CN114815872B - Constellation intelligent autonomous orbit control method for collision avoidance - Google Patents
Constellation intelligent autonomous orbit control method for collision avoidance Download PDFInfo
- Publication number
- CN114815872B CN114815872B CN202210673197.XA CN202210673197A CN114815872B CN 114815872 B CN114815872 B CN 114815872B CN 202210673197 A CN202210673197 A CN 202210673197A CN 114815872 B CN114815872 B CN 114815872B
- Authority
- CN
- China
- Prior art keywords
- spacecraft
- neural network
- state
- optimal
- controller
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000013528 artificial neural network Methods 0.000 claims abstract description 57
- 230000006870 function Effects 0.000 claims abstract description 40
- 238000003062 neural network model Methods 0.000 claims abstract description 32
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000012546 transfer Methods 0.000 claims abstract description 20
- 238000011217 control strategy Methods 0.000 claims abstract description 15
- 230000004913 activation Effects 0.000 claims abstract description 13
- 238000013135 deep learning Methods 0.000 claims abstract description 4
- 230000001133 acceleration Effects 0.000 claims description 16
- 238000012549 training Methods 0.000 claims description 14
- 239000000446 fuel Substances 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 4
- 210000002569 neuron Anatomy 0.000 claims description 4
- 150000001875 compounds Chemical class 0.000 claims description 3
- 238000005457 optimization Methods 0.000 claims description 3
- ORILYTVJVMAKLC-UHFFFAOYSA-N Adamantane Natural products C1C(C2)CC3CC1CC2C3 ORILYTVJVMAKLC-UHFFFAOYSA-N 0.000 claims description 2
- 230000005484 gravity Effects 0.000 claims description 2
- 230000010355 oscillation Effects 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 230000017105 transposition Effects 0.000 claims description 2
- 230000001537 neural effect Effects 0.000 claims 1
- 238000004364 calculation method Methods 0.000 abstract description 6
- 238000013461 design Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
- G05D1/0808—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft
- G05D1/0816—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability
- G05D1/0833—Control of attitude, i.e. control of roll, pitch, or yaw specially adapted for aircraft to ensure stability using limited authority control
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
Abstract
The invention discloses a constellation intelligent autonomous orbit control method for collision avoidance, and relates to a constellation intelligent autonomous orbit control method for collision avoidance. The invention aims to solve the problems that the traditional optimal control problem solving method has large calculation amount and is extremely sensitive to an initial value, or the result precision depends on the selection of the shape of the track, and the like. The process is as follows: s1, constructing a controller neural network model based on deep learning: firstly, the method comprises the following steps: solving the optimal track transfer problem by an indirect method, and constructing an optimal control database; II, secondly, the method comprises the following steps: designing a neural network structure, wherein the neural network structure comprises the number of layers of a neural network, the number of nodes of each layer and an activation function; thirdly, the method comprises the following steps: obtaining an optimal controller model of the spacecraft, and generating an optimal control strategy according to the current and expected state information in real time; and S2, constructing the intelligent autonomous controller of the satellite constellation thrust considering collision avoidance based on the neural network model trained in the S1 and the artificial potential function. The invention is used in the field of satellite orbit control.
Description
Technical Field
The invention belongs to the field of satellite orbit control, and particularly relates to a constellation intelligent autonomous orbit control method aiming at collision avoidance.
Background
With the increasing complexity of space tasks, the problem of constellation control gradually becomes a hotspot and difficulty in the field of aerospace engineering. The constellation is a carrier for solving space problems in a multi-satellite cooperative mode including satellite clustering, formation and constellation, and compared with a single independent satellite system, the reliability, task diversity, function expandability and other aspects are remarkably improved, so that the method is an important direction for future satellite technology development.
However, the space environment of the constellation is complex, and as the number of satellites increases, the probability of inter-satellite collision of the satellites in the maneuvering process also increases greatly, and how to better integrate maneuvering targets and collision avoidance problems still is a difficult point in the field of current aerospace engineering.
At present, the traditional optimal control problem solution includes a direct method, an indirect method and a shape method, wherein the direct method directly discretizes state variables and control variables of a system, but the computation amount of the direct method for complex problems is large, the indirect method utilizes an optimal control theory, although a solution with high precision can be obtained, the indirect method is extremely sensitive to an initial value, the shape method is a method for reversely deducing a spacecraft control strategy by estimating a small-thrust-transferred orbit, and although the calculation speed is high, the result precision depends on the selection of the shape of the orbit. In addition, the traditional maneuvering process including collision avoidance mostly needs to calculate the control rate on the ground and then inject planets, so that the limitation that the calculation complexity increases along with the increase of the number of the satellites and the potential collision danger caused by delay generated by injecting information on the ground are necessarily faced, and therefore, the design of an intelligent controller which is flexible, can generate optimal control information in real time and can realize autonomous collision avoidance needs to be completed.
Disclosure of Invention
The invention aims to solve the problems that the traditional optimal control problem solving method has large calculation amount and is extremely sensitive to an initial value, or the accuracy of a result depends on the selection of an orbit shape and the like, and provides an intelligent autonomous orbit control method for a constellation aiming at collision avoidance.
