CN115709809B

CN115709809B - An intelligent collaborative control method for autonomous orbiting and collision avoidance of spacecraft

Info

Publication number: CN115709809B
Application number: CN202211351476.0A
Authority: CN
Inventors: 袁利; 李敏; 刘磊; 张聪; 汤亮; 耿远卓; 王英杰; 李佳兴
Original assignee: Beijing Institute of Control Engineering
Current assignee: Beijing Institute of Control Engineering
Priority date: 2022-10-31
Filing date: 2022-10-31
Publication date: 2025-02-07
Anticipated expiration: 2042-10-31
Also published as: CN115709809A

Abstract

The present invention provides an intelligent collaborative control method for autonomous fly-by and collision threat avoidance of spacecraft, comprising the following steps: using a hybrid state machine to establish a hybrid model of the autonomous fly-by and collision threat avoidance control system, giving the control target corresponding to each state, and monitoring and managing the behavior state of the fly-by spacecraft based on the state machine model; designing a motion planner to give a reference motion trajectory of the fly-by spacecraft to ensure the flexibility of the conversion of each behavior state of the spacecraft; designing a reduced-order observer to estimate the unknown maneuvering parameters for the target spacecraft in view of the uncertainty of the maneuvering parameters for the controller design; designing a multi-modal motion planning and control strategy according to the control targets of different states to realize multi-modal adaptive tracking control. The present invention solves the problem that the spacecraft can effectively avoid collision threat events of space targets while realizing autonomous fly-by, thereby improving the mission execution capability and space operation safety of the spacecraft.

Description

Intelligent cooperative control method for autonomous detour and collision threat avoidance of spacecraft

Technical Field

The invention belongs to the technical field of autonomous control of a spacecraft, and particularly relates to an intelligent cooperative control method for autonomous detour and collision threat avoidance of the spacecraft.

Background

The autonomous fly-around has important roles in supporting on-orbit service and emergency treatment, on-orbit control of anti-satellite combat weapons, space target tracking and monitoring, recognition and investigation, space intersection and docking, on-orbit filling, space debris removal and other activities. For on-orbit control of the anti-satellite combat weapon, a path and a capture control point which are favorable for approaching a target are found through autonomous fly-around. For activities such as space target tracking monitoring, recognition and investigation, the omnidirectional detailed observation and monitoring of the target spacecraft are realized through autonomous round-the-fly and accurate and stable gesture pointing control, the target information is fully acquired, and further, the accurate recognition of the target characteristic information is realized, and the target type and the capability are determined. For activities such as on-orbit service and emergency treatment, space intersection butt joint, on-orbit filling, space debris removal and the like, a safe approach corridor is found through winding flying, collision between a non-cooperative target is avoided, approaching safety is guaranteed, and meanwhile, proper service and operation positions are found through winding flying to identify target characteristic information and motion states. The autonomous fly-around process is divided into a plurality of stages of approaching guidance, maintaining relative positions (fixed-point observation, photographing imaging), fly-around and the like.

However, as space debris increases year by year, on-orbit deployment of a large number of large constellations causes space orbit environments to be increasingly crowded, collision risks are increasingly increased, and an autonomous round-the-fly process is likely to face collision threat events from space debris or other spacecrafts, so that the round-the-fly process requires the spacecrafts to have collision threat autonomous evasion capability, orbit adjustment maneuver can be performed autonomously, and collision threat events possibly occurring with the space debris or other spacecrafts in the autonomous round-the-fly process are effectively treated.

The existing fly-around control method only aims at a fly-around section, has a single control mode, does not consider the problem of space collision danger avoidance, cannot adjust a behavior strategy according to the change of space conditions, such as autonomous decisions of rapid illumination conditions, increased collision risks and the like, plans a motion track, and changes the mode of a controller, so that a spacecraft can adapt to the change of the space conditions, autonomously completes corresponding control tasks and ensures the safety of the spacecraft, and therefore, the autonomous capability of a spacecraft control system needs to be increased. In order to enable the spacecraft to autonomously complete the whole around-flight process and to autonomously avoid collision risk events, and ensure task execution and spacecraft safety, the invention designs an intelligent cooperative control method for autonomous around-flight and collision risk avoidance based on a hybrid state machine method.

Disclosure of Invention

The intelligent cooperative control method for avoiding the collision threat and the autonomous winding flight of the spacecraft solves the technical problems of overcoming the defects of the prior art, and provides an intelligent cooperative control method for avoiding the collision threat and the autonomous winding flight of the spacecraft, so that the collision threat can be avoided while the task is ensured to be conducted and the safety of the spacecraft is ensured.

The technical scheme of the invention is as follows:

an intelligent cooperative control method for autonomous fly-around and collision threat avoidance of a spacecraft comprises the following steps:

(1) Setting behavior states of the spacecraft in the process of autonomous wraparound and collision threat avoidance, setting control targets in the behavior states, setting discrete events and triggering conditions for starting and ending of the discrete events in the process of autonomous wraparound and collision threat avoidance of the spacecraft, and setting a conversion function between the behavior states according to the discrete events and the triggering conditions for starting and ending of the discrete events, wherein the value of the conversion function is used for identifying whether to perform conversion between the behavior states;

(2) After receiving the autonomous winding start instruction, the winding spacecraft enters an initial behavior state, the triggering state of each discrete event is determined according to the relative motion parameters of the winding spacecraft and the target spacecraft, the requirement of the autonomous winding task and the collision threat situation, the value of a conversion function is calculated according to the triggering state of each discrete event, when the corresponding conversion function is met, the behavior state is converted, and otherwise, the current behavior state is maintained;

(3) And estimating the unknown maneuver parameters of the target spacecraft according to the current behavior state, and further determining the control parameters by combining the reference trajectory and the estimated unknown maneuver parameters to realize the control target in the current behavior state.

