CN115755598A - Intelligent spacecraft cluster distributed model prediction path planning method - Google Patents
Intelligent spacecraft cluster distributed model prediction path planning method Download PDFInfo
- Publication number
- CN115755598A CN115755598A CN202211352289.4A CN202211352289A CN115755598A CN 115755598 A CN115755598 A CN 115755598A CN 202211352289 A CN202211352289 A CN 202211352289A CN 115755598 A CN115755598 A CN 115755598A
- Authority
- CN
- China
- Prior art keywords
- spacecraft
- model
- control
- prediction
- state
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention provides a method for planning a prediction path of an intelligent spacecraft cluster distributed model, which comprises the following steps: determining target orbit parameters and initial orbit parameters of each spacecraft in a cluster according to a path planning target of the spacecraft cluster; establishing a spacecraft dynamics model, and describing the motion state and track information of a spacecraft by adopting a state vector; determining constraint conditions of path planning of each spacecraft, respectively constructing a model prediction controller of each spacecraft, optimizing controller parameters by adopting a Metterol Bolus-Heisynes sampling algorithm, setting a sampling period, discretizing a spacecraft orbit dynamics model, taking the discretized spacecraft orbit dynamics model as a prediction model, taking target orbit parameters as a control target, and respectively performing model prediction control on the orbit of each spacecraft by combining the constraint conditions. The invention solves the problem of generating the track of the spacecraft cluster on line in real time, reduces the calculated amount and improves the robustness of the controller.
Description
Technical Field
The invention belongs to the technical field of spacecraft path planning, and particularly relates to an intelligent spacecraft cluster distributed model prediction path planning method.
Background
Under the promotion of continuous traction of space application requirements and the development of aerospace technologies, on one hand, the range of exploring and developing space resources by human beings gradually extends from orbital space to deep space fields such as moon and mars and extends to the adjacent space between the traditional aviation and aerospace, and the environment faced by space activities is more complex; on the other hand, the mode of the spacecraft for carrying out the task is also continuously developed, the task is independently implemented by a single main body, and the task is cooperatively completed by a plurality of main bodies, so that the control difficulty is further increased. The spacecraft trajectory planning is to design a flight trajectory which starts from an initial state and meets terminal conditions and has practical physical feasibility by considering factors such as spacecraft dynamic characteristics, constraint conditions and the like aiming at a specified flight task. The overall efficiency of the system can be improved to the maximum extent by planning the spacecraft track, the characteristics of the spacecraft can be reflected most by planning and controlling the task track, and the method is also the key point for reducing the risk of the spacecraft for carrying out the task and improving the reliability and the benefit of the task operation in a complex environment. In fact, trajectory planning and control are important problems throughout the whole life cycle of a spacecraft system, and play irreplaceable roles in scheme demonstration and design, and flight mission operation stages.
The spacecraft trajectory design often needs to optimize performance indexes such as fuel/energy consumption and mission time while meeting complex constraint conditions such as mission environment and spacecraft physical performance. Such optimal trajectory planning problems can be generally described as constrained nonlinear optimal control problems. The spacecraft trajectory planning and control has the characteristics of high state variable dimension, more constraints, strong coupling and uncertainty and the like, and provides certain challenges in both theoretical and technical realization. Spacecraft trajectory planning and control mainly faces the following challenges: firstly, a system dynamics model generally has the characteristics of high dimension and strong nonlinearity, so that the increase of the computational complexity and the reduction of the planning efficiency are easily caused, and the difficulty of the design of a controller is increased; secondly, the faced constraint conditions are numerous and relate to a plurality of aspects such as spacecraft performance, task environment, system safety and the like; thirdly, a plurality of planning elements which need to be considered are mutually coupled, so that not only are the state variables mutually coupled, but also serious coupling relations exist among the control input, the index/constraint condition and the state variables; fourthly, the complexity of the system is increased, the uncertainty is enhanced, and the requirements on the robustness of the planning result and the robustness of the trajectory control algorithm are further improved.