A constellation intelligent autonomous orbit control method for avoiding collision specifically comprises the following processes:
s1, constructing a controller neural network model based on deep learning; the specific process is as follows:
the method comprises the following steps: solving the optimal track transfer problem by an indirect method, and constructing an optimal control database;
step two: designing a neural network structure, including the number of layers of the neural network, the number of nodes of each layer and an activation function;
step three: obtaining an optimal controller model of the spacecraft to realize real-time operation according to the current and expected state information (x) c ,m c ,x t ) Generating an optimal control strategy (u, α);
s2, constructing a satellite constellation thrust intelligent autonomous controller considering collision avoidance based on the neural network model trained in the S1 and an artificial potential function; the specific process is as follows:
step 1: constructing a collision avoidance controller by using a potential function;
step 2: judging whether the current state of the spacecraft meets the state allowable deviation or not; if not, executing the step 3; if yes, executing step 4;
and 3, step 3: controlling the spacecraft by using the neural network model trained in the S1, and executing the step 4;
and 4, step 4: judging whether the current state of the spacecraft has collision risks or not in real time, and if the current state of the spacecraft has collision risks, executing the step 5; if the spacecraft has no collision risk, executing step 6;
and 5: utilizing the potential function to carry out collision avoidance maneuvering control on the spacecraft until the spacecraft does not have collision risk, and executing step 6
Step 6: and judging whether the current state of the spacecraft meets the state allowable deviation again, if so, finishing the control, if not, returning to the step 2, repeating the step 2 to the step 6 until all the current states of the spacecraft meet the state allowable deviation and no collision risk exists, and finishing the control.
The invention has the beneficial effects that:
the invention provides a design scheme of an optimal spacecraft orbit transfer controller based on a neural network. In the invention, the neural network with a designed structure is trained by utilizing the neural network data generated on the ground so as to realize the purpose of generating optimal control on the satellite intelligently and autonomously, and in addition, the neural network controller is combined with a potential function so as to realize the purpose of avoiding the autonomous collision on the satellite in the constellation maneuvering process. The invention not only can solve the problem that the traditional control needs to be re-planned when disturbance occurs, but also can avoid the trouble that a proper window needs to be waited for because the ground station uploads an instruction to the satellite, and the controller also provides an effective design idea for the design problem of the orbit transfer controller of other deep space tasks in the solar system and the problem of autonomously avoiding inter-satellite collision of a constellation and even a giant constellation.
Drawings
FIG. 1 is a flow chart of the design of an optimal controller for spacecraft orbit transfer based on a neural network according to the present invention;
FIG. 2 is a flow chart of the operation of the autonomous maneuvering and collision avoidance controller on the spacecraft based on the neural network of the invention;
FIG. 3a is a schematic diagram of the neural network structure of the spacecraft orbit transfer optimal controller based on the neural network 1 of the present invention, wherein L (0) Representing the input layer of a neural network, L (l) Denotes the subsequent hidden layer, L (l+1) Representing an output layer;
FIG. 3b is a schematic diagram of the neural network structure of the spacecraft orbit transfer optimal controller based on the neural network 2 according to the present invention;
FIG. 4a is a diagram of satellite distribution after the time of operation of the constellation configuration of the present invention is 0 s;
FIG. 4b shows that the constellation configuration of the present invention has a runtime of 9.68 × 10 4 s back satellite distribution chart;
FIG. 4c shows the constellation configuration of the present invention having a runtime of 1.94X 10 5 s back satellite distribution chart;
FIG. 4d shows the constellation configuration of the present invention having a runtime of 2.90X 10 5 s rear satellite distribution situation graph;
FIG. 4e shows the constellation configuration of the present invention having a runtime of 3.87X 10 5 s back satellite distribution chart;
FIG. 4f shows the constellation configuration of the present invention having a runtime of 4.84X 10 5 S rear satelliteA distribution situation graph;
fig. 5 is a relative distance image between spacecrafts according to the present invention.
Detailed Description
The first embodiment is as follows: a constellation intelligent autonomous orbit control method for avoiding collision specifically comprises the following processes:
the invention designs an intelligent controller design based on a neural network. The algorithm overcomes the problems of large calculation amount and the like of the traditional method, the current state and the expected state of the spacecraft are directly used as input quantities, the current optimal control strategy is obtained through a fitted neural network, and the optimal control strategy is combined with a collision avoidance link based on a potential function, so that the design of an on-satellite intelligent autonomous controller with collision avoidance is realized.
