Disclosure of Invention
The invention provides an in-orbit autonomous three-axis quick maneuvering control method for a satellite, aiming at realizing in-orbit autonomous quick imaging of the satellite and solving the contradiction between quickness and stability when the satellite needs three-axis maneuvering.
An on-orbit autonomous three-axis fast maneuvering control method for a satellite is realized by the following steps:
calculating an expected attitude;
calculating an expected quaternion and calculating a rotating shaft and a rotating angle corresponding to three-axis maneuvering of the satellite according to the initial attitude and the target attitude of the satellite; obtaining the desired quaternion qQDesired rotation angle thetaQAnd the direction of the axis of rotation en;
Step two, planning a three-axis attitude;
under the constraint of meeting the satellite rotational inertia I, the reaction flywheel torque T and the angular momentum H, amplitude limiting and constraint setting are carried out on the three-axis angular acceleration and the angular velocity to obtain an angular acceleration limit value alpha
LGAnd angular velocity limit ω
LG(ii) a And ensuring the direction of a rotating shaft to be unchanged in the three-axis large-angle maneuvering process of the satellite, and simultaneously designing an eight-section attitude planner with continuous angular acceleration to expect the rotating angle theta
QAngular acceleration limit α
LGAngular velocity limit ω
LGAs input, the angular acceleration generates a function
The definition is as follows:
wherein, Δ tAThe value is a set value, the rising time of the angular acceleration is limited, and the value is reasonably selected according to the dynamic performance of the satellite actuating mechanism; Δ tB,ΔtCValue of (a) to the desired angle thetaQIs related to the size of the cell;
obtaining a real-time planning angle theta epsilon [0, theta ] through the attitude planner
Q]Real-time planning of angular velocity
And planning angular acceleration in real time
And obtaining a planning quaternion q
G=[cosθ;sinθe
n]Three-axis planning angular velocity of the system
And three axes planning angular acceleration
Step three, quick maneuvering control;
step three, calculating deviation angular velocity omegaEAnd deviation quaternion qE;
The rotational quaternion, i.e. the deviation quaternion, of the real-time attitude of the satellite relative to the planned attitude
The quaternion q under the satellite inertial system is a rotation quaternion of the satellite body coordinate system relative to the inertial system; initial quaternion q
CA rotation quaternion of the initial attitude of the satellite relative to the inertial system;
deviation angular velocity, which is a representation of the deviation of the real-time angular velocity of the satellite from the planned angular velocity in the inertial system
Wherein the track angular velocity omega
guiThe representation of the rotation vector of the satellite orbit system relative to the inertial system under the orbit system is shown, and the satellite angular velocity omega is the rotation angular velocity of the satellite body system relative to the inertial system;
a rotation quaternion being the initial attitude of the orbital relative to the satellite; r (q)
E) Is q
EA corresponding rotation matrix;
is composed of
A corresponding rotation matrix;
step three or two, obtaining the deviation angular speed omegaEDeviation quaternion qEAnd three-axis planned angular acceleration alphaGInputting the PD controller to realize the on-orbit autonomous three-axis quick maneuvering control of the satellite;
the PD controller is designed as follows:
u=KqαG-KPqE-KdωE
in the formula, Kq,KP,KdRespectively a feedforward control gain matrix, a proportional control increment matrix and a differential control gain matrix.
The invention has the beneficial effects that:
the invention designs a three-axis attitude planning scheme aiming at the situation of three-axis large-angle maneuvering of the satellite, shortens the required maneuvering time under the same rotation angle, improves the maneuvering performance of the satellite, and simultaneously designs a quick maneuvering algorithm to ensure the stability while realizing the rapidity of the satellite.
After the on-orbit autonomous triaxial fast maneuvering control of the satellite is carried out, the satellite is ensured to fast acquire data and simultaneously ensure the imaging stability and acquire high-quality image data when aiming at emergency tasks such as maritime search and rescue, post-disaster wide area search and rescue, emergency geographic investigation and the like. Therefore, the imaging capability of the low-orbit remote sensing satellite is improved, and the high timeliness of the image data acquired in the orbit is ensured.
Detailed Description
In the first embodiment, the present embodiment is described with reference to fig. 1 to 5, and an in-orbit autonomous three-axis maneuvering control method for a satellite relates to the following definitions:
related coordinate system definition
In the present embodiment, a body coordinate system O is usedbXbYbZbOrbital coordinate system ObXoYoZoAnd the inertia system CeXeIYeIZeIThree coordinate systems.
(1) Body coordinate system ObXbYbZb: origin of coordinates ObLocated at the center of mass of the satellite, three-axis orientation is related to the installation of the star body, and X is definedbThe axis pointing in the direction of the sailboard, ZbThe axis pointing in the direction of the camera, YbAxis and XbAxis and ZbThe axes form a right-handed rectangular coordinate system.
(2) Orbital coordinate system ObXoYoZo: the origin of coordinates is the center of mass O of the satellitebThe Y axis pointing in the opposite direction of the track angular velocity, ZoThe axis pointing to the center of the earth, XoAxis and YoAxis and ZoThe axes constitute a right-hand rectangular coordinate system (flight direction) which is a ground-oriented reference.
(3) System of inertia CeXeIYeIZeI: the origin of the coordinate system is the earth centroid Ce,XeIThe axis points to the spring (2000 years)1 month, 1 day, 12 hours), ZeIThe axis points to the flat north pole (1/12/2000, JD-2451545.0), YeIAxis and XeIAxis, ZeIThe axes form a right-handed rectangular coordinate system, also known as the J2000 Earth inertial coordinate system.
In this embodiment, the satellite attitude is described in a quaternion form, and the correlation properties are defined as follows:
description mode of satellite attitude, quaternion expression:
wherein
q
0A scale part, which is a quaternion, represents the rotation angle phi,
vector part of quaternion, representing direction e of rotation axis
n=[i;j;k]Satisfy i
2+j
2+k
2=1。
The four parameters satisfy the constraint equation:
vector product rule:
quaternion multiplication:
the specific implementation steps of the embodiment are as follows:
the method comprises the following steps: calculating an expected attitude;
and calculating an expected quaternion and a rotating shaft and a rotating angle corresponding to the three-axis maneuvering of the satellite according to the initial attitude and the target attitude of the satellite.
As can be obtained from the definition of quaternion, the attitude transformation of the initial coordinate system Oxyz relative to the target coordinate system Ox ' y ' z ' is expressed as
As shown in fig. 2.
The desired quaternion of the target attitude of the satellite relative to the initial attitude is
Wherein, an initial quaternion q
CA rotation quaternion of the initial attitude of the satellite relative to the inertial system; target quaternion q
FA rotation quaternion of the target attitude of the satellite relative to the inertial system;
by definition of quaternion, from q
QMark part q of
Q0The rotation angle phi is obtained in reverse, phi is 2arccos (q)
Q0). At the same time, from
Can obtain
When Φ is 0, the corresponding quaternion is qQ=[1;0;0;0]The target pose coincides with the initial pose.
Step two: planning a three-axis attitude;
the maneuvering process of the satellite is planned in real time according to the performance constraint of the satellite, and the one-dimensional rotation angle is generated through the attitude planner, so that the maneuvering capability of the satellite can be improved.
Under the constraint of meeting the rotational inertia I of the satellite, the moment T of a reaction flywheel and the angular momentum H, in order to realize the initial quaternion qCTo the target quaternion qFI.e. the whole desired quaternion qQTo ensure the rotation axis e in the maneuvering processnWhen the attitude planning is performed, amplitude limiting and constraint setting are required to be performed on the triaxial angular acceleration and the angular velocity.
The input and the output of the attitude planner are one-dimensional, the input is an expected rotation angle, an angular acceleration limit value and an angular velocity limit value, and the output is a real-time angle, a real-time angular velocity and a real-time angular acceleration. The three-axis attitude planning diagram is shown in figure 3.
Angular acceleration limit αLGIs calculated as follows:
αLG=||αLG||2;
wherein the rotational inertia of the satellite
Reaction flywheel triaxial moment T ═ T
x;T
y;T
z]N·m,M
max=10
20For a set larger number, the angular acceleration limit α is programmed
LG=[α
LGx;α
LGy;α
LGz]°/s
2,[·]
minTo calculate the minimum value, | ·| non-conducting phosphor
2The vector is modulo.
Inputting three-axis angular velocity limit value omegaLim=[ωLimx;ωLimy;ωLimz](ii) DEG/s; angular velocity limit ωLGIs calculated as follows:
ωLG=||ωLG||2;
wherein the reaction flywheel angular momentum H ═ Hx;Hy;Hz]N m s, three-axis angular velocity limit ωLG=[ωLGx;ωLGy;ωLGz]°/s。
In order to avoid the sudden change problem of angular acceleration, realize the stable change of the moment of the flywheel and simultaneously consider the rapidity so as to expect the rotation angle theta
QAngular acceleration limit α
LGAngular velocity limit ω
LGAs input, an eight-segment attitude planner with continuous angular acceleration is designed, and the angular acceleration generates a function
The definition is as follows:
wherein, Δ t
AThe rising time of the angular acceleration is limited for the set value and can be reasonably selected according to the dynamic performance of the satellite actuator. Δ t
B,Δt
CValue of (a) to the desired angle theta
QIs related to the magnitude of,. DELTA.t
B,Δt
C,
The specific calculation process is as follows:
(1) when the desired angle of rotation is achieved
Time, plan angular acceleration
Reaches a maximum value of
LGPlanning angular velocity
Is as maximum asTo omega
LG。
(2) When in use
Time, plan angular acceleration
Reaches a maximum value of
LGPlanning angular velocity
Has a maximum value of not more than ω
LG。Δt
BBy a quadratic equation of unity
The solution is obtained by solving the above-mentioned problems,
(3) when in use
Time, plan angular acceleration
Has a maximum value of not more than alpha
LGPlanning angular velocity
Has a maximum value of not more than ω
LG. Maneuvering angle θ
QThe corresponding time is 4 delta t
A。Δt
B=0,Δt
C=0,
By a manoeuvre angle theta
QAngular acceleration limit of 60 DEGValue alpha
LG=0.1161°/s
2Angular velocity limit ω
LG=1.5°/s,Δt
AGenerated by the attitude planner as an input for 5s
It is briefly described as
The planned angular velocities and angles are shown in fig. 4.
The angular acceleration is totally 8 segments, and is divided into an ascending segment 2 segment, a stable segment 4 segment and a descending segment 2 segment. Wherein, the ascending section 1, the stable section 1 and the descending section 1 of the angular acceleration correspond to the ascending section of the angular velocity; the stationary segment 2 of angular acceleration corresponds to the stationary segment of angular velocity; an ascending section 2, a stationary section 3 and a descending section 2 of angular acceleration correspond to a descending section of angular velocity; the value of the stationary segment 4 of the angular acceleration is zero, the value of the corresponding angular velocity is also zero, and the angle value reaches the desired angle.
Real-time planning angles theta ∈ [0, theta ] generated by attitude planner
Q]Angular velocity
And angular acceleration
Solving to obtain a planning quaternion q
G=[cosθ;sinθe
n]Three-axis planning angular velocity of body system
Three-axis planned angular acceleration
Step three: fast maneuver control
The equations for the dynamics and kinematics of a rigid body satellite are described as:
wherein u is the control moment, S (-) is an antisymmetric matrix,
in maneuvering control of a satellite according to the present invention, a conversion map of attitude and angular velocity in a plurality of corresponding coordinate systems is shown in fig. 5; deviation angular velocity ωEAnd deviation quaternion qEThe calculation is as follows:
rotational quaternion of the initial attitude of the orbital system relative to the satellite
Wherein, the orbital quaternion q
guiIs a rotation quaternion of the orbital system relative to the inertial system;
the rotational quaternion, i.e. the deviation quaternion, of the real-time attitude of the satellite relative to the planned attitude
The quaternion q under the satellite inertial system is a rotation quaternion of the satellite body coordinate system relative to the inertial system;
deviation angular velocity, which is a representation of the deviation of the real-time angular velocity of the satellite from the planned angular velocity in the inertial system
Wherein the track angular velocity omega
guiThe satellite angular velocity ω is a rotational angular velocity of the satellite body system relative to the inertial system.
R(q
E) Is q
EThe corresponding rotation matrix is then used to determine,
is composed of
A corresponding rotation matrix.
In order to further improve the rapidity of satellite maneuvering, a feedforward design is added on the basis of PD control, and a controller is designed as follows:
u=KqαG-KPqE-KdωE
wherein, Kq,KP,KdRespectively a feedforward control gain matrix, a proportional control increment matrix and a differential control gain matrix, Kq=KqI,KP=KPI,Kd=KdI,Kq,KP,KdGain matrix coefficients greater than 0.
In a second embodiment, the present embodiment is described with reference to fig. 6 to 10, where the first embodiment is a method for performing simulation verification by using an in-orbit autonomous three-axis fast maneuver control method for a satellite, and a verification result is compared with a conventional PD control scheme without path planning, and the satellite and control parameters of the present embodiment are selected as follows: moment of inertia of satellite
Flywheel angular momentum H ═ 0.01; 0.01; 0.01]N.m.s; flywheel torque T ═ 0.003; 0.003; 0.003]N.m; input angular velocity limit ω
Lim=[1.2;1.3;1.1](ii) DEG/s; feedforward control gain matrix coefficient K
q0.75; coefficient K of proportional control gain matrix
p1.55; coefficient K of differential control gain matrix
d1.5; initial attitude quaternion of satellite is q
C=[1;0;0;0](ii) a The target is imaged to the ground, coinciding with the orbital system, and the target attitude is q
F=q
guiInitial angular velocity ω
C=[0;0;0]°/s。
In the PD control scheme, u ═ K
P1q
ed-K
D1ω
ed,
ω
ed=ω-R(q
E)ω
gui,K
P1=K
P1I,K
d1=K
d1I, proportional control gain matrix coefficient K
p10.12; coefficient K of differential control gain matrix
d1=0.58。
The desired rotation angle from the initial attitude to the target attitude is 100.74 deg., and the angle curve is planned from the three-axis attitude as shown in fig. 6. The three-axis attitude angles of the PD control scheme, the three-axis attitude angles corresponding to the three-axis attitude planning angle of the present embodiment, and the three-axis attitude angles of the post-planning feed-forward control scheme of the present embodiment (euler angles obtained by rotating quaternion q in the ZYX order) are shown in fig. 7, and correspond to those shown in the PD control, expected planning, and planning control labels, respectively. The corresponding angular velocities are shown in fig. 8. The deviation angle and the deviation angular velocity of the present embodiment are shown in fig. 9 and 10. As can be seen from fig. 7 and 8, the time required for the method according to the present embodiment is shorter for the same angle of rotation.