The traditional spacecraft trajectory planning problem is described in a mathematical language as a nonlinear control problem with constraints. The development of control theory and computer technology has led to the development of a variety of numerical methods for solving optimal control problems, which can be broadly classified into indirect and direct methods. The direct method converts the original continuous optimal control problem into a high-dimensional nonlinear parameter optimization problem by discretization. The indirect method is to convert the constrained optimal control problem of the control variable into a two-point boundary value problem meeting the first-order optimal necessary condition according to the Pontryagin extreme value principle, and then solve the shooting function by a numerical method to obtain the optimal control. The common methods for optimal control all need to be optimized in the whole time domain, generally only linear models can be processed, relatively simple constraint conditions exist, and the computational complexity is very high. The optimal control method has the disadvantages that the nonlinear situation containing complex constraints is difficult to solve and the system model needs to be accurately described due to over-emphasis on optimality, so that the optimal control method is large in calculation amount, low in convergence speed and easy to converge to a local optimal point under the common situation. In recent years, the proposed algorithm based on artificial intelligence algorithm iterative optimization such as particle swarm optimization can solve the problems to a certain extent, but real-time trajectory planning cannot be guaranteed.
Model predictive control is an emerging powerful advanced process control technique. Model predictive control based on rolling time domain optimization is based on iterative, finite time domain rolling optimization for a model of the controlled volume. The state of the controlled object is sampled within a time window and a control strategy (numerical minimization algorithm) that minimizes the cost is calculated for a short period of time of the rolling horizon in the future. In particular, online or on the fly calculations are used to explore the state trajectory evolving from the current state and the cost minimization strategy before the time window is calculated via the euler-lagrange equation. The control strategy only implements the first step, then re-samples the state of the system, computes a new control strategy from the new state, and predicts a new state path. Although this method is not necessarily optimized, the simplification in the model and time domain makes it a great improvement in real-time over the optimal control method.
The cost function design of the traditional model predictive control needs to give a proper set of parameters, and the parameter design usually depends on the experience of a controller designer, however, the performance of the controller is greatly dependent on the controller parameters, which presents a great challenge to the parameter design.
At present, a spacecraft clustering algorithm needs a central node to obtain initial states of all spacecrafts, tracks of all spacecrafts from a starting point to a final point are calculated at one time, the calculated amount is large, and meanwhile robustness is poor. Most of the existing spacecraft cluster path planning adopt an optimal control algorithm, the calculation amount in the optimization process is large, the convergence speed is low, and the local optimal point is easy to converge. The method is applied to the existing model prediction control algorithm of the spacecraft, the parameter design of a target function depends on manual adjustment, and a method for automatically setting the parameters intelligently is not used temporarily.
Disclosure of Invention
The invention solves the technical problems that: the method overcomes the defects of the prior art, reduces the calculation amount of the spacecraft cluster path planning, and effectively improves the calculation precision and stability.
The technical solution of the invention is as follows:
an intelligent spacecraft cluster distributed model prediction path planning method comprises the following steps:
(1) Determining a reference track of each spacecraft in the cluster from an initial track to a target track according to a path planning target of the spacecraft cluster;
(2) Assuming that each spacecraft only considers the weight and not the shape and only receives the control force generated by the gravity and the propulsion device of the central celestial body, and the central celestial body is an ideal central gravity body, establishing a spacecraft dynamic model, and describing the motion state and track information of the spacecraft by adopting a state vector;
(3) Determining constraint conditions of path planning of each spacecraft, including determining a feasible state vector set of each spacecraft according to position boundary constraint of each spacecraft, determining a feasible control quantity set of each spacecraft according to fuel limit carried by each spacecraft, and determining safety distance constraint of each spacecraft and other spacecrafts in a spacecraft cluster;
(4) Respectively constructing model prediction controllers of each spacecraft, setting a sampling period, discretizing a spacecraft orbit dynamic model, taking the discretized spacecraft orbit dynamic model as a prediction model, taking a reference track as a control target, and combining constraint conditions to respectively perform model prediction control on each spacecraft.