Aiming at the defects of the prior art, the invention provides a design scheme of a spacecraft on-satellite autonomous maneuvering and collision avoidance controller based on a neural network, and in order to achieve the purpose, the technical scheme adopted by the invention comprises the following two parts:
s1, constructing a controller neural network model based on deep learning; the specific process is as follows:
the method comprises the following steps: solving the optimal track transfer problem by an indirect method, and constructing an optimal control database;
step two: designing a neural network structure, including the number of layers of the neural network, the number of nodes of each layer and an activation function;
step three: obtaining an optimal controller model of the spacecraft to realize real-time operation according to the current and expected state information (x) c ,m c ,x t ) Generating an optimal control strategy (u, α);
s2, constructing a satellite constellation low-thrust intelligent autonomous controller considering collision avoidance based on the neural network model trained in the S1 and the artificial potential function; the specific process is as follows:
step 1: constructing a collision avoidance controller by using a potential function;
step 2: judging whether the current state of the spacecraft meets the state allowable deviation or not; if not, executing the step 3; if yes, executing step 4;
and step 3: controlling the spacecraft by using the neural network model trained in the S1 (inputting the current and expected state information into the neural network model trained in the S1, outputting an optimal control strategy, and controlling the spacecraft based on the optimal control strategy), and executing the step 4;
and 4, step 4: judging whether the current state of the spacecraft has collision risks or not in real time (formula 11), and if so, executing the step 5; if the spacecraft has no collision risk, executing step 6;
and 5: performing collision avoidance maneuver control (formula 12-18) on the spacecraft by using the potential function (performing collision avoidance maneuver control on the state and the like of the spacecraft controlled based on the optimal control strategy in the step 3 by using the potential function to obtain the control rate updated each time) until the spacecraft does not have collision risk, and executing the step 6;
and 6: and judging whether the current state of the spacecraft meets the state allowable deviation again, if so, finishing the control, if not, returning to the step 2, repeating the step 2 to the step 6 until all the current states of the spacecraft meet the state allowable deviation and no collision risk exists, and finishing the control.
The second embodiment is as follows: the first embodiment is different from the first embodiment in that the optimal orbit transfer problem is solved through an indirect method in the first step, and an optimal control database is constructed;
the specific process is as follows:
step one, selecting a dynamic model of the spacecraft as a two-body dynamic model, wherein the adopted two-body dynamic model can be expressed as follows in a cylindrical coordinate system:
wherein D and B are intermediate variables, and the expression is as follows:
wherein the content of the first and second substances,
is the first derivative of state x with respect to time; x = [ r, theta, z, v = r ,v θ ,v z ] T R, θ and z are the radial distance, azimuth and altitude of the spacecraft, respectively; v. of r 、v θ And v z Representing the first derivatives of r, θ and z with respect to time, respectively; r is the distance from the center of the spacecraft to the central celestial body,
T max maximum thrust of the spacecraft, I sp And g 0 Respectively representing the specific impulse of the propeller and the average gravity acceleration of the earth; u is the ratio of the actual thrust to the maximum thrust of the engine, and u belongs to [0,1 ]]；α＝[α r ,α θ ,α z ] T Is the direction of thrust, α r 、α θ 、α z Components of the thrust direction in the radial direction, the primary normal direction and the secondary normal direction are respectively; m is the spacecraft mass; μ is the gravitational constant of the central celestial body, μ =398600.4415km for a near earth satellite 3 /s 2 (ii) a The superscript "T" represents the matrix transposition; t is time;
the second step, time-free index J of the fuel optimal control problem can be expressed as:
wherein, t f Is the transfer time of the task;
obtaining an initial state and an expected state of a nominal track, respectivelyIs shown as
And
step three, solving the time-free fuel optimal control problem by using an indirect method based on the initial state and the expected state of the nominal orbit to obtain initial values and transfer time of the covariates, which are expressed as the optimal solution lambda * ；
Step one and four, acquiring the initial state of a group of new nominal tracks
And the expected state
The value of (c):
wherein, the first and the second end of the pipe are connected with each other,
are state quantities given in the form of six numbers; δ x co ，δx to Represents a sufficiently small random quantum;
step one or five, the optimal solution lambda is obtained * As an optimal solution for solving the new state
An initial value of (1);
initial state based on new nominal track
And the expected state
Value of (a), time-free fuel optimization using an indirect methodSolving the control problem to obtain a new initial value of the covariate and the transfer time, which are expressed as the optimal solution in the new state
Step six, repeating the step four and the step five to obtain a plurality of optimal solutions, wherein the optimal solutions correspond to a plurality of optimal tracks;
when δ x co ，δx to When the time is small enough, the target practice of solving the indirect method is converged quickly;
establishing a neural network database, sampling each obtained optimal track in M time discrete points, and obtaining current and expected state-optimal control pairs (x) by sampling c ,m c ,x t ,U*)；
Obtaining a plurality of groups of current and expected state-optimal control pairs (x) by sampling a plurality of optimal tracks c ,m c ,x t U), building an optimal control database;
wherein x c ,m c Is the current state of the spacecraft, x t U is the optimal control for the desired state.
Other steps and parameters are the same as those in the first embodiment.