Preferably, the relative motion between the space-around spacecraft and other space targets is described by using a C-W equation under a reference particle orbit coordinate system, wherein the origin of the reference particle orbit coordinate system is a reference particle, the Z _o axis points to the earth center from the reference particle, the Y _o axis is perpendicular to the instantaneous orbit plane of the reference particle and points to the negative normal direction of the orbit plane, the X _o axis points to form a right-hand coordinate system with other two axes, and the relative motion between the space-around spacecraft and other space targets meets the following expression:

Wherein P represents the around-the-fly spacecraft, i represents other space targets, ρ _Pi＝[x_Pi,y_Pi,z_Pi]^T represents the relative position vector of the around-the-fly spacecraft and the other space targets; a relative velocity vector representing the velocity of the orbiting spacecraft with respect to other spatial targets; Representing relative acceleration vectors around the spacecraft and other spatial targets, u _P、u_i represents the control input of P, i, Indicating the reference particle orbit angular velocity.

Preferably, in the step (1), a behavior state and a control target in each behavior state in the autonomous fly-around and collision threat avoidance process of the spacecraft are set, specifically:

The guidance approach state q ₀ is that the around-flying spacecraft reaches the expected around-flying position of the target spacecraft through guidance, and the control target is ρ _PE＝ρ_PE,ref, wherein E represents the target spacecraft, ρ _PE represents the relative position between the around-flying spacecraft and the target spacecraft, and ρ _PE,ref represents the expected around-flying position of the around-flying spacecraft to the target spacecraft;

The relative position maintaining state q ₁ is that the relative position of the surrounding spacecraft and the target spacecraft is maintained stable, the target spacecraft is observed, and the control target is Wherein ρ _Pi,k represents the observation position,Representing a relative velocity between the orbiting spacecraft and the target spacecraft; representing a relative acceleration between the orbiting spacecraft and the target spacecraft;

the winding flight transfer state q ₂ is that the winding flight spacecraft is transferred from one observation position to another observation position, and the control targets are as follows: Wherein ρ _r,ref denotes a wraparound fly transition reference position, Representing a reference speed of the fly-around transition;

A collision threat avoidance state q ₃, wherein a space target with collision threat is avoided around the spacecraft, so that the distance between the around-the-spacecraft and the spacecraft with collision threat meets the safety distance, and the control target is [ rho _Pi||≥d_s ] (rho _Pi) represents the distance between the around-the-spacecraft and the spacecraft with collision threat, and d _s represents the safety distance between the spacecraft;

Returning to the evacuation state q ₄, after the winding task is finished, returning to a set target track by the winding device, wherein the control targets are as follows: Wherein ρ _P, Respectively representing the position, velocity and acceleration of the orbiting spacecraft in a reference particle orbit coordinate system, ρ _c,ref,Respectively representing the position of a given target track, the track speed and the track acceleration under a reference particle track coordinate system.

Preferably, in the step (1), setting a discrete event in autonomous fly around and collision threat avoidance of the spacecraft and triggering conditions for starting and ending each discrete event specifically includes:

The winding flight task e ₁：e₁ =1 represents the starting of the winding flight task, the triggering condition is task= 1;e ₁ =0 represents the ending of the winding flight task, the triggering condition is n is equal to or greater than n _ref||t≥t_t,ref |task=0, wherein n represents the number of winding flights, n _ref represents the number of expected winding flights, t represents the total time spent on winding flights, t _t,ref represents the total time expected by the winding flight task, task=1 represents the receipt of an autonomous winding flight starting instruction, and task=0 represents the receipt of an autonomous winding flight ending instruction;

the guidance approach e ₂：e₂ =1 indicates the guidance approach start, the trigger condition ρ _PE≠ρ_PE,ref;e₂ =0 indicates the guidance approach end, the trigger condition ρ _PE＝ρ_PE,ref,

The relative position hold e ₃：e₃ =1 indicates that the relative position hold starts, the trigger condition is ρ _PE＝ρ_PE,ref;e₃ =0 indicates that the relative position hold ends, the trigger condition isWherein, t _o is the observation time spent on a certain observation position, and t _o,ref is the expected observation time spent on a certain observation position;

The target observation e ₄：e₄ =1 indicates that the target observation starts, and the trigger condition is that E ₄ =0 indicates that the target observation is finished, and the triggering condition is that

Collision threat avoidance e ₅：e₅ =1 indicates that collision threat avoidance starts, the triggering condition is ||ρ _Pi||-d_c≤0&d||ρ_Pi||/dt<0;e₅ =0 indicates that collision threat avoidance ends, and the triggering condition is ||ρ _Pi||≥d_s, wherein d _c is the collision pre-warning distance between spacecrafts;

the fly-around transition e ₆：e₆ =1 indicates the start of the fly-around transition, and the trigger condition is that E ₆ =0 indicates that the winding transition is finished, and the triggering condition is thatAnd is also provided withWherein ψ=5pi/, represents an impermissible backlight angle between ρ _PE and a solar vector, N represents the number of observation interval segments divided in a round-the-fly period, s represents vector projection of the solar vector under a reference orbit coordinate system under a geocentric inertial system,S _g denotes the projection of the solar vector in the geocentric inertial coordinate system,Representing a transfer matrix of the geocentric inertial coordinate system to the reference orbit coordinate system, wherein phi represents the perigee argument,Represents the inclination angle of the track,Ρ _PE(t_k-1) represents the relative position of the last observation position around the spacecraft and the target spacecraft;

The return evacuation e ₇：e₇ =1 indicates the start of the return evacuation, the trigger condition is e ₁＝0;e₇ =0 indicates the end of the evacuation, and the trigger condition is In the case of ρ _c,ref,Respectively representing the position of a given target track, the track speed and the track acceleration under a reference track coordinate system.

Preferably, in the step (1), the step of setting a transfer function between behavior states according to trigger conditions of a discrete event and start and end thereof specifically includes:

conversion function T ₀:

δ(q₀＝1&e₁＝1&e₇＝0&e₂＝1&e₅＝0||q₀＝1&e₁＝0&e₅＝0&e₇＝1||q₂＝1&e₁＝1&e₃＝1&e₅＝0&e₆＝0&e₇＝0||q₂＝1&e₁＝0&e₅＝0&e₇＝1||q₃＝1&e₁＝1&e₅＝0&e₇＝0||q₃＝1&e₁＝0&e₅＝0&e₇＝1)＝1;

Conversion function T ₁:

δ(q₀＝1&e₅＝1||q₁＝1&e₅＝1||q₂＝1&e₅＝1||q₄＝1&e₅＝1)＝1;

Conversion function T ₂:

δ(q₁＝1&e₁＝1&e₂＝0&e₅＝0&e₆＝1&e₇＝0)＝1;

Conversion function T ₃:

δ(q₁＝1&e₁＝1&e₂＝1&e₅=0)=1;

Conversion function T ₄:

δ(q₁＝1&e₁＝0&e₅＝0&e₇=1)=1;

where δ (·) represents the coincidence function and δ represents all the variables inside the function.

Preferably, in the step (2), the behavior state transition is performed according to the behavior state and the transition function, specifically:

The initial behavior state is q ₀, if the conversion function T ₀ is satisfied, the behavior state is converted into q ₁, and if the conversion function T ₁ is satisfied, the behavior state is converted into q ₃;

When the current behavior state is q ₁, if the conversion function T ₁ is met, the behavior state is converted into q ₃, if the conversion function T ₂ is met, the behavior state is converted into q ₂, if the conversion function T ₃ is met, the behavior state is converted into q ₀, and if the conversion function T ₄ is met, the behavior state is converted into q ₄;

When the current behavior state is q ₂, if the conversion function T ₀ is satisfied, the behavior state is converted into q ₁, and if the conversion function T ₁ is satisfied, the behavior state is converted into q ₃;

When the current behavior state is q ₃, if the conversion function T ₀ is satisfied, the behavior state is converted into q ₁;

When the current behavior state is q ₄, if the transfer function T ₁ is satisfied, the behavior state is transferred to q ₃.

Preferably, in the step (3), the motion planning is performed on the surrounding spacecraft according to the relative motion parameter with the target spacecraft and the control target of the current behavior state, specifically:

When the current behavior state is q ₀ or q ₄, adopting multi-pattern programming, and the expression is as follows:

ρ_PE＝b₀+b₁t+b₂t²+b₃t³+b₄t⁴+b₅t⁵

Wherein ρ _PE (0), Respectively representing the relative position, relative speed and relative acceleration between the surrounding spacecraft and the target spacecraft at the termination time of the last behavior state q _l, ρ _PE(t_f),Respectively representing the relative position, the relative speed and the relative acceleration between the current behavior state termination moment and the target spacecraft;

when the current behavior state is q ₃, potential function planning is adopted, and the expression is as follows:

Wherein U _s represents a total potential energy function, U _ia and U _ir represent a gravitational potential energy function and a repulsive potential energy function between a surrounding spacecraft and an ith space object respectively, h represents the total number of collision threat pre-warning space objects, eta ₁ and eta ₂ represent attractive force factors, eta ₃ represents repulsive force factors, alpha and beta are constants greater than zero, rho _i1 represents a relative position vector of the ith space object, Representing the relative velocity vector of the ith spatial object.

Preferably, in the step (3), the unknown maneuver parameters of the target spacecraft are estimated according to the current behavior state, specifically:

Wherein, Representation pairIs used for the estimation of (a),Representing an estimate of f, f representing an unknown maneuver parameter of the target,Representing the error of the estimation,Σ _1k、σ_2k is a constant larger than 0, and k represents the behavior state according to the actual situation.

Preferably, in the step (3), the control parameter is determined by combining the motion planning result and the estimated unknown maneuver parameter, specifically:

when k is {0,1,2,3},

When k=4, the number of the groups,And

Where k e {0,1,2,., 4} represents the control modality, the behavior state of the corresponding spacecraft,AndAnd planning the obtained reference track for the motion under each control mode, wherein delta _k >0 is a positive constant.

Compared with the prior art, the invention has the advantages that:

(1) According to the invention, a hybrid model of the autonomous detour and collision threat avoidance control system of the spacecraft is established based on the hybrid state machine, so that the behavior change rule of the autonomous detour and collision threat process can be effectively described;

(2) The state transfer function designed by the invention realizes the monitoring and management of the behavior states of the spacecraft, and ensures the transfer coordination among the behavior states of the spacecraft;

(3) The method adapts to the change of the behavior state of the spacecraft through the multimode reduced order observer, and effectively estimates uncertain parameters in a control system;

(4) According to the invention, the reference track can be effectively tracked through a multi-mode self-adaptive control strategy, so that the control target corresponding to the behavior state is realized, and the method has stronger adaptability and robustness;

Drawings

FIG. 1 is a schematic flow diagram of an intelligent cooperative control method for autonomous fly-around and collision threat avoidance of a spacecraft;

FIG. 2 is a schematic diagram of a hybrid model of the autonomous wraparound and collision threat avoidance control system of the present invention;

FIG. 3 is a schematic diagram of a reference particle orbit coordinate system according to the present invention;

FIG. 4 is a schematic diagram of a control system for a wraparound spacecraft of the present invention.

Detailed Description

The features and advantages of the present invention will become more apparent and clear from the following detailed description of the invention.

An intelligent cooperative control method for autonomous fly-around and collision threat avoidance of a spacecraft is shown in fig. 1, and comprises the following specific implementation steps:

1) According to analysis of autonomous detour and collision threat avoidance tasks of a spacecraft, a hybrid state machine is adopted to establish a hybrid model of an autonomous detour and collision threat avoidance control system, and as shown in fig. 2, the hybrid model is mainly divided into a discrete event dynamic system model, a relative orbit dynamics model, and a mutual conversion relation between discrete events and relative orbit positions:

(1) Discrete event dynamic system model

According to the fly-around procedure and the threat avoidance procedure, an event set e= { E _i }, i=1, 2,..7, where E ₁: fly-around task, E ₂: guidance approach, E ₃: relative position hold, E ₄: target observation, E ₅: collision threat avoidance, E ₆: fly-around transfer, E ₇: return evacuation is defined. e _i =1 or e _i =0 (i=1, 2,..7) indicates the occurrence or end of an event. All events have their triggering conditions, specifically defined by the relevant design in the transfer function (3).

And dividing the autonomous fly-around process and threat avoidance process into five behavior states of guidance approaching, relative position maintaining, fly-around transferring, collision threat avoidance and return evacuation according to the event set and the process state to form a complete state machine, and establishing a finite state machine model. State machine transitions as shown in fig. 2, q= { Q _i }, i=0, 1,2,..4 represents a finite state set, where Q ₀ represents guidance proximity, Q ₁ represents relative position maintenance, Q ₂ represents fly-around transition, Q ₃ collision threat avoidance, Q ₄ represents return evacuation. T= { T _i }, i=0, 1,2,3,4 represents the transfer function between states, see (3) for specific definition, wherein T ₀ represents the transfer functions of q ₀ to q ₁、q₂ to q ₁ and q ₃ to q ₁, T ₁ represents the transfer functions of q ₀ to q ₃、q₁ to q ₃、q₂ to q ₃ and q ₄ to q ₃, T ₂ represents the transfer function of q ₁ to q ₂, T ₃ represents the transfer function of q ₁ to q ₀, and T ₄ represents the transfer function of q ₁ to q ₄.

(2) Relative orbit dynamics model

In order to describe the relative orbiting behavior of the spacecraft, a relative coordinate system is first constructed, as shown in fig. 3, wherein F _I represents a geocentric inertial coordinate system, which is defined as an origin being the earth center, an X-axis pointing to the spring point, a Z-axis pointing to the polar of the celestial sphere, and a Y-axis and the other two axes forming a right-hand coordinate system. F _o represents a reference particle orbit coordinate system, the origin is the reference particle, the Z _o axis points to the earth center from the reference particle, the Y _o axis is perpendicular to the instantaneous orbit plane and points to the negative normal direction of the orbit plane, and the reference particle is generally selected on the circular orbit near the spacecraft or on the circular orbit near the target spacecraft. The X _o axis points to form a right hand coordinate system with the other two axes.

In the reference particle orbit coordinate system, the relative motion between spacecraft P and spacecraft E relative to the reference particles and the relative motion between other spatial target pairs and the reference particles are described by using the C-W equation:

Where i= { P, E, C }, P represents spacecraft P, E represents spacecraft E, and C represents other spacecraft. ρ _i1＝[x_i,y_i,z_i]^T represents a relative position vector, ρ _i2＝[v_xi,v_yi,v_zi]^T represents a relative velocity vector, and u _i＝[u_xi,u_yi,u_zi]^T represents a control input of the spacecraft i. Wherein a ₁ and a ₂ are respectively:

In the middle of Indicating the reference particle orbit angular velocity.

Definition ρ _PE＝ρ_P1-ρ_E1 Representing the relative position and velocity between the spacecraft P and E, defining ρ _PC＝ρ_P1-ρ_C1 andRelative position and velocity between spacecraft P and other objects in space, according to the existence of

Where i= { E, C }, ρ _Pi＝[x_Pi,y_Pi,z_Pi]^T,

The spacecraft P can avoid collision threat to the spacecraft E and other target spacecrafts C while finishing autonomous winding flight to the spacecraft E. In consideration of the fact that the spacecraft P has various behavior states in the autonomous winding process, corresponding behavior state control targets are required to be established, and then the control targets are guaranteed to be achieved by designing a control strategy, so that the spacecraft P enters the corresponding behavior states.

The relative motion relation under each behavior state is analyzed, and the corresponding behavior state control targets are established as follows:

① Guidance approaching state

The guidance approaching state refers to the situation that the spacecraft P reaches the expected flying position of the spacecraft E through guidance, wherein the initial expected flying position, the expected flying position after collision threat avoidance and the like are included, namely

ρ_PE＝ρ_PE,ref

Where ρ _PE,ref is the desired fly around position for spacecraft E.

② Relative position maintaining state

The relative position maintaining state is mainly that the spacecraft P is used for state adjustment or imaging observation of the spacecraft E in a favorable position, and the relative position between the spacecraft P and the spacecraft E or other targets in the space is required to be stable in the state

Where i= { E, C }, ρ _Pi,k represents a certain fixed relative position.

③ State of transition around fly

The wraparound transition is to realize multidirectional observation of a target, and a better cognitive target is generally coplanar wraparound, namely the spacecraft P and the spacecraft E are always coplanar. In order to ensure that spacecraft P and spacecraft E are coplanar, as shown in fig. 3, assuming that the fly-around radius is r _r,ref and the fly-around plane is a plane where z=0, the following coplanar fly-around trajectory can be designed.

Solving the above method to obtain the parameters rho _r,ref and rho _r,ref of the fly-around track curveThe fly-around transfer process is actually a reference track curve tracking process, namely:

④ Collision avoidance state

In a collision avoidance state, mainly performing collision threat avoidance control, setting the safety distance between spacecrafts as d _s, the collision early warning distance between the spacecrafts as d _c, and when the distance between the two is equal to rho _Pi||≤d_c and d rho _Pi/dt is less than or equal to 0, the collision threat is present, and at the moment, the spacecrafts avoid the collision threat by changing the behavior state, so that the relative distance is satisfied, namely

||ρ_Pi||≥d_s

⑤ Returning to the evacuation state

When the winding mission is finished, the spacecraft needs to return to a set target orbit, namely:

In the case of ρ _c,ref, Respectively representing the position of a given target track, the track speed and the track acceleration under a reference track coordinate system.

(3) Establishing a conversion relationship between discrete events and relative track positions

And (3) combining the event set, the relative motion parameters and the control targets in the step (1), designing triggering conditions of the events e _i, i=1, 2, and 7, and ensuring that the spacecraft can autonomously respond to the events.

E ₁ =1 denotes the start of the winding flight task, the trigger condition is task=1, that is, a winding flight task instruction is received, e ₁ =0 denotes the end of the winding flight task, the trigger condition is n.gtoreq.n _ref||t≥t_t,ref |task=0, the target observation is completed, n denotes the number of winding flight turns, n _ref denotes the expected number of winding flight turns, t denotes the total time spent on winding flight, and t _t,ref denotes the expected total time spent on winding flight tasks.

E ₂ =1 indicates the start of the approach of guidance, ρ _PE≠ρ_PE,ref;e₂ =0 indicates the end of the approach, and ρ _PE＝ρ_PE,ref is the trigger condition.

E ₃ =1 indicates that the relative position holding starts, ρ _PE＝ρ_PE,ref;e₃ =0 indicates that the relative position holding ends, and ρ _PE＝ρ_PE,ref;e₃ =0 indicates that the relative position holding endsT _o denotes an observation period spent in a certain observation orientation, and t _o,ref denotes a desired observation period in a certain observation orientation.

E ₄ =1 indicates that the target observation starts, and the trigger condition is thatE ₄ =0 indicates that the target observation is finished, and the triggering condition is that

E ₅ =1 represents the start of collision threat avoidance, the trigger condition is ρ _Pi||-d_c≤0&d||ρ_Pi/dt <0, i= { E, O }, O represents the spacecraft that poses a collision threat to the spacecraft P, E ₅ =0 represents the end of collision threat avoidance, the relative distance between the target and the target is safe, and the trigger condition is ρ _Pi||≥d_s.

E ₆ =1 indicates the start of the wraparound transition, and the trigger condition is thatE ₆ =0 indicates that the winding transition is finished, and the triggering condition is that

And is also provided with

Wherein ψ=5pi/6, representing the impermissible backlight angle between ρ _PE and the solar vector, N being a positive integer, s representing the vector projection of the solar vector under the geocentric inertial system under the reference orbit coordinate system according to the actual situation design,S _g denotes the projection of the solar vector in the geocentric inertial coordinate system,Representing a transfer matrix of the geocentric inertial coordinate system to the orbital coordinate system, < p _PE(t_k-1 > representing the relative position of the last observed object.

E ₇ =1 indicates return to the start of evacuation, and the trigger condition is e ₁＝0;e₇ =0 indicates the end of evacuation, and the trigger condition is

According to the designed event triggering conditions, in order to coordinate the system operation, ensure the completion of the winding mission and the safety of the spacecraft, the design behavior states q _i to q _j i are not equal to j, i, j=0, 1,2, & gt, and the transfer function between 5 is as follows:

11 Q ₀ to q ₁、q₂ to q ₁ and q ₃ to q ₁ as a conversion function T ₀:

δ(q₀＝1&e₁＝1&e₇＝0&e₂＝1&e₅＝0||q₀＝1&e₁＝0&e₅＝0&e₇＝1||q₂＝1&e₁＝1&e₃＝1&e₅＝0&e₆＝0&e₇＝0||q₂＝1&e₁＝0&e₅＝0&e₇＝1||q₃＝1&e₁＝1&e₅＝0&e₇＝0||q₃＝1&e₁＝0&e₅＝0&e₇＝1)＝1

12 Q ₀ to q ₃、q₁ to q ₃、q₂ to q ₃ and q ₄ to q ₃ as a conversion function T ₁:

δ(q₀＝1&e₅＝1||q₁＝1&e₅＝1||q₂＝1&e₅＝1||q₄＝1&e₅＝1)＝1

13 Q ₁ to q ₂, the transfer function T ₂ is:

δ(q₁＝1&e₁＝1&e₂＝0&e₅＝0&e₆＝1&e₇＝0)＝1;

14 Q ₁ to q ₀, the transfer function T ₃ is:

δ(q₁＝1&e₁＝1&e₂＝1&e₅=0)=1;

15 Q ₁ to q ₄, the transfer function T ₄ is:

δ(q₁＝1&e₁＝0&e₅＝0&e₇=1)=1。

According to logic analysis, all the state transfer functions are mutually exclusive, namely, all the state transfer functions are not true at the same time, no conflict exists, the logic is self-consistent, and the coordination of the system can be ensured.

According to the descriptions of (1), (2) and (3), the hybrid model of the autonomous wraparound control system can be described by a hybrid state machine model as follows:

H=(Q,X,E,U,T,Y,f,Init,Inv)

Where q= { Q _i }, i=0, 1,2,..4 represents a finite state set, Representing continuous variables, e= { E _i }, i=1, 2,..7 represents a set of discrete events, u= { U _pi (T) } represents a continuous dynamic system segment control strategy, t= { T _i }, i=0, 1,2,..4 state transfer functions, Y, f, init, inv are a set of continuous dynamic system output variables, a continuous dynamic system differential equation, a system initial value, and a continuous variable constraint boundary, respectively.

2) According to the hybrid model established in 1), the control targets of the spacecraft P in different behavior states are different, so that the adopted control strategies are also different, and in order to reduce the influence on the dynamic performance of the control system in the control strategy switching process, motion planning is required, so that the motion curve is soft. Here, the motion planning adopts a method of combining polynomial planning and potential function planning, adopts polynomial planning in states q ₀ and q ₄, and adopts potential function planning in state q ₃

① Motion planning in the approach guidance state q ₀ and the return evacuation state q ₄

ρ_Pi＝b₀+b₁t+b₂t²+b₃t³+b₄t⁴+b₅t⁵

Where b ₀、b₁、b₂、b₃、b₄ and b ₅ are parameter vectors. The initial time constraint is the relative position, speed and acceleration information of the last state q _l at the end time, namely

The constraint condition of the termination moment is the relative position, speed and acceleration information corresponding to the control target of the current behavior state.

The end constraint for the guidance approaching state q ₀ is:

If it is

Then

If it is

Then

The end constraint for returning to evacuation state q ₄ is:

Based on the initial conditions and end constraints, given a length of time t

Wherein the method comprises the steps of

② Motion planning for collision avoidance state q ₃

The spacecraft P is required to ensure the completion of tasks and the safety of the spacecraft P in the autonomous flight around process, so that when a collision threat avoidance early warning event occurs, the spacecraft P needs to enter a threat avoidance state to carry out re-planning so as to ensure the avoidance of the collision threat and ensure the safe operation of the spacecraft. In the threat avoidance state, a potential function planning method based on a model is adopted for re-planning, a C-W equation is adopted for the model, and the potential function is constructed as follows:

Wherein U _s represents a total potential energy function, U _ia and U _ir represent an gravitational potential energy function and a repulsive potential energy function between the target and an ith target respectively, h represents the total number of collision threat early-warning space targets, eta ₁ and eta ₂ represent attractive force factors, and eta ₃ represents repulsive force factors.

In the motion planning, what planning strategy is adopted needs to be determined according to the behavior state of the spacecraft P, the motion behavior is given in the wraparound transition process q ₂ and the relative position maintaining state q ₁, and therefore the planning is not needed. After the reference motion trail is planned and given, the spacecraft P can track the given motion trail to realize the corresponding control target.

3) Uncertain parameter estimation

Before the controller is designed, taking into account that the maneuvering parameters of the spacecraft E and other space target spacecraft O are unknown, the maneuvering parameters need to be estimated and then fed back into the controller. The relative distance and relative speed between the target and the target can be obtained through a navigation system, and then on the basis, an uncertainty parameter is estimated by designing a reduced order observer, wherein the reduced order observer is in the following form:

In the middle of Representation pairIs used for the estimation of (a),Representing an estimate of f, f representing an unknown maneuver parameter of the target,Representing the estimated error, σ _2k representing the parameter to be designed,Σ _1k represents the parameters to be designed, and k represents the behavior state.

4) Adaptive tracking controller design

And designing an adaptive tracking control strategy according to the motion planning result given by the planner and the estimation of the uncertainty parameter given by the observer so as to realize control targets in different behavior states. In consideration of the fact that the control targets and the control objects are changed in different behavior states, the spacecraft controller is designed to be multi-modal so as to adapt to tracking of reference tracks in different behavior states, and the whole continuous variable closed-loop control system framework is shown in fig. 4. Assuming state q _k, k e {0,1,..4 } the reference trajectory given by the planner isAndThe multi-mode self-adaptive tracking controller designed by the invention is in the following form:

wherein delta _k >0 is a normal number, k epsilon {0,1,2,.,. 4} represents a controller mode, and corresponds to a behavior state of the spacecraft, and under the condition of k epsilon {0,1,2,3}, AndIn the case of k=4And

What is not described in detail in the present specification is a well known technology to those skilled in the art.

Claims

1. An intelligent collaborative control method for autonomous orbiting and collision threat avoidance of a spacecraft, characterized in that it comprises the following steps:

(1) Setting the behavior states of the spacecraft in the process of autonomous flyby and collision threat avoidance, the control targets in each behavior state, setting the discrete events in the process of autonomous flyby and collision threat avoidance and the trigger conditions for the start and end of each discrete event, and setting the transition function between the behavior states according to the discrete events and their start and end trigger conditions. The value of the transition function is used to identify whether to perform the transition between the behavior states;

(2) After receiving the autonomous flyby start command, the flyby spacecraft enters the initial behavior state, determines the trigger state of each discrete event according to the relative motion parameters between the flyby spacecraft and the target spacecraft, the autonomous flyby mission requirements, and the collision threat situation, and calculates the value of the conversion function according to the trigger state of each discrete event. When the corresponding conversion function is satisfied, the behavior state is converted, otherwise the current behavior state is maintained;

(3) In each behavior state, the flyby spacecraft performs motion planning based on the relative motion parameters with the target spacecraft and the control target of the current behavior state to obtain the reference trajectory of the current behavior state; then, the unknown maneuvering parameters of the target spacecraft are estimated based on the current behavior state, and the control parameters are further determined by combining the reference trajectory and the estimated unknown maneuvering parameters to achieve the control target in the current behavior state;

The CW equation in the reference particle orbit coordinate system is used to describe the relative motion between the flyby spacecraft and other space targets. The origin of the reference particle orbit coordinate system is the reference particle, the Z _o axis points from the reference particle to the center of the earth, the Y _o axis is perpendicular to the instantaneous orbit plane of the reference particle and points to the negative normal direction of the orbit plane, and the X _o axis points to form a right-handed coordinate system with the other two axes. The relative motion of the flyby spacecraft and other space targets satisfies the following expression:

Wherein, P represents the flyby spacecraft, i represents other space targets, ρ _Pi = [x _Pi , y _Pi , z _Pi ] ^T represents the relative position vector between the flyby spacecraft and other space targets; Represents the relative velocity vector between the orbiting spacecraft and other space targets; represents the relative acceleration vector between the orbiting spacecraft and other space targets, u _P and u _i represent the control inputs of P and i respectively, represents the reference particle orbital angular velocity;

In step (1), the behavior states of the spacecraft during autonomous circling and collision threat avoidance and the control targets under each behavior state are set, specifically:

Guidance approach state q ₀ : through guidance, the flyby spacecraft reaches the expected flyby position of the target spacecraft, and the control target is ρ _PE =ρ _PE,ref ; where E represents the target spacecraft, ρ _PE represents the relative position between the flyby spacecraft and the target spacecraft, and ρ _PE,ref represents the expected flyby position of the flyby spacecraft with respect to the target spacecraft;

Relative position maintenance state q ₁ : The relative position between the flyby spacecraft and the target spacecraft remains stable, the target spacecraft is observed, and the control target is ρ _PE =ρ _PE,k , Among them, ρ _Pi,k represents the observation position, represents the relative speed between the flyby spacecraft and the target spacecraft; represents the relative acceleration between the flyby spacecraft and the target spacecraft;

Flyby transfer state q ₂ : The flyby spacecraft transfers from one observation position to another observation position. The control target is: ρ _PE = ρ _{r, ref} , Where ρ _r,ref represents the fly-by transfer reference position, represents the fly-by transfer reference speed;

Collision threat avoidance state q ₃ : The flyby spacecraft avoids the space target with collision threat, so that the distance between the flyby spacecraft and the spacecraft with collision threat meets the safe distance. The control target is: ||ρ _Pi ||≥d _s ; where ρ _Pi represents the distance between the flyby spacecraft and the spacecraft with collision threat, and d _s represents the safe distance between spacecrafts;

Return to evacuation state q ₄ : After the circling mission is completed, the circling vehicle returns to the predetermined target orbit, and the control target is: ρ _P = ρ _{c, ref} , Among them, ρ _P ,

denote the position, velocity and acceleration of the flyby spacecraft in the reference particle orbit coordinate system, ρ _c,ref , They represent the predetermined target orbital position, orbital velocity and orbital acceleration in the reference particle orbital coordinate system respectively;

In the step (1), discrete events in the autonomous flyby and collision threat avoidance of the spacecraft and trigger conditions for the start and end of each discrete event are set, specifically including:

Circling task e ₁ : e ₁ =1 indicates that the circling task starts, and the trigger condition is task=1; e ₁ =0 indicates that the circling task ends, and the trigger condition is n≥n _ref ||t≥t _t,ref ||task=0, wherein n indicates the number of circling circles, n _ref indicates the expected number of circling circles, t indicates the total time spent on circling, t _t,ref indicates the expected total time of the circling task, task=1 indicates that the autonomous circling start instruction is received, and task=0 indicates that the autonomous circling end instruction is received;

Guidance approach e ₂ : e ₂ = 1 indicates that the guidance approach starts, and the trigger condition is ρ _PE ≠ρ _PE,ref ; e ₂ = 0 indicates that the guidance approach ends, and the trigger condition is ρ _PE =ρ _PE,ref ,

Relative position holding e ₃ : e ₃ = 1 indicates the start of relative position holding, and the trigger condition is ρ _PE = ρ _PE,ref ; e ₃ = 0 indicates the end of relative position holding, and the trigger condition is Among them, t _o is the observation time spent at a certain observation position, t _o,ref is the expected observation time at a certain observation position;

Target observation e ₄ : e ₄ = 1 indicates the start of target observation. The trigger condition is e ₄ = 0 indicates the end of target observation, and the trigger condition is

Collision threat avoidance e ₅ : e ₅ =1 indicates the start of collision threat avoidance, and the triggering condition is ||ρ _Pi ||-d _c ≤0&d||ρ _Pi ||/dt<0; e ₅ =0 indicates the end of collision threat avoidance, and the triggering condition is ||ρ _Pi ||≥d _s , where d _c is the collision warning distance between spacecraft;

Fly-by transfer e ₆ : e ₆ = 1 indicates the start of fly-by transfer. The triggering condition is e ₆ = 0 means the fly-by transfer is finished, and the triggering condition is and Where ψ = 5π/, which indicates the unallowed backlight angle between ρ _PE and the sunlight vector, N indicates the number of observation intervals divided in a flyby cycle, and s indicates the vector projection of the sunlight vector in the reference orbit coordinate system in the geocentric inertial system. s _g represents the projection of the sunlight vector in the geocentric inertial coordinate system, represents the transfer matrix from the geocentric inertial coordinate system to the reference orbit coordinate system, where φ represents the perigee argument, represents the orbital inclination, represents the right ascension of the ascending node; ρ _PE (t _k-1 ) represents the relative position of the flyby spacecraft and the target spacecraft at the last observation position;

Return evacuation e ₇ : e ₇ = 1 indicates the start of return evacuation, and the trigger condition is e ₁ = 0; e ₇ = 0 indicates the end of evacuation, and the trigger condition is Where ρ _c,ref , They represent the predetermined target orbital position, orbital velocity and orbital acceleration in the reference orbital coordinate system respectively;

In the step (1), the transition function between the behavior states is set according to the discrete event and the trigger conditions of its start and end, specifically including:

Transformation function T ₀ :

δ(q ₀ ＝1&e ₁ ＝1&e ₇ ＝0&e ₂ ＝1&e ₅ ＝0||q ₀ ＝1&e ₁ ＝0&e ₅ ＝0&e ₇ ＝1||

q ₂ ＝1&e ₁ ＝1&e ₃ ＝1&e ₅ ＝0&e ₆ ＝0&e ₇ ＝0||q ₂ ＝1&e ₁ ＝0&e ₅ ＝0&e ₇ ＝1||

q ₃ ＝1&e ₁ ＝1&e ₅ ＝0&e ₇ ＝0||q ₃ ＝1&e ₁ ＝0&e ₅ ＝0&e ₇ ＝1)＝1;

Transformation function T ₁ :

δ(q ₀ =1&e ₅ =1||q ₁ =1&e ₅ =1||q ₂ =1&e ₅ =1||q ₄ =1&e ₅ =1)=1;

Transformation function T ₂ :

δ(q ₁ =1&e ₁ =1&e ₂ =0&e ₅ =0&e ₆ =1&e ₇ =0)=1;

Transformation function T ₃ :

δ(q ₁ =1&e ₁ =1&e ₂ =1&e ₅ =0)=1;

Transformation function T ₄ :

δ(q ₁ =1&e ₁ =0&e ₅ =0&e ₇ =1)=1;

Among them, δ(·) represents the conforming function, · represents all the variables in the function;

In the step (2), the behavior state conversion is performed according to the behavior state and the conversion function, specifically:

The initial behavior state is q ₀ . If the conversion function T ₀ is satisfied, the behavior state is converted to q ₁ . If the conversion function T ₁ is satisfied, the behavior state is converted to q ₃ .

When the current behavior state is q ₁ , if the conversion function T ₁ is satisfied, the behavior state is converted to q ₃ ; if the conversion function T ₂ is satisfied, the behavior state is converted to q ₂ ; if the conversion function T ₃ is satisfied, the behavior state is converted to q ₀ ; if the conversion function T ₄ is satisfied, the behavior state is converted to q ₄ ;

When the current behavior state is q ₂ , if the conversion function T ₀ is satisfied, the behavior state is converted to q ₁ ; if the conversion function T ₁ is satisfied, the behavior state is converted to q ₃ ;

When the current behavior state is q ₃ , if the transition function T ₀ is satisfied, the behavior state is converted to q ₁ ;

When the current behavior state is q ₄ , if the transition function T ₁ is satisfied, the behavior state is converted to q ₃ ;

In step (3), the flyby spacecraft performs motion planning according to the relative motion parameters with the target spacecraft and the control target of the current behavior state, specifically:

When the current behavior state is q ₀ or q ₄ , multi-style planning is adopted, and the expression is as follows:

ρ _PE =b ₀ +b ₁ t+b ₂ t ² +b ₃ t ³ +b ₄ t ⁴ +b ₅ t ⁵

b ₀ =ρ _PE (0),

Among them, ρ _PE (0), They represent the relative position, relative velocity, and relative acceleration between the flyby spacecraft and the target spacecraft at the end time of the previous behavior state q _l ; ρ _PE (t _f ), They respectively represent the relative position, relative velocity, and relative acceleration between the flyby spacecraft and the target spacecraft at the end of the current behavior state;

When the current behavior state is q ₃ , potential function planning is adopted, and the expression is as follows:

Where U _s represents the total potential energy function, U _ia and U _ir represent the gravitational potential energy function and repulsive potential energy function between the orbiting spacecraft and the i-th space target, respectively, h represents the total number of collision threat warning space targets, η ₁ and η ₂ represent the attraction factors, η ₃ represents the repulsion factor, α and β are both constants greater than zero, ρ _i1 represents the relative position vector of the i-th space target, represents the relative velocity vector of the i-th space target;

In step (3), the unknown maneuvering parameters of the target spacecraft are estimated according to the current behavior state, specifically:

in, represents the estimate of ρ _PE , represents the estimation of f, f represents the unknown maneuvering parameters of the target, represents the estimation error, σ _1k and σ _2k are constants greater than 0 and are determined according to actual conditions. k represents the behavior state.

2. According to claim 1, the intelligent collaborative control method for autonomous orbiting and collision threat avoidance of a spacecraft is characterized in that, in step (3), the control parameters are determined by combining the motion planning results and the estimated unknown maneuvering parameters, specifically:

When k∈{0,1,2,3},

When k = 4, and

Among them, k∈{0,1,2,…,4} represents the control mode, corresponding to the behavior state of the spacecraft, and is the reference trajectory obtained by motion planning under each control mode, and δ _k >0 is a positive constant.