Preferably, the model prediction control is performed on each spacecraft, specifically: setting a prediction time domain and a control time domain, acquiring the current state vector of the spacecraft at each sampling moment, predicting the trajectory of the spacecraft in the prediction time domain according to the control quantity and the prediction model in the control time domain, establishing a model prediction control optimization equation in the prediction time domain by combining the reference trajectory and the constraint condition in the prediction time domain, solving the optimization equation to obtain an optimal control quantity sequence in the control time domain, and applying the first control quantity in the sequence to the spacecraft.
Preferably, the model predictive control optimization equation specifically includes:
s.t.x i,k + ＝f(x i,k )
x i,k ∈X i
u i,k ∈U i
s(x i,k )-s(x j,k )≥R safe ,for j∈N i
wherein x is i,k Represents the state vector, x, of the spacecraft i after k sampling periods i,k,r Represents the expected state vector, u, of the spacecraft i after k sampling periods i,k Represents the control quantity of the spacecraft i after k sampling periods, N represents the number of sampling periods contained in the prediction time domain, and x i + ＝f(x i ) Spacecraft dynamics model representing discretization, X i Representing a set of feasible state vectors, U, for a spacecraft i i Set of feasible control quantities, s (x), representing spacecraft i i,k ),s(x j,k ) Representing the position information of the spacecraft i, j after k sampling periods, R safe Represents the minimum safe distance, N i Indicating that the distance to the spacecraft i is less than a warning threshold R warn Set of constituent spacecraft, R warn ＞R safe The semi-positive definite matrix Q, P, R is the controller parameter of the model predictive controller.
Preferably, the controller parameters of the model predictive controller are determined by:
randomly generating different controller parameters, carrying out simulation, using a model prediction controller to control the spacecraft to track the reference track under the condition of different controller parameters, obtaining the accumulated error between the simulation track and the reference track, and obtaining the probability distribution between the controller parameters and the accumulated error;
and sampling from the probability distribution by adopting a Metterol Boris-Helsteins algorithm to determine the controller parameters.
Preferably, the sampling from the probability distribution by using the merterol bolis-blacksmith algorithm to determine the controller parameter specifically includes:
(11) Selecting an initial state ω 0 Let iteration number t =0;
(12) Assume the current state ω t Transition to candidate State ω' t Satisfying Gaussian distribution, and determining candidate state omega 'according to the following expression' t Probability of acceptance:
d(ω)＝exp(-m(ω))
wherein m (ω) represents the accumulated error of the model predictive controller under the controller parameter corresponding to the state;
(13) Generating a random number u from the uniform distribution of [0,1],
if u is less than or equal to alpha, accepting the candidate state and updating omega t+1 ＝ω‘ t ，
If u>Alpha, rejecting the candidate state and keeping the original state omega t+1 ＝ω t ；
(14) And (5) repeating the steps (102) and (103) until the iteration time t reaches a preset time, and taking the controller parameter corresponding to the state with the minimum accumulated error as the final controller parameter of the model predictive controller.
Preferably, the spacecraft dynamics model specifically comprises:
wherein r is a position vector from the earth center to the spacecraft, a represents a control force acceleration vector, t represents time, mu is a universal gravitation constant, F max To control the magnitude of the force, I sp Is the specific impulse of the engine, g 0 Is the sea level gravitational acceleration.
Preferably, the position boundary constraint of each spacecraft is implemented by limiting the value range of the orbit element, and specifically includes:
setting the value range of the semi-major axis as [ a ] min ,a max ](ii) a Setting the eccentricity value range as [ e min ,e max ]The value range of the track inclination angle is [ i ] min ,i max ]The range of the ascending intersection red meridian is 0,180 DEG]The amplitude and angle of the near place is [ -180 DEG, 180 DEG)]True near pointThe angle being in the range of [0,360 ° ]]。
Preferably, the state vector is composed of a position vector and a velocity vector of the spacecraft.
Preferably, the state vector is formed by kepler orbit parameters of the spacecraft.
Compared with the prior art, the invention has the advantages that:
(1) According to the invention, a distributed model predictive control algorithm is adopted, the spacecraft can calculate the optimal control law only by acquiring the information of the neighboring spacecraft, and compared with the traditional centralized control, the distributed control has the advantages of better expandability, robustness, adaptability and the like; meanwhile, the control algorithm adopts a rolling optimization strategy, optimization calculation is carried out on line, the calculation cost is low, and the dynamic performance is good;
(2) According to the method, the optimal control parameters of the model predictive controller are estimated by utilizing the Metterol Boris-Heisothians algorithm, the control parameters do not need to be manually set, and the usability of the controller is improved; meanwhile, the optimal control parameters obtained by the algorithm are functions related to the state of the spacecraft, and the controller can dynamically adjust the parameters under different motion states to enable the performance to be optimal.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention;
FIG. 2 is a block diagram of the method architecture of the present invention;
FIG. 3 is a schematic diagram of a spacecraft cluster path planning of the present invention;
FIG. 4 is a schematic diagram of modeling a coordinate system according to an embodiment of the present invention.
Detailed Description
The features and advantages of the present invention will become more apparent and appreciated from the following detailed description of the invention.
Those skilled in the art will appreciate that those matters not described in detail in the present specification are well known in the art.
As shown in fig. 1 to fig. 2, the present embodiment discloses an intelligent spacecraft cluster distributed model prediction path planning method, which generates a trajectory on line based on distributed model prediction control, and optimizes model prediction controller parameters by using a metterol bolis-blacksmith sampling algorithm, including:
In this embodiment, as shown in fig. 4, to facilitate modeling of the spacecraft, three coordinate systems are established: the earth center inertial coordinate system, the earth center orbit coordinate system and the spacecraft orbit coordinate system. The geocentric inertial coordinate system takes the earth mass center as the origin of coordinates, the x axis points to the spring equinox point from the earth mass center along the intersection line of the ecliptic plane and the equatorial plane, the z axis points to the north from the earth mass center along the earth rotation axis, and the y axis, the x axis and the z axis meet the right-hand spiral rule. The earth center orbit coordinate system takes the earth mass center as the coordinate origin, the x axis points to the near point along the orbit plane from the earth mass center, the y axis points to the true near point angle direction from the earth mass center along the semi-path direction, and the z axis meets the right-hand spiral rule with the x axis and the y axis along the angular momentum vector from the earth mass center. The spacecraft orbit coordinate system takes the spacecraft barycenter as a coordinate origin, the x-axis points to the nose direction from the spacecraft barycenter along an orbit plane, the y-axis moves from the spacecraft barycenter along the spacecraft and is perpendicular to the x-axis, the z-axis from the spacecraft barycenter along the normal direction of the orbit plane meets the right-hand spiral rule with the x-axis and the y-axis.
Preferably, the state vector comprises a set of position vectors r and velocity vectors or a set of kepler orbit parameters. The position and speed information and the Kepler orbit parameter information can be mutually converted to obtain spacecraft motion information and characteristics of a spacecraft orbit. The Kepler orbit parameter describes the motion state of the spacecraft by a group of vectors [ a e i omega theta ], wherein a is a semimajor axis, e is an eccentricity, i is an orbit inclination angle, omega is a ascension of a rising intersection point, omega is an amplitude angle of a near place, and theta is a true near point angle.
And 102, establishing a spacecraft dynamic equation on the assumption that only the weight of the space spacecraft is considered and the shape of the space spacecraft is not considered, the space spacecraft is only controlled by the central gravity and the control force generated by a propulsion device of the central gravity, and the central celestial body is an ideal central gravity body.
As a further preferred technical scheme, based on the assumption, the kinetic equation of the non-Kepler motion of the spacecraft is as follows
Wherein r is a position vector from the geocentric to the spacecraft, a represents a control force acceleration vector, t represents time, and mu is a universal gravitation constant;
under the action of finite continuous thrust, the mass change differential equation of the spacecraft is as follows
Wherein, F max Amplitude of thrust of the engine, I sp Is the specific impulse of the engine, in seconds, g 0 Is the sea level gravitational acceleration. Since the thrust of the engine is often not adjustable, generally F is set max Is a constant value.
In the embodiment, as shown in fig. 3, the spacecraft type and mass determination is implemented by establishing a dynamic equation and a mass change differential equation under finite continuous thrust for each spacecraft, and determining the motion state change of the spacecraft under the action of the specified thrust generated by the propulsion device.
And 103, determining the position boundary constraint of the spacecraft by limiting the value range of each orbit element.
It should be noted that, the specific numerical range of each track element is set as the value range [ a ] of the semi-major axis of the track parameter min ,a max ](ii) a In order to avoid the occurrence of singular values when the eccentricity e or the orbit inclination angle i is zero or smaller in an orbit dynamics model established according to the classical orbit element, the eccentricity constraint range [ e ] of the space spacecraft is set min ,e max ]The value range of the track inclination angle is [ i ] min ,i max ]The range of the ascending crossing point right ascension is [0,180 °]The range of amplitude and angle of near place is [ -180 deg. and 180 deg. °]The value range of true proximal angle is [0,360 °]。
And 104, determining an energy boundary value of the spacecraft according to the limit of the fuel carried by the spacecraft.
It should be noted that the thrust required by the spacecraft orbit transfer cluster in the flight process is provided by fuel combustion, and the range of the available energy of the spacecraft is determined to meet the requirement of the spacecraft on the available energy according to the limit of the fuel carried by the spacecraft
E≤E max
Wherein E represents the available energy of the spacecraft, E max It represents the total energy that the spacecraft can provide by carrying fuel for combustion.
In this embodiment, the fuel conversion rate is determined according to the type of the spacecraft, the total available energy is determined according to the initial fuel type and the mass, the energy boundary value of the spacecraft is further obtained, and the energy required by the spacecraft is limited within the energy boundary value.
And 105, constructing a network topological graph of the spacecraft cluster, and establishing space safety constraint of the spacecraft.
As a further preferable technical solution, the spacecraft distance alert threshold R is set warn The alarm threshold value R warn Should be greater than a minimum safe distance.
Further, when the distance between the spacecraft and other surrounding spacecrafts is smaller than a warning threshold value R in the moving process of the spacecraft warn And establishing mutual communication and broadcasting self position information at regular time, and considering the self position information in subsequent track generation. The constraints are expressed as:
s(x i )-s(x j )≥R safe ,for j∈N i
wherein s (x) i ),s(x j ) Representing position information of the spacecraft, R safe Represents the minimum safe distance, N i Indicating that the distance to the spacecraft i is less than the warning threshold value R warn And (4) forming a spacecraft assembly.
And 106, determining target orbit parameters according to the set path planning target, and determining initial orbit parameters of each spacecraft.
In the present embodiment, the parameter selection is as shown in the following table.
Table 1 initial orbit parameters and target orbit parameters of a spacecraft cluster in an embodiment of the present invention
| Parameter(s) | Aircraft 1 | |
Aircraft 1 | Aircraft 4 | Target track |
| Semi-major axis | 14000 | 13000 | 11000 | 10000 | 12000 |
| Eccentricity ratio | 0.2 | 0.13 | 0.15 | 0.19 | 0.14 |
| Inclination angle of track | 0.16 | 0.25 | 0.15 | 0.22 | 0.20 |
| Ascending crossing point of the right ascension | 1.00 | 1.00 | 1.00 | 1.00 | 1.00 |
| Argument of near place | 0.10 | 0.10 | 0.10 | 0.10 | 0.10 |
| True proximal angle | 3.00 | 0 | 1.00 | 0.70 | - |
| Quality of | 1000 | 1000 | 1000 | 1000 | - |
And step 107, at each sampling moment, using a Metterol Polish-Black-Stentines algorithm to sample the optimized parameters.
In this embodiment, the step 107 may specifically include the following sub-steps:
substep 1071, initialization of parameters, selecting initial state ω according to the parameters in step 106 0 Let time t =0.
Substep 1072, selecting the initial parameter distribution and assuming the transfer model t (ω) t+1 ,ω t ) Obeying a gaussian distribution.
Substep 1073, the probability that the new distribution is accepted is calculated as follows:
wherein the scoring function d (ω) t ) In the present embodiment, the following are defined
d(ω)＝exp(-m(ω))
Where m (ω) represents a cumulative measure of error accumulated by the controller during trajectory planning, and each measurement error can be derived from a state estimate.
Substep 1074, from [0, 1%]Generates a random number u, accepts the state and updates ω if u is less than or equal to α t+1 If u is>Alpha is receiving the state and keeping the original state;
substep 1075, let t = t +1, repeat the above procedure.
And step 108, solving a track optimization problem on line by using distributed model predictive control according to the sampling information.
As a further preferred technical scheme, a distributed model is established by using a dynamic equation of a discretized spacecraft after each sampling to solve the problem of predictive control optimization, and the specific control requirements are expressed as follows:
s.t.x i,k + ＝f(x i,k )
x i,k ∈X i
u i,k ∈U i
s(x i,k )-s(x j,k )≥R safe ,for j∈N i
wherein x i,k Represents the state of the spacecraft i at step k, x i,k,r Denotes the reference of the spacecraft i at step kState u i,k Representing the input, x, of the spacecraft i at step k i + ＝f(x i ) Discrete State transfer equation, X, representing a spacecraft i And U i Representing a feasible state and a set of inputs for the spacecraft. The semi-positive definite matrix Q, P, R is a parameter of the cost function of the model predictive control.
And step 109, numerically solving an optimization equation of model predictive control, obtaining a control sequence and applying the first output to the spacecraft.
In this embodiment, the equations of the model predictive controller in step 108 are solved, and the output control quantity acts on the spacecraft in a manner that the propulsion device generates thrust, so as to plan the trajectory of the spacecraft on line in real time.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.
Those skilled in the art will appreciate that the invention may be practiced without these specific details.
Claims (9)
1. An intelligent spacecraft cluster distributed model prediction path planning method is characterized by comprising the following steps:
(1) Determining a reference track from an initial track to a target track of each spacecraft in a cluster according to a path planning target of the spacecraft cluster;
(2) Assuming that each spacecraft only considers the weight and not the shape and is only controlled by the gravity of a central celestial body and the control force generated by a propulsion device of the central celestial body, and the central celestial body is an ideal central gravity body, establishing a spacecraft dynamics model, and describing the motion state and track information of the spacecraft by adopting a state vector;
(3) Determining constraint conditions of path planning of each spacecraft, including determining a feasible state vector set of each spacecraft according to position boundary constraint of each spacecraft, determining a feasible control quantity set of each spacecraft according to fuel limit carried by each spacecraft, and determining safety distance constraint of each spacecraft and other spacecrafts in a spacecraft cluster;
(4) Respectively constructing model prediction controllers of each spacecraft, setting a sampling period, discretizing a spacecraft orbit dynamic model, taking the discretized spacecraft orbit dynamic model as a prediction model, taking a reference track as a control target, and combining constraint conditions to respectively perform model prediction control on each spacecraft.
2. The method for planning the distributed model prediction path of the intelligent spacecraft cluster according to claim 1, wherein the model prediction control is performed on each spacecraft, specifically: setting a prediction time domain and a control time domain, acquiring the current state vector of the spacecraft at each sampling moment, predicting the trajectory of the spacecraft in the prediction time domain according to the control quantity and the prediction model in the control time domain, establishing a model prediction control optimization equation in the prediction time domain by combining the reference trajectory and the constraint condition in the prediction time domain, solving the optimization equation to obtain an optimal control quantity sequence in the control time domain, and applying the first control quantity in the sequence to the spacecraft.
3. The method for planning the distributed model prediction path of the intelligent spacecraft cluster according to claim 2, wherein the model prediction control optimization equation specifically comprises:
s.t.x i，k + ＝f(x i，k )
x i，k ∈X i
u i，k ∈U i
S(x i，k )-S(x j，k )≥R saf e，for j∈N i
wherein x is i,k Represents the state vector, x, of the spacecraft i after k sampling periods i,k,r Represents the expected state vector, u, of the spacecraft i after k sampling periods i,k Represents the control quantity of the spacecraft i after k sampling periods, N represents the number of sampling periods contained in the prediction time domain, and x i + ＝f(x i ) Representing a discretized spacecraft dynamics model, X i Representing a set of feasible state vectors, U, for a spacecraft i i Set of feasible control quantities, s (x), representing spacecraft i i,k ),s(x j,k ) Representing the position information of the spacecraft i, j after k sampling periods, R safe Represents the minimum safe distance, N i Indicating that the distance to the spacecraft i is less than a warning threshold R warn Set of spacecraft formed, R warn ＞R safe The semi-positive definite matrix Q, P, R is the controller parameter of the model predictive controller.
4. The intelligent spacecraft cluster distributed model predictive path planning method of claim 3, wherein the controller parameters of the model predictive controller are determined by:
randomly generating different controller parameters and carrying out simulation, using a model prediction controller to control the spacecraft to track the reference track under the condition of different controller parameters, and obtaining the accumulated error of the simulation track and the reference track to obtain the probability distribution between the controller parameters and the accumulated error;
and sampling from the probability distribution by adopting a Metterol Boris-Helsteins algorithm to determine the controller parameters.
5. The intelligent spacecraft cluster distributed model predictive path planning method of claim 4, wherein the controller parameters are determined by sampling from the probability distribution by using a merterol bolis-blacknstein algorithm, and specifically comprises:
(11) Selecting an initial state ω 0 Let the iteration number t =0;
(12) Assume the current state ω t Transition to candidate State ω' t Satisfying Gaussian distribution, and determining candidate state omega 'according to the following expression' t Accepted probability:
d(ω)＝exp(-m(ω))
wherein m (ω) represents the accumulated error of the model predictive controller under the controller parameter corresponding to the state;
(13) Generating a random number u from the uniform distribution of [0,1],
if u is less than or equal to alpha, accepting the candidate state and updating omega t+1 ＝ω‘ t ，
If u>Alpha, rejecting the candidate state and keeping the original state omega t+1 ＝ω t ；
(14) And (5) repeating the steps (102) and (103) until the iteration time t reaches a preset time, and taking the controller parameter corresponding to the minimum accumulated error state as the final controller parameter of the model predictive controller.
6. The method for planning the prediction path of the intelligent spacecraft cluster distributed model according to claim 5, wherein the spacecraft dynamics model is specifically:
wherein r is a position vector from the earth center to the spacecraft, a represents a control force acceleration vector, t represents time, mu is a universal gravitation constant, F max To control the force amplitude, I sp Is the specific impulse of the engine, g 0 Is the sea level gravitational acceleration.
7. The intelligent spacecraft cluster distributed model predictive path planning method of claim 6, wherein the position boundary constraints of each spacecraft are implemented by limiting the value ranges of orbital elements, and specifically comprises:
setting the value range of the semi-major axis as [ a ] min ,a max ](ii) a Setting the eccentricity value range as [ e ] min ,e max ]The value range of the track inclination angle is [ i ] min ,i max ]The range of the ascending crossing point right ascension is [0,180 °]The amplitude and angle of the near place is [ -180 DEG, 180 DEG)]The value range of true proximal angle is [0,360 °]。
8. An intelligent spacecraft cluster distributed model predictive path planning method according to any one of claims 1 to 7, wherein the state vector is composed of a position vector and a velocity vector of a spacecraft.
9. An intelligent spacecraft cluster distributed model predictive path planning method according to any one of claims 1 to 7, wherein the state vector is formed by Kepler orbit parameters of a spacecraft.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352289.4A CN115755598A (en) | 2022-10-31 | 2022-10-31 | Intelligent spacecraft cluster distributed model prediction path planning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211352289.4A CN115755598A (en) | 2022-10-31 | 2022-10-31 | Intelligent spacecraft cluster distributed model prediction path planning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115755598A true CN115755598A (en) | 2023-03-07 |
Family
ID=85354793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211352289.4A Pending CN115755598A (en) | 2022-10-31 | 2022-10-31 | Intelligent spacecraft cluster distributed model prediction path planning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115755598A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116738893A (en) * | 2023-08-10 | 2023-09-12 | 北京国星创图科技有限公司 | Spacecraft simulation prediction system based on analysis of synchronous transmitter |
| CN117647933A (en) * | 2024-01-26 | 2024-03-05 | 中国人民解放军国防科技大学 | Track planning method for precision improvement |
-
2022
- 2022-10-31 CN CN202211352289.4A patent/CN115755598A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116738893A (en) * | 2023-08-10 | 2023-09-12 | 北京国星创图科技有限公司 | Spacecraft simulation prediction system based on analysis of synchronous transmitter |
| CN116738893B (en) * | 2023-08-10 | 2023-11-03 | 北京国星创图科技有限公司 | Spacecraft simulation prediction system based on analysis of synchronous transmitter |
| CN117647933A (en) * | 2024-01-26 | 2024-03-05 | 中国人民解放军国防科技大学 | Track planning method for precision improvement |
| CN117647933B (en) * | 2024-01-26 | 2024-03-29 | 中国人民解放军国防科技大学 | Track planning method for precision improvement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110806759B (en) | Aircraft route tracking method based on deep reinforcement learning | |
| CN111351488B (en) | Intelligent trajectory reconstruction reentry guidance method for aircraft | |
| CN115755598A (en) | Intelligent spacecraft cluster distributed model prediction path planning method | |
| CN106697333B (en) | A kind of robust analysis method of spacecraft orbit control strategy | |
| Shi et al. | Onboard generation of optimal trajectories for hypersonic vehicles using deep learning | |
| Zhao et al. | Progress in reentry trajectory planning for hypersonic vehicle | |
| CN111813146B (en) | Reentry prediction-correction guidance method based on BP neural network prediction voyage | |
| CN114253296B (en) | Hypersonic aircraft airborne track planning method and device, aircraft and medium | |
| CN112550770B (en) | Rocket soft landing trajectory planning method based on convex optimization | |
| CN113126643A (en) | Intelligent robust reentry guidance method and system for hypersonic aircraft | |
| Bairstow | Reentry guidance with extended range capability for low L/D spacecraft | |
| CN113110560B (en) | Satellite formation reconstruction model prediction control method based on Chebyshev inequality | |
| CN112000132A (en) | Spacecraft obstacle avoidance control method based on ellipsoid description | |
| CN114370793A (en) | Rocket sublevel return and vertical landing guidance method | |
| CN114415730B (en) | Intelligent planning method for escape trajectory of spacecraft | |
| CN114721261A (en) | Rocket sublevel attitude overturning and landing online guidance method | |
| Li et al. | Closed-loop deep neural network optimal control algorithm and error analysis for powered landing under uncertainties | |
| CN116125811A (en) | Spacecraft multi-space fragment collision avoidance autonomous decision-making method based on near-end strategy optimization | |
| CN115258196A (en) | Low-orbit satellite constellation networking electric propulsion orbit transfer strategy optimization method and system | |
| Zhang et al. | Spacecraft hovering flight in a binary asteroid system by using fuzzy logic control | |
| CN109325288A (en) | A kind of Solid Launch Vehicle population parameter based on uncertainty optimization determines method and system | |
| CN113093246A (en) | Ground multi-target point imaging rapid judgment and task parameter calculation method | |
| Jun et al. | Application of collaborative optimization using response surface methodology to an aircraft wing design | |
| CN115289917B (en) | Rocket sublevel landing real-time optimal guidance method and system based on deep learning | |
| Wang et al. | Deep learning based missile trajectory prediction |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |