which relates to the technical field of spacecraft control. Spacecraft parameters and spacecraft attitude control torque commands are obtained; a spacecraft attitude dynamics model is constructed; a state space equation is constructed according to the spacecraft attitude dynamics model; an estimated value of the external disturbance torque acting on the spacecraft is obtained according to the state space equation; an error conversion state space equation is constructed according to the finite-time performance function and the estimated value of the external disturbance torque acting on the spacecraft; a super-twisting sliding mode controller is constructed according to the error conversion state space equation; a spacecraft attitude dynamics model after chattering suppression is constructed according to the super-twisting sliding mode controller and the saturation function; an RCS optimal allocation result is obtained according to the spacecraft attitude control torque command and the spacecraft attitude dynamics model after chattering suppression. The present invention realizes the precise tracking control of the spacecraft attitude within a finite time, and optimizes the control allocation with the minimum command error and the overall optimal energy consumption.

The present invention relates to a human-in-the-loop non-cooperative target approaching tracking control method, belonging to the field of spacecraft control technology. By adding the control input of a ground operator to the automatic control process during the approaching process of a chasing spacecraft in space, the human-in-the-loop approaching rendezvous control of the spacecraft is realized, ensuring that the chasing spacecraft can still implement a safe and reliable rendezvous approaching strategy under the condition that there are deviations in the dynamic and orbital information of the target spacecraft during the approaching process, and ensuring the safety and reliability of the approaching process. It breaks through the operation interaction method based on the image tracking of the target spacecraft and solves the problem of generating a spacecraft maneuvering strategy that integrates the active perception of the operator under the condition that the state of the target spacecraft is unknown.

The application discloses a kind of management methods of large batch simulation of spacecraft control system, including the data structure of specification simulation case, storage forms simulation case node file;Through establishing simulation case management tool, load and analyze simulation case node file, realize the management and difference comparison of simulation case node;Batch generates simulation case node file, and independently carries out simulation.The present application is based on the engineering demand of spacecraft control system scheme simulation verification, can carry out the design generation of simulation case, and the designed simulation case is easy to artificial interpretation, easy to quickly reuse and modify.Can carry out the centralized management and automatic execution of massive simulation case, with the simulation verification ability of large batch task scene of unattended, ensure the high reliability and high efficiency of spacecraft control system simulation verification.

The present invention discloses a method for automatically generating test cases for a spacecraft controller software based on a model, including: 1), formalizing and modeling the test requirements of a space vehicle, and modeling the states of the spacecraft and their transitions in the test requirements with a UML state diagram to obtain a standard model; 2), preprocessing the standard model, and converting the test information contained in the state diagram into a directed graph for storage; 3), obtaining test paths under different coverage criteria through the directed graph; 4), dividing the system input and output for each test path, and performing an equivalent class partitioning on the input space of this path. 5), performing an assignment operation on each equivalent class to obtain a sequence of specific test cases on this logical sequence. The present invention solves the disadvantage that there is no unified description method for spacecraft software requirements; it solves the disadvantage that in the current test process, the design of test scenarios completely depends on the experience of testers and cannot guarantee that the designed test scenarios effectively cover all the requirements of autonomous functions.

The application discloses a spacecraft control method and system considering residual vibration, and the method comprises the following steps: based on Euler rotation theorem, an Euler axis expression of a spacecraft attitude maneuver process is constructed; a dynamic model of a flexible spacecraft is established; a residual vibration expression of a flexible accessory in the flexible spacecraft is constructed; according to the characteristics and threshold limit of the residual vibration, a trajectory planning is carried out, and a spacecraft angular acceleration trajectory planning curve is generated; and according to the Euler axis expression of the spacecraft attitude maneuver process, the dynamic model of the flexible spacecraft and the spacecraft angular acceleration trajectory planning curve, a control loop is composed. The system comprises an Euler axis calculation module, a dynamic model construction module, a residual vibration construction module, an angular acceleration trajectory planning module and an overall control module. By using the application, the vibration generated in the maneuver process can be inhibited. The application can be widely applied in the field of spacecraft control.

This application provides an electromagnetic spacecraft control method, device, product, and electronic device, relating to the field of spacecraft control technology. This method can obtain the directed connected topology network of an electromagnetic spacecraft formation system and the dynamic equations of each electromagnetic spacecraft. Based on the dynamic equations and the directed connected topology network, a distributed control protocol is constructed. When the state update error of the distributed control protocol meets preset conditions, position control is performed on each electromagnetic spacecraft based on the distributed control protocol. This method can save communication resources while precisely controlling the electromagnetic spacecraft and can solve the actuator saturation problem involved in maintaining the formation configuration of a multi-electromagnetic spacecraft formation system. Compared with related technologies, this method considers the inherent characteristics of electromagnetic spacecraft, eliminating the need to trigger communication at every moment; communication only needs to be triggered when the state update error of the distributed control protocol meets preset conditions.

A spacecraft non-cooperative game method based on a model predictive control deep neural network belongs to the technical field of spacecraft control, and comprises the following steps: establishing a multi-spacecraft relative motion model; designing a composite interference estimator with decision variables; obtaining a non-cooperative game equilibrium solution under an offline condition; obtaining a non-cooperative game equilibrium solution under an online condition; according to the invention, through designing the composite interference estimator with the decision variable, accurate estimation of multi-source interference can be realized; a multi-spacecraft non-cooperative game problem is solved offline through model prediction control, and a large amount of data of solutions of the non-cooperative game problem is obtained; a deep neural network is designed, and the neural network is trained by using the obtained data of a large amount of solutions of the non-cooperative game problem, so that the neural network can be applied to the solution of the multi-spacecraft non-cooperative game problem online; according to the method, rapid solving of the multi-spacecraft non-cooperative game problem is achieved, and meanwhile the influence of multi-source interference on multi-spacecraft control is greatly restrained.

The application relates to a method for establishing a high-stability inertial reference of a satellite star sensor, belonging to the technical field of spacecraft control, which comprises the following steps: integrating the star sensor and a load accelerometer; using a separated two-stage light shield to suppress the stray light of the star sensor in the cabin; using a sectional heat control coating outside the two-stage light shield, and performing anti-stray-light treatment on the mounting surface of the light shield; calculating the inertial attitude of the load accelerometer by using the star sensor output, and establishing an inertial attitude model of the load accelerometer; calibrating the installation matrix of the star sensor based on the load accelerometer, and establishing a unified inertial measurement reference between the star sensor and the load accelerometer; and establishing a unified inertial measurement reference between multiple star sensors.

The invention provides a high-precision inter-satellite pointing control method fusing laser interference measurement, and relates to the field of spacecraft control. The method comprises the following steps: firstly, carrying out validity judgment and processing on an attitude error value provided by the measurement of the laser interferometer in the period, carrying out attitude angular velocity differential calculation, and when the attitude error value provided by the measurement of the laser interferometer in the period is valid, carrying out reference error calculation in combination with an attitude error value provided by orbit extrapolation and on-satellite attitude determination; then, according to attitude error value validity provided by measurement of the laser interferometer in the current period and the last period, on the basis of a reference introduction / direct introduction selection mark set on the ground, determining to implement reference introduction calculation or direct introduction calculation, calculating an attitude control quantity and an angular velocity control quantity, and finally, according to the attitude control quantity and the angular velocity control quantity, calculating the attitude control quantity and the angular velocity control quantity. An attitude control moment is calculated by using a control law and is implemented by an execution mechanism. According to the invention, high-precision pointing error measurement of the laser interferometer is introduced, so that high-precision inter-satellite pointing control of satellites is realized.

The present invention provides a rendezvous and docking planning and guidance method and device based on dynamic double anchor points, which relates to the technical field of spacecraft control. The method includes: obtaining a normal rendezvous and docking planning scheme based on a near anchor point and an emergency rendezvous and docking planning scheme based on a far anchor point; the relative distance corresponding to the far anchor point is greater than the relative distance corresponding to the near anchor point; starting the rendezvous and docking process according to the normal rendezvous and docking planning scheme, and predicting the rendezvous error at the end of the long-distance guidance section in real time; determining whether the accuracy requirement is met based on the rendezvous error predicted in real time. If so, continue the rendezvous and docking process according to the normal rendezvous and docking planning scheme. If not, interrupt the rendezvous and docking process of the normal rendezvous and docking planning scheme and switch to the emergency rendezvous and docking planning scheme for the subsequent rendezvous and docking process. The present invention can improve the rapidity and safety of rendezvous and docking.

The present invention relates to a method for acquiring fault features of aerospace control systems based on an autoencoder, comprising: preprocessing data samples of a spacecraft control system; constructing an autoencoder, and performing network training using the preprocessed data samples; optimizing the number of hidden layer nodes of the trained autoencoder, thereby completing the acquisition of fault features of aerospace control systems based on the autoencoder. The method of the present invention can fully explore the correlation between high-dimensional data, extract the implicit fault features between the outputs of multiple components, and thus achieve efficient and accurate diagnosis of faults in aerospace control systems.

This invention discloses a method and apparatus for real-time onboard determination of interference in the field of view of a star sensor, belonging to the field of spacecraft control technology. The method includes: pre-determining the obstruction rotation range of the dual-axis solar array's field of view based on its geometry, installation position, and rotation method, as well as the star sensor's field of view angle, installation position, and orientation; determining that the star sensor's field of view is obstructed by the dual-axis solar array when the rotation angles of the A-axis and B-axis of the dual-axis solar array are within the obstruction rotation range; acquiring the rotation angle measurements of the A-axis and B-axis of the dual-axis solar array in real-time on-orbit, and determining whether the star sensor's field of view is obstructed by the dual-axis solar array based on whether the rotation angle measurements are within the obstruction rotation range; and determining whether the star sensor is affected by solar interference based on the change of the solar vector in the satellite orbital system; thus obtaining the result of interference in the star sensor's field of view. This invention enables real-time onboard determination of the star sensor's field of view interference results.

The present invention relates to a satellite sun-tracking sensing and driving device, belonging to the technical field of spacecraft control, and comprising a central rotating shaft body and a first driving assembly; the central rotating shaft body is used for connecting with the shaft body of the solar wing of the spacecraft; the first driving assembly includes a first shaft body, a first elastic element and a first shape memory alloy elastic element; the first shaft body is sleeved outside the central rotating shaft body and is coaxial with the central rotating shaft body; the first shaft body is used for connecting with the shell of the spacecraft; a first contact point is arranged on the first shaft body, and a second contact point is arranged on the central rotating shaft body; a first hollow through groove is formed in a partial area of the first shaft body, and a first moving light-shielding plate is arranged inside the first shaft body; the first moving light-shielding plate is used for blocking the sunlight irradiating the first shape memory alloy elastic element through the first hollow through groove within a set time. The present invention realizes the automatic sun-tracking of the solar wing, and reduces the overall structural complexity and cost.

A kind of stop method of space vehicle target access optimization task based on local search and exploration enhancement belongs to space vehicle control technical field.To solve the problem of optimization termination control with robustness and self-adaptation, the present application designs optimization termination criterion as when the maximum prediction standard deviation of Gaussian process regression surrogate model in current local area is lower than preset threshold, terminate the iteration optimization of bayesian optimization;The design optimization termination criterion is embedded into BO;The prediction standard deviation of GPR surrogate model is used to measure the cognitive uncertainty of GPR surrogate model to objective function at a point;Set the judgment condition to judge whether the exploration enhancement ability of acquisition function is enhanced and whether local evaluation is triggered.The present application realizes the optimization termination control with robustness and self-adaptation.

The invention discloses a satellite autonomous rendezvous control method based on adaptive time sequence hidden space learning, and relates to the technical field of deep reinforcement learning and spacecraft control. According to the method, a satellite long-distance rendezvous task is modeled into a partially observable Markov decision process, and the core innovation is that the decision time step length is converted from a fixed hyper-parameter to a key decision variable which can be autonomously optimized by an intelligent agent. A continuous step size scaling factor is introduced by expanding an action space, so that end-to-end collaborative optimization of a control strategy and a decision time sequence is realized; designing a self-adaptive time sequence decision-making mechanism based on hidden state perception, and enabling an intelligent agent to dynamically adjust decision-making frequency according to a task stage; and constructing a multi-target composite reward function containing step length efficiency reward and punishment, and guiding the intelligent agent to intelligently balance the control precision and task time consumption under multiple constraints. Experiments prove that the method is obviously superior to a fixed step length baseline algorithm in task success rate, fuel economy and time efficiency, shows strong generalization ability, and provides an efficient and robust solution for intelligent autonomous control of spacecrafts.