The third concrete implementation mode: the second step is different from the first or second specific embodiment in that a neural network structure is designed, including the number of layers of the neural network, the number of nodes of each layer and an activation function;
the specific process is as follows:
the neural network model of the controller is a feedforward fully-connected neural network and comprises a neural network 1 model and a neural network 2 model;
the inputs of the neural network 1 and 2 models of the controller are the current state and the expected state [ x ] of the spacecraft c ；m c ；x t ]The output of the neural network 1 model of the controller is the thrust amplitude u of the spacecraft, belonging to [0, 1' ]]The output of the neural network 2 model is the radial thrust direction angle and the sub-normal thrust direction angle [ theta ] r ,θ z ] T Wherein, theta r ∈[-π,π]，θ z ∈[-π/2,π/2]Comprises the following steps:
the neural network 1 sequentially comprises an input layer, a 3-layer hidden layer and an output layer, wherein each hidden layer comprises 128 neurons, a Sigmoid function is selected as an activation function of the output layer, and ReLU is selected as the activation function in the hidden layer; the neural network 2 sequentially comprises an input layer, 9 hidden layers and an output layer, wherein each hidden layer comprises 128 neurons, a Tanh function is selected as an activation function of the output layer, and a ReLU is selected as an activation function in the hidden layer;
the neural network is parameterized by a weight omega and an offset b, and a Mean Square Error (MSE) between training data and a network prediction result is adopted as a loss function:
where Net represents the neural network that needs to be trained, N is the total number of samples used for training X i In order to input the data, the data is,
for the neural network to expect output, | | | | represents the two-norm of the vector;
parameters in the neural network are trained by using an Adam optimization algorithm, and the learning rate is set to be 0.0001.
Other steps and parameters are the same as those in the first or second embodiment.
The fourth concrete implementation mode: the difference between the first embodiment and the third embodiment is that the optimal controller model of the spacecraft is obtained in the third step, and the current and expected state information (x) is obtained in real time c ,m c ,x t ) Generating an optimal control strategy (u, α); the specific process is as follows:
extracting data in a database as a training set, inputting the training set into the constructed controller neural network models 1 and 2, and training the constructed controller neural network models to obtain two trained neural network models and obtain an optimal spacecraft controller model;
associating current and expected state information [ x ] c ；m c ；x t ]Inputting the trained neural network models 1 and 2, and outputting the trained neural network model 1 as u net The output of the trained neural network model 2 is [ theta ] r ,θ z ] T The thrust direction vector that the spacecraft is subjected to at this time
Comprises the following steps:
in the formula (I), the compound is shown in the specification,
the components of the thrust direction calculated by the neural network in the radial direction, the primary normal direction and the secondary normal direction are respectively represented, so that the optimal control strategy (u, alpha) is as follows:
other steps and parameters are the same as those in one of the first to third embodiments.
The fifth concrete implementation mode: the present embodiment is different from one of the first to fourth embodiments in that, in the step 1, a collision avoidance controller is constructed by using a potential function; the specific process is as follows:
setting the safety constraints of a spacecraft is expressed as:
||r min ||＞L (10)
wherein r is min Representing the nearest distance between two spacecrafts, and L representing the minimum distance allowed between the spacecrafts;
the repulsion potential of a spacecraft to its nearest neighbor is:
wherein, U o (x i ,x j ) Is the repulsive force, x, of the spacecraft j closest to the spacecraft i i ,x j Is the state of the i, j two spacecrafts, d 0 Is the repulsive force field radius, and k is the repulsive force gain coefficient;
representing the gradient of the function, when the closest distance between two spacecrafts is greater than d 0 When the two spacecrafts are not in collision, the closest distance between the two spacecrafts is less than or equal to d 0 When there is a risk of collision between two spacecraft, F o (x i ,x j ) The amplitude of the repulsion force generated by the final repulsion field to the spacecraft;
assuming that the direction of the tangential acceleration experienced by the spacecraft i is alpha ui Thus, it can be expressed as:
wherein a is i Denotes the semi-major axis of the ith spacecraft, a j Denotes the semi-major axis, da, of the spacecraft closest to the ith spacecraft 0 To prevent the thrust direction oscillation generated by the over-small difference of the semi-major axes;
in order to transfer the tangential acceleration direction vector to the cylindrical coordinate system, firstly, the direction vector is transposed to the radial direction S, and in the transverse direction T and the orbital plane normal direction W, the intermediate variables A and B are expressed as follows:
comprises the following steps:
S＝Aα ui
T＝Bα ui (15)
W＝0
wherein e represents the orbital eccentricity and the orbital eccentricity,
representing true proximal angles;
thus, will [ S, T, W] T Transfer to the geocentric inertial frame, with:
in the formula, a x 、a y 、a z Representing the acceleration component of the spacecraft under the geocentric inertial coordinate system;
and
representing the angle of rotation of the vector about the x, z axes, respectively
The rotation matrix of (a); omega represents the ascension of the intersection point, i represents the track inclination angle, and omega represents the argument of the perigee;
each represents-omega, -i or
Obtaining the acceleration direction vector alpha under the cylindrical coordinate system by the acceleration direction vector under the earth center inertia coordinate system o-rθz ＝[α r ,α θ ,α z ] T The following:
α r ＝a x cosθ+a y sinθ
α θ ＝-a x sinθ+a y cosθ (17)
α z ＝a z
in the formula, alpha r 、α θ 、α z Components of thrust direction in radial direction, primary normal direction and secondary normal direction, respectively
The acceleration obtained by the method is summed with the acceleration obtained by the neural network controller, and the control rate of the collision avoidance spacecraft can be obtained, namely:
other steps and parameters are the same as in one of the first to fifth embodiments.
The sixth specific implementation mode is as follows: the difference between this embodiment and one of the first to fifth embodiments is that the state allowable deviation of the spacecraft in the steps 2 to 6 is given in the form of cylindrical coordinates, and the state deviation of the spacecraft is specifically expressed as follows:
wherein, | Δ r |, | Δ θ |, | Δ z |, | Δ v |, etc r |、|Δv θ I and | Δ v z Respectively representing the current state
And the expected state
Rem (p, q) represents the remainder of dividing p by q;
if the current state of the spacecraft is less than or equal to the allowable deviation, the spacecraft is still likely to deviate from the expected state of the spacecraft due to collision avoidance maneuver, and at the moment, in order to avoid the frequent switch state switching of the controller, the state allowable deviation is used for starting the limit
And a stop limit
The form of (1) is given;
when all state deviations of the spacecraft are smaller than or equal to the stopping limit, the state allowable deviation is considered to be met, and when any state deviation of the spacecraft is larger than the starting limit, the state allowable deviation is considered to be not met any more;
the start limits are each greater than the stop limits.
Other steps and parameters are the same as those in one of the first to fifth embodiments.
The following examples were used to demonstrate the beneficial effects of the present invention:
the first embodiment is as follows:
an orbital plane in a large constellation is selected, assuming that the phases of the orbital plane are uniformly distributed among 100 satellites. Each satellite has a mass of 270kg, and adopts an electric propeller and a specific impulse I sp =3000s, maximum thrust T that can be provided max =100mN, common initial semi-major axis of the spacecraft a =7378km, eccentricity e =0.1, and common orbit inclination i, ascension (RAAN) Ω, perigee argument ω and true perigee angle
Are all 0, the maximum time of the limiting task is 4.841 multiplied by 10 5 s, the satellite is adjusted according to its initial phase
The numbers from small to large are 1-100, five satellites are randomly selected from the numbers of [1,23,58,75,88] to rearrange the phases]And rearranged to [58,88,1,23,75 ] by rail-powered operation]。
TABLE 1 Artificial potential function parameters and control limits
The satellite position change during the maneuver is shown in FIGS. 4a, 4b, 4c, 4d, 4e, 4f, where number 12345 corresponds to the selected satellite number [1,23,58,75,88]:
through 4.841X 10 5 The s satellite realizes configuration reconstruction, and the minimum relative distance change between the spacecrafts is shown in figure 5: and after the control evasion link is executed, the minimum distance between the spacecrafts is 10.12km which is larger than a given 10km boundary, and under the condition that potential function collision evasion control is not carried out, the total fuel consumption of the spacecrafts for completing the process is 0.08kg-0.31kg, 1.1863kg, and the minimum distance between the spacecrafts is 8.28km which is smaller than the given 10km collision boundary.
Meanwhile, the fuel consumption of the spacecraft maneuver containing collision avoidance is 0.09kg-0.32kg, and the total fuel consumption is 1.2917kg. It can be seen that the artificial intelligence controller based on the potential function has reasonable energy consumption, and meanwhile, the effectiveness of the collision avoidance algorithm is proved.
And compared with an indirect method that the average duration needs about 2000s of calculation time, the neural network only needs about 100s to calculate a complete track once, and the average neural network controller only needs 0.0095s to calculate the control quantity once. It also represents a sufficient advantage in terms of computation time.
The present invention is capable of other embodiments and its several details are capable of modifications in various obvious respects, all without departing from the spirit and scope of the present invention.
Claims (4)
1. A constellation intelligent autonomous orbit control method aiming at collision avoidance is characterized by comprising the following steps: the method comprises the following specific processes:
s1, constructing a controller neural network model based on deep learning; the specific process is as follows:
the method comprises the following steps: solving the optimal track transfer problem by an indirect method, and constructing an optimal control database;
the specific process is as follows:
step one, selecting a dynamic model of the spacecraft as a two-body dynamic model, and expressing the adopted two-body dynamic model in a cylindrical coordinate system as follows:
wherein D and B are intermediate variables, and the expression is as follows:
wherein the content of the first and second substances,
is the first derivative of state x with respect to time; x = [ r, theta, z, v = r ,v θ ,v z ] T R, θ and z are the radial distance, azimuth and altitude of the spacecraft, respectively; v. of r 、v θ And v z Representing the first derivatives of r, θ and z with respect to time, respectively; r is the distance from the center of the spacecraft to the central celestial body,
T max maximum thrust of the spacecraft, I sp And g 0 Respectively representing the specific impulse of the propeller and the average gravity acceleration of the earth; u is the ratio of the actual thrust to the maximum thrust of the engine, and u belongs to [0,1 ]]；α＝[α r ,α θ ,α z ] T Is the direction of thrust, α r 、α θ 、α z The components of the thrust direction in the radial direction, the primary normal direction and the secondary normal direction are respectively; m is the spacecraft mass; mu is the gravitational force of the central celestial bodyAmount, μ =398600.4415km for a near earth satellite 3 /s 2 (ii) a The superscript "T" represents the matrix transposition; t is time;
the second step, time-free index J of the fuel optimal control problem is expressed as:
wherein, t f Is the transfer time of the task;
the initial state and the expected state of the nominal track are respectively represented as
And
step three, solving the index J of the time-free fuel optimal control problem by using an indirect method based on the initial state and the expected state of the nominal orbit to obtain the initial value of the co-modal variable and the transfer time, which are expressed as the optimal solution Lambda * ；
Step one and four, acquiring the initial state of a group of new nominal tracks
And the expected state
The value of (c):
wherein, δ x co ，δx to Represents a sufficiently small random quantity;
step one or five, the optimal solution lambda is obtained * As an optimal solution for solving the new state
An initial value of (1);
initial state based on new nominal track
And the expected state
The time-free fuel optimal control problem J is solved by using an indirect method to obtain a new initial value and transfer time of the co-state variables, which are expressed as an optimal solution in a new state
Step six, repeating the step four and the step five to obtain a plurality of optimal solutions, wherein the optimal solutions correspond to a plurality of optimal tracks;
sampling each obtained optimal track in M time discrete points to obtain a current and expected state-optimal control brake pair (x) c ,m c ,x t ,U*)；
Obtaining a plurality of groups of current and expected state-optimal control brake pairs (x) by sampling a plurality of optimal tracks c ,m c ,x t U), building an optimal control database;
wherein x is c ,m c Is the current state of the spacecraft, x t In the expected state, U is the optimal control;
step two: designing a neural network structure, wherein the neural network structure comprises the number of layers of a neural network, the number of nodes of each layer and an activation function; the specific process is as follows:
the neural network model of the controller is a feedforward fully-connected neural network and comprises a neural network 1 model and a neural network 2 model;
the inputs of the neural network 1 and 2 models of the controller are the current state and the expected state [ x ] of the spacecraft c ；m c ；x t ]The output of the neural network 1 model of the controller is the thrust amplitude u E [0,1 ] of the spacecraft]Neural netThe output of the collateral 2 model is a radial thrust direction angle and a sub-normal thrust direction angle [ theta ] r ,θ z ] T Wherein, θ r ∈[-π,π]，θ z ∈[-π/2,π/2]Comprises the following steps:
the neural network 1 sequentially comprises an input layer, a 3-layer hidden layer and an output layer, wherein each hidden layer comprises 128 neurons, a Sigmoid function is selected as an activation function of the output layer, and a ReLU is selected as an activation function in the hidden layer;
the neural network 2 sequentially comprises an input layer, 9 hidden layers and an output layer, wherein each hidden layer comprises 128 neurons, a Tanh function is selected as an activation function of the output layer, and a ReLU is selected as an activation function in the hidden layer;
the neural network is parameterized by a weight omega and an offset b, and the mean square error between training data and a network prediction result is taken as a loss function:
where Net represents the neural network that needs to be trained, N is the total number of samples used for training, X i In order to input the data, the data is,
for the neural network to expect output, | | | | represents a two-norm of the vector;
training parameters in the neural network by using an Adam optimization algorithm, and setting the learning rate to be 0.0001;
step three: obtaining an optimal controller model of the spacecraft to realize real-time operation according to the current and expected state information (x) c ,m c ,x t ) Generating an optimal control strategy (u, a); the specific process is as follows:
extracting data in a database as a training set, inputting the training set into the constructed controller neural network models 1 and 2, and training the constructed controller neural network models to obtain two trained neural network models and obtain an optimal spacecraft controller model;
associating current and expected state information [ x ] c ；m c ；x t ]Inputting the trained neural network models 1 and 2, and outputting the trained neural network model 1 as u net The output of the trained neural network model 2 is [ theta ] r ,θ z ] T Thrust direction vector to which the spacecraft is subjected at the moment
Comprises the following steps:
in the formula (I), the compound is shown in the specification,
the components of the thrust direction calculated by the neural network in the radial direction, the main normal direction and the secondary normal direction are respectively represented, so that the optimal control strategy (u, α) is as follows:
s2, constructing a satellite constellation thrust intelligent autonomous controller considering collision avoidance based on the neural network model trained in the S1 and the artificial potential function; the specific process is as follows:
step 1: constructing a collision avoidance controller by using a potential function;
and 2, step: judging whether the current state of the spacecraft meets the state allowable deviation or not; if not, executing the step 3; if yes, executing step 4;
and step 3: controlling the spacecraft by using the neural network model trained in the S1, and executing the step 4;
and 4, step 4: judging whether the current state of the spacecraft has collision risks or not in real time, and if the current state of the spacecraft has collision risks, executing the step 5; if the spacecraft has no collision risk, executing step 6;
and 5: utilizing the potential function to carry out collision avoidance maneuvering control on the spacecraft until the spacecraft does not have collision risks, and executing the step 6;
and 6: and judging whether the current state of the spacecraft meets the state allowable deviation or not again, if so, ending the control, otherwise, returning to the step 2, repeating the step 2 to the step 6 until all the current states of the spacecraft meet the state allowable deviation and no collision risk exists, and ending the control.
2. The intelligent constellation autonomous trajectory control method for collision avoidance according to claim 1, characterized in that: the optimal controller model of the spacecraft is obtained in the third step, and the current and expected state information (x) is obtained in real time c ,m c ,x t ) Generating an optimal control strategy (u, α); the specific process is as follows:
extracting data in a database as a training set, inputting the training set into the constructed controller neural network models 1 and 2, and training the constructed controller neural network models to obtain two trained neural network models and obtain an optimal spacecraft controller model;
comparing current and expected state information [ x ] c ；m c ；x t ]Inputting the trained neural network models 1 and 2, and outputting the trained neural network model 1 as u net The output of the trained neural network model 2 is [ theta ] r ,θ z ] T Thrust direction vector to which the spacecraft is subjected at the moment
Comprises the following steps:
in the formula (I), the compound is shown in the specification,
the components of the thrust direction calculated by the neural network in the radial direction, the main normal direction and the secondary normal direction are respectively represented, so that the optimal control strategy (u, α) is as follows:
3. the intelligent constellation autonomous trajectory control method for collision avoidance according to claim 2, characterized in that: in the step 1, a collision avoidance controller is constructed by using a potential function; the specific process is as follows:
setting the safety constraints of a spacecraft is expressed as:
||r min ||＞L (10)
wherein r is min The minimum distance between two spacecrafts is represented, and L represents the minimum distance allowed between the spacecrafts;
the repulsion force of a spacecraft closest to the spacecraft is as follows:
wherein, U o (x i ,x j ) Is the repulsive force, x, of a spacecraft j closest to the spacecraft i i ,x j Is the state of the i, j two spacecrafts, d 0 Is the repulsive force field radius, and k is the repulsive force gain coefficient;
representing the gradient of the function, when the closest distance between two spacecrafts is greater than d 0 When the two spacecrafts are in use, the risk of collision is considered to be avoided,when the closest distance between the two spacecrafts is less than or equal to d 0 When there is a risk of collision between two spacecraft, F o (x i ,x j ) The amplitude of the repulsion force generated by the final repulsion field to the spacecraft;
assuming that the direction of the tangential acceleration experienced by the spacecraft i is alpha ui Thus, it can be expressed as:
wherein a is i Denotes the semi-major axis of the ith spacecraft, a j Denotes the semi-major axis, da, of the spacecraft closest to the ith spacecraft 0 To prevent the thrust direction oscillation generated by the over-small difference of the semi-major axes;
in order to transfer the tangential acceleration direction vector to the cylindrical coordinate system, firstly, the direction vector is transposed to the radial direction S, and in the transverse direction T and the orbital plane normal direction W, the intermediate variables A and B are expressed as follows:
comprises the following steps:
wherein e represents the orbital eccentricity and the orbital eccentricity,
representing true proximal angles;
thus, will [ S, T, W] T Transfer to the geocentric inertial frame, with:
in the formula, a x 、a y 、a z Representing the acceleration component of the spacecraft in a geocentric inertial coordinate system;
and
representing the angle of rotation of a vector about the x, z axes, respectively
The rotation matrix of (a); omega represents the ascension of the intersection point, i represents the track inclination angle, and omega represents the argument of the perigee;
respectively represent-omega, -i or
Obtaining the acceleration direction vector alpha under the cylindrical coordinate system by the acceleration direction vector under the earth center inertia coordinate system o-rθz ＝[α r ,α θ ,α z ] T The following were used:
in the formula, alpha r 、α θ 、α z Components of thrust direction in radial direction, primary normal direction and secondary normal direction, respectively
The acceleration obtained by the method is summed with the acceleration obtained by the neural network controller, and the control rate of the collision avoidance spacecraft can be obtained, namely:
4. the intelligent constellation autonomous trajectory control method for collision avoidance according to claim 3, characterized in that: the state allowable deviation of the spacecraft in the steps 2 to 6 is given in the form of cylindrical coordinates, and the state deviation of the spacecraft is specifically expressed in the following manner:
wherein, | Δ r |, | Δ θ |, | Δ z |, | Δ v |, etc r |、|Δv θ I and | Δ v z Respectively representing the current state
And the expected state
Rem (p, q) represents the remainder of dividing p by q;
biasing the state allowed to start to limit
And a stop limit
The form of (b) is given; when all state deviations of the spacecraft are smaller than or equal to the stopping limit, considering that the state allowable deviation is met, and when any state deviation of the spacecraft is larger than the starting limit, considering that the state allowable deviation is not met any more;
the start limits are each greater than the stop limits.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210673197.XA CN114815872B (en) | 2022-06-14 | 2022-06-14 | Constellation intelligent autonomous orbit control method for collision avoidance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210673197.XA CN114815872B (en) | 2022-06-14 | 2022-06-14 | Constellation intelligent autonomous orbit control method for collision avoidance |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114815872A CN114815872A (en) | 2022-07-29 |
| CN114815872B true CN114815872B (en) | 2022-11-18 |
Family
ID=82521615
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210673197.XA Active CN114815872B (en) | 2022-06-14 | 2022-06-14 | Constellation intelligent autonomous orbit control method for collision avoidance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114815872B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116331518B (en) * | 2023-01-13 | 2023-09-26 | 天津大学 | Star group intelligent formation collision prevention control method based on safe self-adaptive dynamic programming |
| CN116187104B (en) * | 2023-04-27 | 2023-08-01 | 华侨大学 | Safety analysis and development method and device for rail transit interlocking system |
| CN117022678B (en) * | 2023-08-29 | 2024-01-26 | 北京航天驭星科技有限公司 | Method, system, electronic device and medium for avoiding collision risk |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541070A (en) * | 2012-01-20 | 2012-07-04 | 哈尔滨工业大学 | Collision avoiding method for ground testing system of satellite formation flying |
| CN112000132A (en) * | 2020-07-08 | 2020-11-27 | 中国人民解放军军事科学院国防科技创新研究院 | Spacecraft obstacle avoidance control method based on ellipsoid description |
| WO2021027568A1 (en) * | 2019-08-12 | 2021-02-18 | 华为技术有限公司 | Obstacle avoidance method and device |
| CN113741493A (en) * | 2021-08-11 | 2021-12-03 | 北京空间飞行器总体设计部 | Spacecraft safety approach control method |
-
2022
- 2022-06-14 CN CN202210673197.XA patent/CN114815872B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102541070A (en) * | 2012-01-20 | 2012-07-04 | 哈尔滨工业大学 | Collision avoiding method for ground testing system of satellite formation flying |
| WO2021027568A1 (en) * | 2019-08-12 | 2021-02-18 | 华为技术有限公司 | Obstacle avoidance method and device |
| CN112000132A (en) * | 2020-07-08 | 2020-11-27 | 中国人民解放军军事科学院国防科技创新研究院 | Spacecraft obstacle avoidance control method based on ellipsoid description |
| CN113741493A (en) * | 2021-08-11 | 2021-12-03 | 北京空间飞行器总体设计部 | Spacecraft safety approach control method |
Non-Patent Citations (5)
| Title |
|---|
| Collision avoidance control for formation flying of multiple spacecraft using artificial potential field;JiyoonHwang等;《Advances in Space Research》;20220331;第69卷(第5期);2197-2209 * |
| Low-thrust spacecraft trajectory optimization via a DNN-based method;YIN S等;《Advances in Space Research》;20201231;第66卷(第7期);1635-1646 * |
| 基于遗传算法的卫星攻击轨道优化方法;赵琳等;《系统工程与电子技术》;20160531(第05期);144-150 * |
| 多Euler-Lagrange系统协同跟踪避碰控制;史小平等;《黑龙江大学工程学报》;20180925(第03期);2+40-46 * |
| 深空环境下近距离绳系卫星系统控制方法研究;黄静;《中国博士学位论文全文数据库 工程科技Ⅱ辑》;20160315(第03(2016)期);C031-60 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114815872A (en) | 2022-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114815872B (en) | Constellation intelligent autonomous orbit control method for collision avoidance | |
| Zhou et al. | Event-triggered approximate optimal path-following control for unmanned surface vehicles with state constraints | |
| Chen et al. | Output consensus and collision avoidance of a team of flexible spacecraft for on-orbit autonomous assembly | |
| Wang et al. | Coordinated stabilization of tumbling targets using tethered space manipulators | |
| Chen et al. | On-orbit assembly of a team of flexible spacecraft using potential field based method | |
| Mancini et al. | Sliding mode control techniques and artificial potential field for dynamic collision avoidance in rendezvous maneuvers | |
| CN104590588A (en) | Flexible satellite attitude orbit coupling control method based on isolation allowance method and pulse width fusion strategy | |
| CN106021784A (en) | Full-trajectory optimization design method based on two-layer optimization strategy | |
| CN113602535A (en) | Method for controlling in-orbit autonomous intersection of micro/nano satellite and computer equipment | |
| CN115509248A (en) | Satellite cluster reconstruction control method based on sequence convex optimization and model prediction control | |
| CN116203835A (en) | Spacecraft orbit chase-escaping game method based on pulse thrust | |
| Xia et al. | Neuroadaptive saturated control for relative motion based noncooperative spacecraft proximity with prescribed performance | |
| Nazmy et al. | Shielded deep reinforcement learning for multi-sensor spacecraft imaging | |
| Srivastava et al. | Deep Reinforcement Learning Based Control of Rotation Floating Space Robots for Proximity Operations in PyBullet | |
| Kornfeld | On-board autonomous attitude maneuver planning for planetary spacecraft using genetic algorithms | |
| Liu et al. | Optimal formation control for multiple rotation-translation coupled satellites using reinforcement learning | |
| Wang et al. | Robust control of a dual-arm space robot to capture a non-cooperative target in 3D space | |
| Sabol et al. | Machine learning based relative orbit transfer for swarm spacecraft motion planning | |
| Ortega | Fuzzy logic techniques for rendezvous and docking of two geostationary satellites | |
| CN111596678B (en) | Relative position self-adaptive control method for spacecraft approaching space rolling target | |
| Nehrenz et al. | On the development of spacecraft operating modes for a deep space cubeSat | |
| Xu et al. | Coordinated control method of space-tethered robot system for tracking optimal trajectory | |
| Yoshimura et al. | Global trajectory design for position and attitude control of an underactuated satellite | |
| Yuan et al. | Leveraging Curriculum Reinforcement Learning for Rocket Powered Landing Guidance and Control | |
| CN115583367B (en) | Satellite cluster reconstruction control method based on distributed sequence convex optimization |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |