Multi-node distributed flexible spacecraft configuration and intelligent attachment trajectory planning method

Abstract

The multi-node distributed flexible spacecraft configuration and intelligent attachment trajectory planning method disclosed by the application belong to the field of spacecraft guidance and control. The spacecraft configuration has the advantages of light weight, high energy absorption efficiency, avoidance of contact collision between rigid structure and spacecraft surface, facilitation of multi-point distributed detection, and modularization and rapid assembly, and can avoid rebound escape during spacecraft attachment. The multi-node distributed flexible spacecraft intelligent attachment trajectory planning method adopts a linear spring model to describe the flexible connection constraint force generated by the deformation of the flexible air bag; a multi-node distributed flexible spacecraft cooperative attachment trajectory planning model under multiple constraint conditions is constructed, and is converted into a standard quadratic programming form to reduce the complexity of the planning problem; an original dual neural network algorithm is adopted to construct a trajectory planning solution framework with state coupling at each control node, and the attachment trajectory of each control node is solved, and the safety and stability of the overall spacecraft attachment are improved through distributed cooperation.

Description

TECHNICAL FIELD

[0001] The present application relates to a multi-node distributed flexible spacecraft configuration, and also relates to an intelligent attachment trajectory planning method of a multi-node distributed flexible spacecraft based on the multi-node distributed flexible spacecraft configuration, and belongs to the field of spacecraft guidance and control. BACKGROUND

[0002] Asteroid exploration is a field with far-reaching significance, which not only helps to study major scientific issues such as the origin of the universe, but also helps to develop important technologies such as space resource exploitation and planetary defense. Therefore, in recent years, major spacefaring countries in the world have been actively carrying out asteroid exploration missions. Among all asteroid exploration missions, the European Space Agency's Philae comet landing mission and the asteroid sampling return mission of Japan and the United States are representative. In 2014, the Philae probe successfully docked with the 67P comet after a 10-year flight, but due to equipment failure, multiple bounces occurred during landing, and finally it was parked in the shade of a rugged valley on the comet surface, and the mission was partially successful. In June 2010, the Japanese Hayabusa-1 probe completed the world's first asteroid sampling return mission. In December 2020, the Hayabusa-2 probe successfully returned to Earth carrying samples of the Ryugu asteroid. In October 2020, the American Osiris-Rex probe successfully sampled the Bennu asteroid and will return to Earth in September 2023. Compared with flyby and orbit exploration, attached exploration directly extracts samples from the surface of an asteroid for research, which has higher scientific value. However, the attached exploration mission faces severe challenges: on the one hand, the surface topography of an asteroid is very complex, the gravitational field is weak and irregular, and there are many disturbance factors in the attachment process; on the other hand, traditional rigid probes are prone to overturning or bouncing escape during the process of attaching to an asteroid, thereby causing mission failure. The multi-node distributed flexible spacecraft utilizes a special flexible structure to dissipate impact energy during the attachment process and performs collaborative trajectory planning based on multiple control nodes, effectively improving the stability of attachment. Therefore, it is of great significance to study the multi-node distributed flexible spacecraft and the intelligent attachment trajectory planning method for asteroid exploration missions. SUMMARY

[0003] To solve the problems of traditional rigid probes in asteroid exploration missions, such as easy bouncing escape during attachment to an asteroid, low success rate, and limited exploration range of a single probe, one of the main purposes of the present application is to provide a multi-node distributed flexible spacecraft configuration for asteroid attached exploration. The spacecraft configuration has the advantages of light weight, high energy absorption efficiency, avoidance of contact collision between rigid structures and asteroid surfaces, facilitation of multi-point distributed exploration, and modularization and rapid assembly. The flexible airbag is used to absorb energy and avoid bouncing escape during spacecraft attachment.

[0004] Based on the multi-node distributed flexible spacecraft configuration for asteroid attachment, the second main purpose of the present application is to provide a multi-node distributed flexible spacecraft intelligent attachment trajectory planning method, which adopts a linear spring model to describe the flexible connection constraint force generated by the deformation of the flexible air bag, accurately reflects the flexible disturbance in the spacecraft attachment process, combines the two-body gravity model to describe the asteroid gravity acceleration suffered by the spacecraft, and constructs a multi-node distributed flexible spacecraft cooperative attachment trajectory planning model under multiple constraint conditions; the cooperative attachment trajectory planning model is converted into a standard quadratic programming form to reduce the planning problem complexity and improve the planning problem solving efficiency; the original dual neural network algorithm is adopted to construct a trajectory planning solving framework with state coupling at each control node, and the attachment trajectory of each control node is quickly and synchronously solved, that is, the multi-node distributed flexible spacecraft intelligent attachment trajectory planning is realized, and the safety and stability of the overall spacecraft attachment are improved through distributed cooperation.

[0005] The purpose of the present application is achieved by the following technical solutions.

[0006] The multi-node distributed flexible spacecraft configuration for asteroid attachment disclosed by the present application mainly consists of flexible air bag units and rigid control units. According to the requirements of actual asteroid exploration missions, n rigid control nodes and n flexible air bag units are flexibly arranged to form a spacecraft with a regular n-polygon configuration. Each two adjacent rigid control units are connected by a flexible air bag unit. The flexible air bag unit is a closed buffer air bag used for absorbing impact energy during attachment, avoiding the contact collision between the control unit and the asteroid surface, causing the probe to bounce and escape or the on-board payload to be damaged, and supporting after landing, and the upper surface layer is covered with flexible solar cell pieces for providing electric energy. The rigid control unit includes a rigid probe platform, a thruster, an optical navigation camera, a sampling device, an air bag storage cabin and other payloads. The rigid probe platform is a box board type prism structure used for carrying payloads, and the outer surface is covered with polyimide multi-layer thermal insulation material; the thruster is used to provide thrust for orbit control of the probe; the optical navigation camera is a wide field of view camera installed at the bottom of the rigid probe platform and used for navigation and close-range imaging; the sampling device is used for collecting samples on the asteroid surface after stable attachment and is installed at the bottom of the rigid probe platform; and the air bag storage cabin is located on the side of the rigid probe platform and used for storing the air bag in an un-inflated state.

[0007] The working method of the multi-node distributed flexible spacecraft configuration for asteroid attachment disclosed by the present application is as follows:

[0008] The specific working process in the asteroid-attaching mission includes four stages: an initial gathering stage, an inflation and unfolding stage, a powered descent stage, and a flexible buffering stage. In the initial gathering stage, multiple rigid control units are connected rigidly to form an aggregate, and the flexible airbag is in a folded and stored state; in the inflation and unfolding stage, the rigid control units are disconnected, the flexible airbag starts to inflate and slowly unfold, until a predetermined air pressure is reached in the airbag, forming a multi-node distributed spacecraft configuration; in the powered descent stage, according to a pre-planned cooperative attachment trajectory, the thrusters distributed on the rigid control units work cooperatively, and the spacecraft is controlled to descend from a circumnavigating orbit or a hovering state to a position close to the surface of the asteroid; in the flexible buffering stage, all the thrusters of the spacecraft are turned off, the spacecraft contacts the surface of the asteroid under the action of the gravity of the asteroid, the bottom of the flexible airbag deforms, the internal gas is compressed, and the kinetic energy of the spacecraft is converted into the internal energy of the gas in the airbag, so that stable attachment without rebound is achieved. After the attachment is completed, the sampling device carried on the spacecraft starts to work, and the samples on the surface and underground of the landing area of the probe are obtained by drilling.

[0009] The intelligent attachment trajectory planning method of the multi-node distributed flexible spacecraft disclosed in the present application is based on the multi-node distributed flexible spacecraft configuration for asteroid attachment. The intelligent attachment trajectory planning method of the multi-node distributed flexible spacecraft includes the following steps:

[0010] Step one: establish a multi-node distributed flexible spacecraft configuration connected by a flexible airbag, and simplify an equivalent control model for establishing a trajectory planning model.

[0011] Step 1.1: establish a multi-node distributed flexible spacecraft configuration connected by a flexible airbag.

[0012] Unlike traditional single-rigid-body spacecraft, multi-node distributed flexible spacecraft consist of flexible airbag units and rigid control units. Based on the specific needs of asteroid exploration missions, n rigid control nodes and n flexible airbag units are flexibly configured into a regular n-gon spacecraft. Each adjacent rigid control unit is connected by a flexible airbag unit. The flexible airbag unit is a sealed buffer airbag used to absorb impact energy during attachment, preventing direct contact and collision between the control unit and the asteroid surface, which could lead to probe bounce and escape or damage to the onboard payload. It also provides support after landing, and its upper surface is covered with flexible solar panels to provide power. The rigid control unit includes a rigid probe platform, thrusters, an optical navigation camera, a sampling device, an airbag storage compartment, and other payloads. The rigid probe platform is a box-plate prism structure used to carry the payload, and its exterior is covered with multi-layer polyimide thermal insulation material. The thruster is used to provide thrust for orbital control of the probe. The optical navigation camera is a wide-field-of-view camera, installed at the bottom of the rigid probe platform, for navigation and close-range imaging. The sampling device is used to collect samples on the asteroid surface after stable attachment, and is installed at the bottom of the rigid probe platform. The airbag storage compartment is located on the side of the rigid probe platform and is used to store the deflated airbag.

[0013] Step 1.2: A linear spring model is used to describe the flexible connection constraint force generated by the deformation of the flexible airbag, resulting in a multi-node distributed flexible spacecraft configuration equivalent control model. This spacecraft configuration equivalent control model accurately reflects the flexible disturbances during the spacecraft attachment process, facilitating the establishment of an attachment trajectory planning model in subsequent steps.

[0014] To facilitate the establishment of the attachment trajectory planning model in subsequent steps, the configuration of the multi-node distributed flexible spacecraft needs to be simplified. During the attachment process, the flexible airbag unit will deform, that is, the spacecraft configuration will change, and flexible connection constraint forces will be generated between the rigid control units. Unlike flexible films, the flexible airbag unit has sufficient thickness and stiffness, so that its deformation is confined within the configuration plane. Therefore, during the attachment process, only the compression and tension deformation of the flexible airbag unit needs to be considered. To accurately describe the flexible connection constraint forces generated during deformation, the initial configuration of the spacecraft is equivalent to an equilateral triangle, and the rigid control units are connected by linear spring components. The flexible connection constraint force between the i-th rigid control unit and the j-th rigid control unit is shown in formula (1):

[0015]

[0016] Among them, F ij Indicates the magnitude of the constraint force in the flexible connection, l ijl0 and l0 represent the actual relative distance and the initial relative distance between the two rigid control units, respectively, k represents the equivalent elastic coefficient, and n represents the number of rigid control units.

[0017] Step 2: Establish a coordinate system commonly used in asteroid attachment scenarios, derive the orbital dynamics equations of a multi-node distributed flexible spacecraft attached to an asteroid using an algebraic method, and give the component forms of the orbital dynamics equations of the attached asteroid in the asteroid fixed coordinate system.

[0018] Step 2.1: Establish the coordinate system commonly used in the asteroid attachment scenario.

[0019] To accurately describe the attachment motion of the spacecraft, the following three coordinate systems need to be established first.

[0020] (1) Heliocentric inertial coordinate system OXYZ. The origin of this coordinate system is fixed at the Sun's center of mass. The OX axis points towards the vernal equinox in the plane of the asteroid's orbit. The OZ axis is along the direction of the angular velocity of the asteroid's orbit. The OY axis is determined by the right-hand screw rule.

[0021] (2) The asteroid is fixed in a coordinate system oxyz. The origin of this coordinate system is fixed at the asteroid's center of mass, and the ox axis, oy axis, and oz axis point to the asteroid's principal axes of maximum, intermediate, and minimum inertia, respectively.

[0022] (3) Body-fixed coordinate system o bi x bi y bi z bi The origin of this coordinate system is fixed at the centroid of the i-th rigid control unit. bi x bi The axis points radially away from the overall center of mass of the spacecraft within the initial configuration plane. bi z bi The axis points to the normal direction of the initial configuration plane, o bi y bi The axis is determined by the right-hand screw rule.

[0023] Step 2.2: The algebraic method is used to derive the orbital dynamics equations of the attached asteroid to the multi-node distributed flexible spacecraft, and the component forms of the orbital dynamics equations of the attached asteroid in the fixed coordinate system of the asteroid are given.

[0024] The orbital motions of the asteroid and the i-th rigid control unit in the heliocentric inertial coordinate system are shown in Equation (2):

[0025]

[0026] Where r a The vector r represents the position of an asteroid relative to the Sun. siLet g represent the position vector of the i-th rigid control unit relative to the Sun, μ represent the solar gravitational constant, and g represent the position vector of the i-th rigid control unit relative to the Sun. ai a represents the asteroid's gravitational acceleration vector. ei u represents the acceleration vector caused by the constraint force of the flexible connection. ci This represents the control acceleration vector.

[0027] Considering that the relative distance between the rigid control unit and the asteroid is much smaller than the relative distance between the asteroid and the Sun, the following approximation is made: r si ≈r a Furthermore, the position vector of the i-th rigid control unit relative to the asteroid in the heliocentric inertial coordinate system is represented as ρ. si =r si -r a .

[0028] Consider ρ si The second derivative of is , and substituting it into formula (2), we get:

[0029]

[0030] The asteroid rotates uniformly around its principal axis of minimum inertia, oz, with a rotational angular velocity of ω. o =[0 0 ω o ] T Considering the relative derivative relationship between the heliocentric inertial coordinate system and the asteroid's fixed coordinate system, Further expressed as:

[0031]

[0032] in, This represents the position vector of the i-th rigid control unit relative to the asteroid in the asteroid's fixed coordinate system.

[0033] Substituting formula (4) into formula (3), we get:

[0034]

[0035] Expanding the above expression in component form, we get:

[0036]

[0037] To further reduce the computational load of the trajectory planning algorithm and improve its efficiency, it is preferable to treat the asteroid as a homogeneous celestial body with a shape approximately that of a regular sphere. In this case, the gravitational field of the asteroid is approximated by the two-body gravitational potential function. The gravitational acceleration of the i-th rigid control unit relative to the asteroid is shown in Equation (7).

[0038]

[0039] Where, x i ,y i ,z i Let μ represent the component coordinates of the position vector of the i-th rigid control unit in the asteroid rigid linkage. a This represents the gravitational constant of an asteroid.

[0040] Step 3: Considering the optimal performance index of fuel consumption, describe the dynamic equality constraints and boundary inequality constraints of the trajectory optimization problem, establish an asteroid attachment trajectory planning model under complex constraints, and convert it into a standard quadratic programming form to reduce the complexity of the planning problem and improve the solution efficiency of the planning problem.

[0041] Step 3.1: Use the integral of the L2 norm of the control acceleration in the time domain to describe the optimal fuel consumption performance index.

[0042] In trajectory planning problems, commonly used performance metrics include time optimization and energy optimization. Unlike low Earth orbit spacecraft, spacecraft performing asteroid exploration often carry limited fuel. Therefore, it is essential to plan an optimal fuel consumption trajectory in advance when performing asteroid attachment missions. For the i-th rigid control unit, the optimal fuel consumption performance metric is shown in Equation (8):

[0043]

[0044] Where t0 and t f These represent the task start time and end time, respectively. ci This represents the control acceleration vector of the i-th rigid control unit.

[0045] Step 3.2: Convert the asteroid attachment trajectory dynamic equations obtained in Step 2 into state equations, and then linearize and discretize them to obtain the dynamic equation constraints of the asteroid attachment trajectory planning model.

[0046] Set the task time interval [t0, t] f The sequence is discretized into N equal parts, resulting in a discrete sequence of N+1 time points [t0, t1, ..., t]. N The discretized sequence of state variables is: The discretized sequence of control variables is The asteroid attachment orbit dynamics equation described by formula (6) can be rewritten in the form of a discrete system state equation, as shown in formula (9):

[0047]

[0048] Among them, the system matrix The control matrix B is calculated by the following formula:

[0049]

[0050] In the above formula, the constant part A of the coefficient matrix const As shown below:

[0051]

[0052] asteroid gravitational component of the coefficient matrix As shown below:

[0053]

[0054] The coefficient matrix of the elastic constraint force on the i-th rigid control unit is used express:

[0055]

[0056] in,

[0057]

[0058]

[0059]

[0060]

[0061]

[0062]

[0063] A e21 =A e12

[0064] A e31 =A e13

[0065] A e23 =A e32

[0066] By the trapezoidal discretization method, t i and t i+1 The state variables and control variables at any given time have the following equation relationship:

[0067]

[0068] In the above formula, Δt represents the time step, and the formula can be further transformed into:

[0069]

[0070] in, As can be seen from the above equation, the system has N discrete dynamic equality constraint equations, which can be further standardized into the following form:

[0071] My = 0 (13)

[0072] in,

[0073]

[0074]

[0075] Step 3.3: Construct an extended-dimensional state boundary constraint vector using the upper and lower boundary values ​​of the state variables and control variables to obtain the boundary inequality constraints of the asteroid attachment trajectory planning model.

[0076] During the attachment process, it is necessary to define the range of values ​​for state variables and control variables, thus forming boundary inequality constraints as shown in formula (14):

[0077] η - ≤y≤η + (14)

[0078] in,

[0079]

[0080]

[0081] Step 3.4: Discretize and standardize the optimal fuel consumption performance index described in Step 3.1, and combine it with the dynamic equality constraints described in Step 3.2 and the boundary inequality constraints described in Step 3.3 to obtain the standard quadratic programming form of the asteroid attachment trajectory planning model.

[0082] Discretize the performance index shown in formula (8), and convert the integral into a summation, as shown in formula (15):

[0083]

[0084] Further standardization is as follows:

[0085] J i =y T Wy (16)

[0086] Wherein, the performance index weight matrix W is

[0087]

[0088] Therefore, the discrete trajectory planning problem is as follows:

[0089]

[0090] By discretization and standardization, the above trajectory optimization problem is transformed into the standard form of a quadratic programming problem, with only quadratic terms in the performance index and constraints consisting of N dynamic equality constraints and 9(N+1) boundary inequality constraints.

[0091] Step 4: Construct the original dual neural network dynamic equation based on linear variational inequalities, and build a trajectory planning solution framework with state coupling at each control node; solve the asteroid attachment trajectory planning model described in Step 3 according to the trajectory planning solution framework, and quickly and synchronously obtain the attachment trajectory of each control node, that is, realize multi-node distributed flexible spacecraft intelligent attachment trajectory planning, and improve the safety and stability of the overall attachment of the spacecraft through distributed collaboration.

[0092] For the standard quadratic programming problem shown below:

[0093]

[0094] Where W is a semi-positive definite form, the formula (19) for the primal-dual neural network dynamic equation based on the linear variational inequality required for optimization is shown:

[0095]

[0096] Where γ is the design parameter, and a larger value results in faster convergence. z represents the optimization variables of the primal and dual problems, I is the identity matrix, H is the coefficient matrix, and P... Ω Let p be the mapping function, and p be the column vector of the constraint parameter configuration in the original quadratic programming problem. The expression is as follows:

[0097]

[0098]

[0099] Where y, u, and v are the optimization variables in the original quadratic programming, corresponding to the dual variables of equality constraints and inequality constraints, respectively.

[0100] According to formula (19), an original dual neural network based on linear variational inequality is constructed, and a trajectory planning solution framework with state coupling is constructed at each control node. The asteroid attachment trajectory planning model described in step three is solved according to the trajectory planning solution framework, and the attachment trajectory of each control node is obtained quickly and synchronously, thus realizing intelligent attachment trajectory planning for multi-node distributed flexible spacecraft.

[0101] It also includes step five: based on the multi-node distributed flexible spacecraft intelligent collaborative attachment trajectory planned in step four, the thrusters distributed on each rigid control unit of the spacecraft work together to control the spacecraft to descend from an orbital or hovering state to a position close to the surface of the asteroid, thereby effectively improving the safety and stability of the overall attachment of the spacecraft.

[0102] Beneficial effects:

[0103] 1. This invention discloses a multi-node distributed flexible spacecraft configuration for asteroid attachment. The spacecraft configuration adopts a modular design approach, consisting of multiple flexible airbag units and rigid control units combined into a regular polygonal configuration. The number of units can be configured differently depending on the mission objectives, facilitating rapid assembly and production of the spacecraft. Compared with rigid structures, airbag units are lighter and simpler, providing higher energy absorption efficiency when contacting the asteroid surface, while avoiding contact collisions and bounces from the rigid structure to the asteroid surface. The multi-node distributed configuration achieves attachment control through the coordinated work of thrusters at each node, which is beneficial for multi-point distributed exploration, has strong adaptability to complex terrain on the asteroid surface, and high attachment stability.

[0104] Based on the beneficial effects of the multi-node distributed flexible spacecraft configuration for asteroid attachment disclosed in this invention, the intelligent attachment trajectory planning method for multi-node distributed flexible spacecraft disclosed in this invention has the following beneficial effects:

[0105] 2. This invention discloses a multi-node distributed flexible spacecraft intelligent attachment trajectory planning method. It employs a linear spring model to describe the flexible connection constraint force generated by the deformation of the flexible airbag, accurately reflecting the flexible disturbances during the spacecraft attachment process. It combines a two-body gravity model to describe the gravitational acceleration from the asteroid acting on the spacecraft, reducing the computational load of the trajectory planning algorithm and improving its solution efficiency. A multi-node distributed flexible spacecraft collaborative attachment trajectory planning model under multiple constraints is constructed, and this model is converted into a standard quadratic programming form, further reducing the complexity of the planning problem and improving its solution efficiency. Using a primal-dual neural network algorithm, a trajectory planning solution framework with state coupling is constructed at each control node. This method features strong global optimization capability, high solution efficiency, and fast convergence speed, thereby quickly and synchronously solving the attachment trajectory of each control node, thus realizing intelligent attachment trajectory planning for multi-node distributed flexible spacecraft. Through distributed collaboration, the overall safety and stability of the spacecraft attachment is improved. Attached Figure Description

[0106] Figure 1 This is a flowchart of a multi-node distributed flexible spacecraft intelligent attachment trajectory planning method according to the present invention;

[0107] Figure 2 This is a schematic diagram of the flexible airbag unit of the present invention, wherein: 1—airbag, 2—connecting mechanism;

[0108] Figure 3 This is a schematic diagram of the rigid control unit of the present invention, wherein: 1—rigid detector platform, 4—thruster, 5—sampling device, 6—optical navigation camera, and 7—airbag storage compartment;

[0109] Figure 4 This is a schematic diagram of a multi-node distributed flexible spacecraft configuration according to the present invention;

[0110] Figure 5 This is a schematic diagram of an equivalent control model for a multi-node distributed flexible spacecraft according to the present invention;

[0111] Figure 6 This invention describes the specific working process of a multi-node distributed flexible spacecraft attaching to an asteroid.

[0112] Figure 7 This is a schematic diagram of a commonly used coordinate system established in this invention;

[0113] Figure 8 This is a schematic diagram of the solution framework of the original dual neural network algorithm used in this invention;

[0114] Figure 9 This is the spacecraft cooperative attachment trajectory obtained in the embodiments of the present invention;

[0115] Figure 10 This is the target distance change curve obtained in the embodiments of the present invention;

[0116] Figure 11 This is the x-axis velocity curve obtained in the embodiment of the present invention;

[0117] Figure 12 This is the y-axis velocity curve obtained in an embodiment of the present invention;

[0118] Figure 13 This is the z-axis velocity curve obtained in an embodiment of the present invention;

[0119] Figure 14 This is the distance variation curve between rigid control units obtained in the embodiments of the present invention. Detailed Implementation

[0120] To better illustrate the purpose and advantages of the present invention, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0121] Figure 6This invention presents a flowchart of the multi-node distributed flexible spacecraft's asteroid attachment mission, comprising four stages: initial swarming, inflation and deployment, powered descent, and flexible buffering. The attachment trajectory planning method described in this invention solves for the attachment trajectory during the powered descent stage, effectively improving the success rate of the attachment mission by planning an optimal fuel consumption attachment trajectory that satisfies various constraints.

[0122] Example 1:

[0123] The multi-node distributed flexible spacecraft configuration and intelligent attachment trajectory planning method described in this embodiment selects the Itokawa asteroid as the attachment target, with a mass of 3.147 × 10⁻⁶. 10 kg, with a reference radius of 300 m and a rotational angular velocity of 1.4424 × 10⁻⁶. -4 rad / s.

[0124] Step 1: Establish a multi-node distributed flexible spacecraft configuration with flexible airbag connections, and simplify it to obtain an equivalent control model for establishing a trajectory planning model.

[0125] Step 1.1: Establish a multi-node distributed flexible spacecraft configuration with flexible airbag connections.

[0126] like Figure 4 The diagram shows the structural model of the multi-node distributed flexible detector described in this invention. In this embodiment, n is set to 3, meaning the spacecraft model consists of 3 rigid control units and 3 flexible airbag units, initially forming an equilateral triangle configuration. Each rigid control unit is equipped with 2 thrusters with specific installation angles, providing power during the attachment process. The flexible airbag units undergo stretching or compression during attachment, generating flexible connection constraint forces that cause changes in configuration.

[0127] Step 1.2: A linear spring model is used to describe the flexible connection constraint force generated by the deformation of the flexible airbag, resulting in a multi-node distributed flexible spacecraft configuration equivalent control model. This spacecraft configuration equivalent control model accurately reflects the flexible disturbances during the spacecraft attachment process, facilitating the establishment of an attachment trajectory planning model in subsequent steps.

[0128] like Figure 5 As shown, to effectively describe the flexible connection constraint force, the structure of the multi-node distributed flexible spacecraft is equivalent to an equilateral triangle configuration, with linear spring components connecting every two nodes. The mass of each node is set to 20 kg, and the mass of the linear spring components is negligible. The equivalent elastic force coefficient k is set to 0.001, and the initial relative distance Δl between nodes is set to 1 m.

[0129] Step 2: Establish a coordinate system commonly used in asteroid attachment scenarios, derive the orbital dynamics equations of a multi-node distributed flexible spacecraft attached to an asteroid using an algebraic method, and give the component forms of the orbital dynamics equations of the attached asteroid in the asteroid fixed coordinate system.

[0130] Step 2.1: Establish the coordinate system commonly used in the asteroid attachment scenario.

[0131] like Figure 7 As shown, the coordinate system required for the asteroid attachment trajectory planning task is established. First, a heliocentric inertial coordinate system OXYZ is established, with its origin fixed at the Sun's center of mass. The OX axis points towards the vernal equinox in the asteroid's orbital plane, the OZ axis follows the angular velocity direction of the asteroid's orbit, and the OY axis is determined by the right-hand screw rule. Next, a fixed asteroid coordinate system oxyz is established, with its origin fixed at the asteroid's center of mass. The ox, oy, and oz axes coincide with the asteroid's maximum, intermediate, and minimum moment of inertia principal axes, respectively. Finally, a fixed body coordinate system o is established. bi x bi y bi z bi The origin of this coordinate system is fixed at the centroid of the i-th rigid control unit. bi x bi The axis points radially away from the overall center of mass of the spacecraft within the initial configuration plane. bi z bi The axis points to the normal direction of the initial configuration plane, o bi y bi The axis is determined by the right-hand screw rule.

[0132] Step 2.2: The algebraic method is used to derive the orbital dynamics equations of the attached asteroid to the multi-node distributed flexible spacecraft, and the component forms of the orbital dynamics equations of the attached asteroid in the fixed coordinate system of the asteroid are given.

[0133] Step 3: Considering the optimal performance index of fuel consumption, describe the dynamic equality constraints and boundary inequality constraints of the trajectory optimization problem, establish an asteroid attachment trajectory planning model under complex constraints, and convert it into a standard quadratic programming form to reduce the complexity of the planning problem and improve the solution efficiency of the planning problem.

[0134] Step 3.1: Use the integral of the L2 norm of the control acceleration in the time domain to describe the optimal fuel consumption performance index.

[0135] Set the task initial time t0 to 0, and the task termination time to t0. f Set to 20s, and the time step Δt is set to 0.1s.

[0136] Step 3.2: Convert the asteroid attachment trajectory dynamic equations obtained in Step 2 into state equations, and then linearize and discretize them to obtain the dynamic equation constraints of the asteroid attachment trajectory planning model.

[0137] The initial state variables of the three rigid control units are set as follows:

[0138] x1(t0)=[29m -60.5m 115.86m 0m / s 0m / s 0m / s] T

[0139] x2(t0)=[29m -61m 115m 0m / s 0m / s 0m / s] T

[0140] x3(t0)=[29m -60m 115m 0m / s 0m / s 0m / s] T

[0141] Set the terminal state variables of the three rigid control units to:

[0142] x1(t f )=[31.41m -62.61m 101.26m 0m / s 0m / s 0m / s] T

[0143] x2(t f )=[31.78m -63.47m 100.83m 0m / s 0m / s 0m / s] T

[0144] x3(t f )=[32.38m -62.8m 101.21m 0m / s 0m / s 0m / s] T

[0145] Step 3.3: Construct an extended-dimensional state boundary constraint vector using the upper and lower boundary values ​​of the state variables and control variables to obtain the boundary inequality constraints of the asteroid attachment trajectory planning model.

[0146] Set the upper and lower boundary values ​​of the state variables and control variables to:

[0147] x max =[1000m 1000m 1000m 2m / s 2m / s 2m / s] T

[0148] x min =[-1000m -1000m -1000m -2m / s -2m / s -2m / s] T

[0149] u max =[1m / s 2 1m / s 2 1m / s 2 ] T

[0150] u min = [-1m / s 2 -1m / s 2 -1m / s 2 ] T

[0151] Step 3.4: Discretize and standardize the optimal fuel consumption performance index described in Step 3.1, and combine it with the dynamic equality constraints described in Step 3.2 and the boundary inequality constraints described in Step 3.3 to obtain the standard quadratic programming form of the asteroid attachment trajectory planning model.

[0152] Step 4: Construct the original dual neural network dynamic equation based on linear variational inequalities, and build a trajectory planning solution framework with state coupling at each control node; solve the asteroid attachment trajectory planning model described in Step 3 according to the trajectory planning solution framework, and quickly and synchronously obtain the attachment trajectory of each control node, that is, realize multi-node distributed flexible spacecraft intelligent attachment trajectory planning, and improve the safety and stability of the overall attachment of the spacecraft through distributed collaboration.

[0153] like Figure 8 As shown, a neural network block diagram is built in Simulink based on the dynamic equation of the original dual neural network. The parameters of the constructed standard quadratic programming model are used as input. After the iterative calculation converges, the optimal fuel consumption attachment trajectory curve and the corresponding state parameter curve that satisfy the constraints are output.

[0154] like Figure 9 As shown, the three-dimensional attachment trajectory of a multi-node distributed flexible spacecraft is intuitively displayed. The triangular facets in the figure represent the simplified spacecraft configuration. During the attachment process, the attitude and configuration of the triangular facets change, indicating that when the spacecraft moves along the planned trajectory, it will continuously adjust its attitude and orbit to achieve the optimal fuel consumption effect.

[0155] like Figure 10 As shown, the distances between each node of the spacecraft and the target position are displayed. Each node reaches the target position within the effective time, indicating that the spacecraft is stably attached to the desired position on the asteroid surface along this trajectory.

[0156] like Figures 11 to 13As shown, the velocity variation curves in the x, y, and z directions at each node of the spacecraft are displayed. The velocities in all three axes satisfy the velocity boundary constraints and decrease to zero at the terminal moment, indicating that the relative velocity of the spacecraft is almost zero at the moment of attachment to the asteroid surface, effectively ensuring a stable attachment.

[0157] like Figure 14 The figure shows the relative distance variation curves between the rigid control units of the spacecraft. The variation amplitude is always within 0.4m, indicating that the tensile or compressive deformation of the flexible airbag unit during the attachment process is within a certain limit, effectively describing the configuration change during the attachment process.

[0158] The above detailed description further illustrates the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligent attachment trajectory planning of a multi-node distributed flexible spacecraft, characterized in that: Includes the following steps, Step 1: Establish a multi-node distributed flexible spacecraft configuration with flexible airbag connections, simplifying it to obtain an equivalent control model for establishing a trajectory planning model; Step 2: Establish the coordinate system commonly used in the asteroid attachment scenario, derive the orbital dynamics equation of the multi-node distributed flexible spacecraft attached to the asteroid using the algebraic method, and give the component form of the orbital dynamics equation of the attached asteroid in the asteroid fixed coordinate system. Step 3: Considering the optimal performance index of fuel consumption, describe the dynamic equality constraints and boundary inequality constraints of the trajectory optimization problem, establish an asteroid attachment trajectory planning model under complex constraints, and convert it into a standard quadratic programming form to reduce the complexity of the planning problem and improve the solution efficiency of the planning problem. Step 4: Construct the dynamic equation of the original dual neural network based on linear variational inequalities, and build a trajectory planning solution framework with state coupling at each control node; solve the asteroid attachment trajectory planning model described in Step 3 according to the trajectory planning solution framework, and quickly and synchronously obtain the attachment trajectory of each control node, that is, realize the intelligent attachment trajectory planning of multi-node distributed flexible spacecraft, and improve the safety and stability of the overall attachment of the spacecraft through distributed collaboration. A multi-node distributed flexible spacecraft configuration for asteroid attachment, used to implement the intelligent attachment trajectory planning method of the aforementioned multi-node distributed flexible spacecraft, mainly consists of flexible airbag units and rigid control units. Based on the requirements of actual asteroid exploration missions, n rigid control nodes and n flexible airbag units are flexibly arranged into a regular n-gon spacecraft configuration. Each adjacent rigid control unit is connected by a flexible airbag unit. The flexible airbag unit is a sealed buffer airbag used to absorb impact energy during attachment, preventing direct contact and collision between the control unit and the asteroid surface, thus avoiding probe bounce and escape or damage to the onboard payload. It also provides support after landing, and its upper surface is covered with a flexible airbag. Solar cells are used to provide electrical energy; the rigid control unit includes a rigid probe platform, a thruster, an optical navigation camera, a sampling device, an airbag storage compartment, and other payloads; the rigid probe platform is a box-plate prism structure used to carry the payload, and its exterior is covered with multi-layer polyimide thermal insulation material; the thruster is used to provide thrust for orbital control of the probe; the optical navigation camera is a wide-field-of-view camera, installed at the bottom of the rigid probe platform, used for navigation and close-range imaging; the sampling device is used to collect samples on the asteroid surface after stable attachment, and is installed at the bottom of the rigid probe platform; the airbag storage compartment is located on the side of the rigid probe platform and is used to store the airbag in its uninflated state.

2. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 1, characterized in that: The specific working process in the asteroid attachment mission is divided into four stages: the initial gathering stage, the inflation and deployment stage, the powered descent stage, and the flexible buffering stage. During the initial convergence phase, multiple rigid control units are rigidly connected to form an aggregate, while the flexible airbag is in a folded and stowed state. During the inflation and deployment phase, the rigid control units disconnect, the flexible airbag begins to inflate and slowly expands until the predetermined air pressure is reached, forming a multi-node distributed spacecraft configuration. During the powered descent phase, according to the pre-planned cooperative attachment trajectory, the thrusters distributed on each rigid control unit work together, and the spacecraft descends in a controlled manner from an orbital or hovering state to a position close to the asteroid's surface. During the flexible buffer phase, all thrusters of the spacecraft are shut down, and the spacecraft contacts the asteroid's surface under the asteroid's gravity. The bottom of the flexible airbag deforms, the internal gas is compressed, and the spacecraft's kinetic energy is converted into the internal energy of the gas inside the airbag, thus achieving stable attachment without rebound. After attachment is completed, the sampling device on board the spacecraft begins to work, obtaining samples from the surface and subsurface of the probe's landing area by drilling.

3. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 1, characterized in that: It also includes step five, in which the thrusters distributed on each rigid control unit of the spacecraft work together to control the spacecraft from its orbital or hovering state to a position close to the surface of the asteroid, thereby effectively improving the safety and stability of the overall attachment of the spacecraft.

4. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 1, characterized in that: The implementation method for step one is as follows: Step 1.1: Establish a multi-node distributed flexible spacecraft configuration with flexible airbag connections; Unlike traditional single-rigid-body spacecraft, multi-node distributed flexible spacecraft consists of flexible airbag units and rigid control units. Based on the specific needs of asteroid exploration missions, n rigid control nodes and n flexible airbag units are flexibly configured into a regular n-gon spacecraft. Each adjacent rigid control unit is connected by a flexible airbag unit. The flexible airbag unit is a sealed buffer airbag used to absorb impact energy during attachment, preventing direct contact and collision between the control unit and the asteroid surface, which could lead to probe bounce and escape or damage to the onboard payload. It also provides support after landing, and its upper surface is covered with flexible solar cells to provide power. The control unit includes a rigid probe platform, a thruster, an optical navigation camera, a sampling device, an airbag storage compartment, and other payloads. The rigid probe platform is a box-plate prism structure used to carry the payloads, and its exterior is covered with multi-layer polyimide thermal insulation material. The thruster provides thrust for orbital control of the probe. The optical navigation camera is a wide-field-of-view camera, mounted on the bottom of the rigid probe platform, used for navigation and close-range imaging. The sampling device, mounted on the bottom of the rigid probe platform, is used to collect samples from the asteroid surface after stable attachment. The airbag storage compartment is located on the side of the rigid probe platform and is used to store the deflated airbags. Step 1.2: A linear spring model is used to describe the flexible connection constraint force generated by the deformation of the flexible airbag, and a multi-node distributed flexible spacecraft configuration equivalent control model is obtained. The spacecraft configuration equivalent control model accurately reflects the flexible disturbance during the spacecraft attachment process, which facilitates the establishment of the attachment trajectory planning model in subsequent steps. To facilitate the establishment of the attachment trajectory planning model in subsequent steps, the configuration of the multi-node distributed flexible spacecraft needs to be simplified. During the attachment process, the flexible airbag unit will deform, that is, the spacecraft configuration changes, and flexible connection constraint forces are generated between the rigid control units. Unlike the flexible film, the flexible airbag unit has sufficient thickness and stiffness, so that its deformation is limited to the configuration plane. Therefore, during the attachment process, only the compression and tension deformation of the flexible airbag unit needs to be considered. To accurately describe the flexible connection constraint forces generated during deformation, the initial configuration of the spacecraft is equivalent to an equilateral triangle, and the rigid control units are connected by linear spring components. The flexible connection constraint force between the i-th rigid control unit and the j-th rigid control unit is shown in formula (1). Among them, F ij Indicates the magnitude of the constraint force in the flexible connection, l ij l0 and l0 represent the actual relative distance and the initial relative distance between the two rigid control units, respectively, k represents the equivalent elastic coefficient, and n represents the number of rigid control units.

5. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 4, characterized in that: The second step is implemented as follows: Step 2.1: Establish the coordinate system commonly used in the asteroid attachment scenario; To accurately describe the attachment motion of the spacecraft, the following three coordinate systems need to be established first; (1) Heliocentric inertial coordinate system OXYZ; The origin of this coordinate system is fixed at the center of mass of the Sun. The OX axis points in the direction of the vernal equinox in the plane of the asteroid's orbit. The OZ axis is along the direction of the angular velocity of the asteroid's orbit. The OY axis is determined by the right-hand screw rule. (2) The asteroid is fixed in a coordinate system oxyz; the origin of this coordinate system is fixed at the center of mass of the asteroid, and the ox axis, oy axis and oz axis point to the principal axes of the asteroid’s maximum, middle and minimum inertia, respectively. (3) Body-fixed coordinate system o bi x bi y bi z bi The origin of this coordinate system is fixed at the centroid of the i-th rigid control unit. bi x bi The axis points radially away from the overall center of mass of the spacecraft within the initial configuration plane. bi z bi The axis points to the normal direction of the initial configuration plane, o bi y bi The axis is determined by the right-hand screw rule; Step 2.2: The algebraic method is used to derive the orbital dynamics equations of the attached asteroid to the multi-node distributed flexible spacecraft, and the component forms of the orbital dynamics equations of the attached asteroid in the asteroid fixed coordinate system are given. The orbital motions of the asteroid and the i-th rigid control unit in the heliocentric inertial coordinate system are shown in Equation (2): Where r a The vector r represents the position of an asteroid relative to the Sun. si Let g represent the position vector of the i-th rigid control unit relative to the Sun, μ represent the solar gravitational constant, and g represent the position vector of the i-th rigid control unit relative to the Sun. ai a represents the asteroid's gravitational acceleration vector. ei u represents the acceleration vector caused by the constraint force of the flexible connection. ci Indicates the control acceleration vector; Considering that the relative distance between the rigid control unit and the asteroid is much smaller than the relative distance between the asteroid and the Sun, the following approximation is made: r si ≈r a Furthermore, the position vector of the i-th rigid control unit in the heliocentric inertial coordinate system relative to the asteroid is represented as ρ. si =r si -r a ; Consider the position vector ρ si The second derivative Substituting this into formula (2), we get: The asteroid rotates uniformly around its principal axis of minimum inertia, oz, with a rotational angular velocity of ω. o =[00ω o ] T Considering the relative derivative relationship between the heliocentric inertial coordinate system and the asteroid's fixed coordinate system, Further expressed as: in, This represents the position vector of the i-th rigid control unit relative to the asteroid in the asteroid-fixed coordinate system; Substituting formula (4) into formula (3), we get: Expanding the above expression in component form, we get: Where, x i ,y i ,z i These represent the component coordinates of the position vector of the i-th rigid control unit in the asteroid rigid linkage.

6. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 5, characterized in that: If the asteroid is equivalent to a homogeneous celestial body with a shape approximately equal to a regular sphere, then the gravitational field of the asteroid is approximated by the two-body gravitational potential function. In this case, the gravitational acceleration of the i-th rigid control unit relative to the asteroid is as shown in formula (7): Where, μ a This represents the gravitational constant of an asteroid.

7. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 6, characterized in that: The method for implementing step three is as follows: Step 3.1: Use the integral of the L2 norm of the control acceleration in the time domain to describe the optimal fuel consumption performance index; In trajectory planning problems, commonly used performance indicators include time optimization and energy optimization. Unlike near-Earth orbit spacecraft, spacecraft performing asteroid exploration often carry limited fuel. Therefore, it is essential to plan an attachment trajectory with optimal fuel consumption in advance when performing asteroid attachment missions. For the i-th rigid control unit, the optimal fuel consumption performance indicator is shown in formula (8): Where t0 and t f These represent the task start time and end time, respectively. ci This represents the control acceleration vector of the i-th rigid control unit; Step 3.2: Convert the asteroid attachment trajectory dynamic equations obtained in Step 2 into state equations, and then linearize and discretize them to obtain the dynamic equation constraints of the asteroid attachment trajectory planning model; Set the task time interval [t0, t] f The sequence is discretized into N equal parts, resulting in a discrete sequence of N+1 time points [t0, t1, ..., t]. N The discretized sequence of state variables is: The discretized sequence of control variables is The asteroid attachment orbit dynamics equation described by formula (6) can be rewritten in the form of a discrete system state equation, as shown in formula (9): Among them, the system matrix The control matrix B is calculated by the following formula: In the above formula, the constant part A of the coefficient matrix const As shown below: asteroid gravitational component of the coefficient matrix As shown below: The coefficient matrix of the elastic constraint force on the i-th rigid control unit is used express: in, A e21 =A e12 A e31 =A e13 A e23 =A e32 By the trapezoidal discretization method, t i and t i+1 The state variables and control variables at any given time have the following equation relationship: In the above formula, Δt represents the time step, and the formula can be further transformed into: in, As can be seen from the above equation, the system has N discrete dynamic equality constraint equations, which can be further standardized into the following form: My = 0 (13) in, Step 3.3: Construct an extended-dimensional state boundary constraint vector using the upper and lower boundary values ​​of the state variables and control variables to obtain the boundary inequality constraints of the asteroid attachment trajectory planning model; During the attachment process, it is necessary to define the range of values ​​for state variables and control variables, thus forming boundary inequality constraints as shown in formula (14): or - ≤y≤η + (14) in, Step 3.4: Discretize and standardize the optimal fuel consumption performance index described in Step 3.1, and combine it with the dynamic equality constraints described in Step 3.2 and the boundary inequality constraints described in Step 3.3 to obtain the standard quadratic programming form of the asteroid attachment trajectory planning model; Discretize the optimal fuel consumption performance index shown in formula (8), and convert the integral into a summation, as shown in formula (15): Further standardization is as follows: J i =y T Wy (16) Wherein, the performance index weight matrix W is Therefore, the discrete trajectory planning problem is as follows: By discretization and standardization, the above trajectory optimization problem is transformed into the standard form of a quadratic programming problem, with only quadratic terms in the performance index and constraints consisting of N dynamic equality constraints and 9(N+1) boundary inequality constraints.

8. The intelligent attachment trajectory planning method for a multi-node distributed flexible spacecraft as described in claim 7, characterized in that: Step four is implemented as follows: For the standard quadratic programming problem shown below: Where W is a semi-positive definite form, the formula (19) for the primal-dual neural network dynamic equation based on the linear variational inequality required for optimization is shown: Where γ is the design parameter, and a larger value results in faster convergence; z is the optimization variable of the primal and dual problems, I is the identity matrix, H is the coefficient matrix, and P is the coefficient matrix. Ω Let p be the mapping function, and p be the column vector of the constraint parameter configuration in the original quadratic programming problem. The expression is as follows: Where y, u, and v are the optimization variables in the original quadratic programming, corresponding to the dual variables of equality constraints and inequality constraints, respectively. According to formula (19), an original dual neural network based on linear variational inequality is constructed, and a trajectory planning solution framework with state coupling is constructed at each control node. The asteroid attachment trajectory planning model described in step three is solved according to the trajectory planning solution framework, and the attachment trajectory of each control node is obtained quickly and synchronously, thus realizing intelligent attachment trajectory planning for multi-node distributed flexible spacecraft.

Images