A two-stage control method, system, and medium for reusable rocket-powered soft landing
By dividing the rocket's powered soft landing process into two stages and using convex optimization algorithms to handle non-convex constraints, the rocket attitude deviation problem was solved, enabling the rocket to land vertically or nearly vertically, thus improving landing reliability and fuel utilization efficiency.
Patent Information
- Application Number
- CN202510786678.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-13
- Publication Date
- 2025-10-31
- Estimated Expiration
- 2045-06-13
AI Technical Summary
Existing powered soft landing algorithms cannot effectively account for changes in rocket attitude, leading to deviations in terminal attitude and affecting the success rate and safety of landing.
The rocket's powered soft landing process is divided into two stages, and optimal control problems are established for each stage. By introducing relaxation variables and variable substitution, non-convex constraints are transformed into convex constraints. Convex optimization algorithms are used to solve the optimization problems to ensure that the rocket lands in a vertical or near-vertical state.
It achieved precise vertical or near-vertical landing of the rocket, improving landing reliability and safety, while optimizing fuel consumption.
Smart Images
Figure CN120295345B_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of rocket guidance technology, and in particular to a two-stage control method, system and medium for a reusable rocket's powered soft landing. Background Technology
[0002] Powered soft landing is one of the core components of reusable rocket technology. During the powered descent phase, by controlling the engine thrust, the rocket can decelerate and accurately locate the target landing point when returning to Earth. Compared with the traditional parachute landing method, powered soft landing has higher precision and flexibility, and can adapt to various complex terrain conditions, thereby ensuring the safe recovery of the rocket.
[0003] Most current powered descent guidance algorithms typically simplify the rocket into a point mass model and establish dynamic equations based on this model to construct an optimal control problem with fuel efficiency as the objective. However, this point mass model ignores the rocket's attitude change characteristics, resulting in the ineffective consideration of terminal attitude constraints. In practice, to ensure a safe landing, the terminal attitude needs to remain vertical or near-vertical. However, existing algorithms often fail to meet this requirement and may even lead to significant attitude deviations, affecting the landing success rate and safety. Summary of the Invention
[0004] To address the above problems, this invention aims to provide a two-stage control method for reusable rocket-powered soft landing, comprising the following steps:
[0005] Step S1: Establish the dynamic model for the first stage, introduce constraints for powered descent and landing, and control the rocket's altitude directly above the landing point to be... The location of the reusable rocket is considered, and an optimization problem P0 is established to minimize fuel consumption.
[0006] Step S2: Transform the non-convex constraints in optimization problem P0 into convex constraints, and establish optimization problem P1;
[0007] Step S3: Introduce variable substitution for P1 to obtain the first-stage convex optimization problem P2;
[0008] Step S4: Based on the displacement and velocity constraints of the first stage, control the reusable rocket vertically to the landing point, and establish a convex optimization problem P3 considering the fuel-efficient operation of the reusable rocket.
[0009] Step S5: Use a convex optimization algorithm to solve convex optimization problems P2 and P3.
[0010] Based on the above scheme, the displacement constraint of the rocket at the end of the first stage is: , This represents the end displacement of the first stage.
[0011] Based on the above scheme, the initial height of the second stage The calculation method is as follows:
[0012] Using the initial mass of the reusable rocket as a reference and the maximum thrust as the vertical force of the rocket, the maximum acceleration during the deceleration and landing phase is obtained.
[0013] Based on the speed constraint of the rocket's descent This gives the upper bound of the rocket's vertical descent time.
[0014] Based on the calculated upper bound of the time and maximum acceleration of the rocket's powered landing phase, the displacement of the vertical descent phase was calculated. .
[0015] Based on the above scheme, the initial displacement in the second stage of step S4 initial velocity , This represents the end displacement of the first stage. This represents the terminal velocity of the first stage.
[0016] Based on the above scheme, the establishment of the optimization problem P0 includes:
[0017] ;
[0018] in, For thrust, For the displacement of the rocket, Let g be the rocket's velocity, g be the acceleration due to gravity, and ω be the Earth's angular velocity due to rotation. For the mass of the rocket, The cross product of the antisymmetric matrix representation of the Earth's rotational angular velocity. rocket burn rate , Earth's standard gravitational acceleration , This refers to the specific impulse of the rocket engine. For the range of thrust amplitude, It is a unit vector in the vertical direction. This indicates the maximum permissible angle between the rocket's longitudinal axis and the vertical direction. This means that the rocket's powered landing trajectory must be contained within a conical region with the landing point as its vertex. , , These are displacements in the x, y, and z directions, respectively. Indicates the semi-apex angle of the conical region. The maximum speed of the rocket, For the dry weight of the rocket, This represents the weight of the rocket at the end of the first stage. This represents the weight of the rocket at the end of the first stage. This means that the weight at the end of the first stage must not be less than the dry weight of the rocket. These represent the initial displacement, initial velocity, and initial weight of the rocket at the start of its first stage. The displacement constraints at the end of the first stage. This indicates the speed constraint at the end of the first phase.
[0019] Based on the above scheme, the convex optimization problem P2 is established as follows:
[0020] By introducing the slack variable σ into the optimization problem P0, the non-convex constraint is transformed into a convex constraint, resulting in P1;
[0021] Introducing variable substitution: , , And using Taylor series to pair Perform an approximate expansion:
[0022] ;
[0023] Here, the reference trajectory z0(t) represents the maximum fuel consumption rate, so z0(t) is a lower bound of z(t);
[0024] Add constraints:
[0025] ;
[0026] The final convex optimization problem P2 that needs to be solved is obtained.
[0027] Based on the same inventive concept, this application discloses a two-stage control system for reusable rocket-powered soft landing, comprising:
[0028] The first-stage control module is used to establish the dynamic model for the first stage, introduce constraints for powered descent and landing, and control the rocket's altitude directly above the landing point. The first-stage control module includes the following positions:
[0029] The first optimization module is used to establish an optimization problem P0 that considers the fuel-saving aspect of reusable rockets.
[0030] The second optimization module transforms the non-convex constraints in optimization problem P0 into convex constraints, and establishes optimization problem P1.
[0031] The third optimization module introduces variable substitution for P1 to obtain the first-stage convex optimization problem P2;
[0032] The second-stage control module is used to control the reusable rocket vertically to the landing point based on the displacement and velocity constraints of the first stage, and to establish a convex optimization problem P3 considering the fuel-efficient operation of the reusable rocket.
[0033] The computation module is used to solve convex optimization problems P2 and P3 using convex optimization algorithms.
[0034] Based on the above scheme, the control height of the first-stage control module As the initial displacement constraints for the second stage, the convex optimization problem for the second stage is established.
[0035] Based on the above scheme, an initial altitude calculation submodule is included to calculate the altitude at the end of the first stage based on the rocket's initial mass and maximum thrust. The initial altitude calculation submodule includes:
[0036] The first calculation unit is used to obtain the maximum acceleration of the deceleration and landing phase based on the initial mass of the reusable rocket and the maximum thrust as the vertical force of the rocket.
[0037] The second calculation unit is used to determine the rocket's descent speed constraint. This gives the upper bound of the rocket's vertical descent time.
[0038] The third calculation unit is used to calculate the displacement of the vertical descent phase based on the calculated upper time limit and maximum acceleration of the rocket's powered landing phase. .
[0039] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the two-stage control method for reusable rocket-powered soft landing as described above.
[0040] Compared with the prior art, the present invention has the following beneficial effects:
[0041] By dividing the rocket's powered soft landing process into two stages and establishing an optimal control problem for each stage, the rocket's descent trajectory can be controlled more precisely.
[0042] First, the rocket is positioned directly above the landing site, and then it descends vertically to the landing site. This takes into account the constraints of the rocket's terminal attitude, ensuring that the rocket's attitude is vertical or nearly vertical during landing, thereby increasing the reliability of vertical landing. Attached Figure Description
[0043] Figure 1 This is a schematic diagram of the two-stage soft landing process of this application;
[0044] Figure 2 This is a flowchart of the two-stage soft landing control method of this application;
[0045] Figure 3 This is a comparison diagram of the trajectory simulation of this application;
[0046] Figure 4 This is a speed simulation comparison curve of this application;
[0047] Figure 5 This is a simulation comparison diagram of rocket mass changes in this application;
[0048] Figure 6 This is a simulation comparison diagram of the rocket tilt angle changes in this application;
[0049] Figure 7 This is a comparison diagram of the simulated trajectory within the rocket's launch surface in this application. Detailed Implementation
[0050] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.
[0051] This invention provides a two-stage control method for reusable rocket-powered soft landing. This method can solve the problem of vertical deviation in terminal attitude by dividing the reusable rocket-powered soft landing phase into two stages, such as... Figure 1 As shown, the first stage controls the rocket to a certain height directly above the landing point, from point A to point C; the second stage then controls the rocket vertically to the landing point, from point C to point D. Then, based on the different constraints of the two stages, a fuel-efficient optimal control problem is established for each stage. For example... Figure 2 As shown, the specific steps are as follows:
[0052] Step S1: Establish the first-stage dynamic model, considering the three-degree-of-freedom fuel-efficient powered descent guidance (PDG) problem P0 for a reusable rocket, introducing constraints for powered descent landing, and controlling the rocket's altitude directly above the landing point to be... The position of the rocket's point mass dynamics corresponds to a double integrator with variable mass moving in a constant gravitational field with reference to the Earth's coordinate system, as described below;
[0053] ;
[0054] in, For thrust, For the displacement of the rocket, Let g be the rocket's velocity, g be the acceleration due to gravity, and ω be the Earth's angular velocity due to rotation. , , g and ω∈ , For the mass of the rocket, The cross product of the antisymmetric matrix representation of the Earth's rotational angular velocity. rocket burn rate , Earth's standard gravitational acceleration , This refers to the specific impulse of the rocket engine. For the range of thrust amplitude, This means that the angle between the rocket's longitudinal axis and the vertical direction cannot be greater than [the angle between the rocket's longitudinal axis and the vertical direction]. , It is a unit vector in the vertical direction. This indicates the maximum permissible angle between the rocket's longitudinal axis and the vertical direction. This means that the rocket's powered landing trajectory must be contained within a conical region with the landing point as its vertex. , , These are displacements in the x, y, and z directions, respectively. Indicates the semi-apex angle of the conical region. The maximum speed of the rocket. This indicates that the rocket's speed does not exceed , This means that the weight at the end of the first stage must not be less than the dry weight of the rocket. This indicates the initial state of the rocket at the start of its first stage. This indicates the displacement constraints of the rocket at the end of the first stage. This indicates the rocket's velocity constraint at the end of the first stage.
[0055] Step S2, in order to handle the non-convex constraints of rocket thrust Introducing slack variables Transform the non-convex constraint into a convex constraint. The transformed optimization problem P1 can be expressed as:
[0056] .
[0057] Step S3, for P1, introduce variable substitution , , Meanwhile, using Taylor series pairs By approximate expansion, we get:
[0058] ;
[0059] Among them, the reference trajectory Represents the maximum fuel consumption rate, therefore yes The lower bound.
[0060] To ensure To prevent the physical constraints from being violated, additional constraints need to be added:
[0061] ;
[0062] In summary, through the above variable substitutions, we can obtain the convex optimization problem P2 that needs to be solved in the first stage:
[0063] .
[0064] Step S4: After the first stage ends, the rocket's terminal state is the initial state of the second stage. The optimization objectives of the two stages are consistent, but the terminal displacement and velocity constraints have changed, while other constraints remain unchanged. The second stage's three-degree-of-freedom fuel-efficient PDG problem P3 for the reusable rocket can be expressed as:
[0065] ;
[0066] in, This is the end displacement of the first stage, which is also the initial height of the second stage. It is the terminal velocity of the first stage, which satisfies... .
[0067] This application, through extensive numerical simulations, reveals that although the rocket's mass continuously decreases during its reusable phase, the change in mass is minimal due to the relatively short duration of the powered landing phase, thus having little impact on the acceleration during deceleration. Therefore, The calculation method is as follows:
[0068] Using the initial mass of the reusable rocket as a reference and the maximum thrust as the vertical force of the rocket, the maximum acceleration during the deceleration and landing phase is obtained. Note that the gravitational acceleration value needs to be subtracted during the calculation.
[0069] Based on the speed constraint of the rocket's descent So if the rocket is at maximum speed If a descent can result in a safe landing, other speeds are also acceptable, thus obtaining the upper bound of the rocket's vertical descent phase time;
[0070] Based on the calculated upper bound of the time and maximum acceleration of the rocket's powered landing phase, the displacement of the vertical descent phase was calculated. ;
[0071] The calculations obtained using the above methods This can ensure that reusable rockets are in At a height less than or equal to the maximum speed It can safely land at any speed to a recoverable landing site.
[0072] According to the present invention, the initial altitude calculation method can be adapted to different rocket models and mission requirements. The calculated values can be adjusted according to the rocket model and mission requirements to ensure that the rocket's terminal attitude is vertically upward, thus meeting the actual application scenarios of reusable rockets.
[0073] Step S5: Based on the above steps, the fuel-saving problem can be transformed into a convex optimization problem. Therefore, existing mature convex optimization algorithms can be used to solve the convex optimization problems P2 and P3, which will not be elaborated here.
[0074] To verify the feasibility of the method in this application, a rocket powered descent simulation mission was selected for verification. The parameters of the reusable rocket were selected as follows: , , , , , , , , The rocket's initial state is set as follows: , .
[0075] Based on the above parameters, the end displacement of the first stage can be obtained, which is the initial height of the second stage. for: .
[0076] Based on the transformed convex optimization problem and parameters, a convex optimization toolkit is used to solve it, enabling control of a reusable rocket. The results are compared with existing single-stage convex optimization algorithms. Figures 3 to 7 As shown, both lossless convex optimization algorithms can adjust the rocket's center of mass to the recovery point, and the rocket's terminal velocity is zero, thus achieving a soft landing.
[0077] The comparison shows that the existing lossless convex optimization results in a tilted rocket terminal attitude with an inclination angle of about 30°. However, the two-stage lossless convex optimization algorithm proposed in this invention can adjust the rocket terminal attitude to a vertical state with an inclination angle of 0°. First, the rocket moves to a position directly above the recovery point, and then lands vertically at the recovery point, ensuring that the rocket's terminal attitude is vertically upward.
[0078] This application reduces the impact load during landing by adjusting the terminal attitude, thereby protecting the structural integrity of the rocket and improving the success rate of recovery; at the same time, it controls the fuel optimally, achieving economy while ensuring attitude constraints.
[0079] Based on the same inventive concept, this application discloses a two-stage control system for reusable rocket-powered soft landing, comprising:
[0080] The first-stage control module is used to establish the dynamic model for the first stage, introduce constraints for powered descent and landing, and control the rocket's altitude directly above the landing point. The first-stage control module includes the following positions:
[0081] The first optimization module is used to establish an optimization problem P0 that considers the fuel-saving aspect of reusable rockets.
[0082] The second optimization module transforms the non-convex constraints in optimization problem P0 into convex constraints, and establishes optimization problem P1.
[0083] The third optimization module introduces variable substitution for P1 to obtain the first-stage convex optimization problem P2;
[0084] The second-stage control module is used to control the reusable rocket vertically to the landing point based on the displacement and velocity constraints of the first stage, and to establish a convex optimization problem P3 considering the fuel-efficient operation of the reusable rocket.
[0085] The computation module is used to solve convex optimization problems P2 and P3 using convex optimization algorithms.
[0086] The control level of the first-stage control module As the initial displacement constraints for the second stage, the convex optimization problem for the second stage is established.
[0087] The system also includes an initial altitude calculation submodule, which calculates the altitude at the end of the first stage based on the rocket's initial mass and maximum thrust. The initial altitude calculation submodule includes:
[0088] The first calculation unit is used to obtain the maximum acceleration of the deceleration and landing phase based on the initial mass of the reusable rocket and the maximum thrust as the vertical force of the rocket.
[0089] The second calculation unit is used to determine the rocket's descent speed constraint. This gives the upper bound of the rocket's vertical descent time.
[0090] The third calculation unit is used to calculate the displacement of the vertical descent phase based on the calculated upper time limit and maximum acceleration of the rocket's powered landing phase. .
[0091] According to the present invention, a computer-readable recording medium storing computer-executable instructions can be provided, which, when executed by a processor, causes the processor to perform the large language model training method as described above.
[0092] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or portion of code containing at least one executable instruction for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.
[0093] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
[0094] While the specific embodiments of the present invention have been described above, they are not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.
Claims
1. A two-stage control method for reusable rocket-powered soft landing, characterized in that, Includes the following steps: Step S1: Establish the dynamic model for the first stage, introduce constraints for powered descent and landing, and control the rocket's altitude directly above the landing point to be... The position is determined, and an optimization problem P0 is established considering the fuel-efficient operation of the reusable rocket; where the terminal velocity constraints of the reusable rocket at the end of the first stage include: the velocities in the x and z directions are 0; initial height r yf The calculation method is as follows: Using the initial mass of the reusable rocket as a reference and the maximum thrust as the vertical force of the rocket, the maximum acceleration during the deceleration and landing phase is obtained. According to the rocket's descent velocity constraint ||v(t)||2 ≤ v max This gives the upper bound of the rocket's vertical descent time. Based on the calculated upper bound of the time and maximum acceleration of the rocket's powered landing phase, the displacement r of the vertical descent phase is calculated. yf ; Step S2: Transform the non-convex constraints in optimization problem P0 into convex constraints, and establish optimization problem P1; Step S3: Introduce variable substitution in P1 to obtain the first-stage convex optimization problem P2; Step S4: Based on the terminal displacement and terminal velocity constraints of the first-stage reusable rocket, control the second-stage reusable rocket to move vertically to the landing point, and establish a convex optimization problem P3 considering the fuel-efficient operation of the reusable rocket. Step S5: Use a convex optimization algorithm to solve convex optimization problems P2 and P3.
2. The two-stage control method for reusable rocket-powered soft landing according to claim 1, characterized in that, The displacement constraint of the rocket at the end of the first stage is: r yf It is the end displacement of the first stage and the initial displacement of the second stage.
3. The two-stage control method for reusable rocket-powered soft landing according to claim 2, characterized in that, In step S4, the initial displacement r(0) of the second stage is r yf Initial velocity v(0) = v yf r yf v is the end displacement of the first stage. yf This represents the terminal velocity of the first stage.
4. The two-stage control method for a reusable rocket-powered soft landing according to claim 1, characterized in that, The establishment of the optimization problem P0 includes: ； in, For thrust, For the displacement of the rocket, Let g be the velocity of the rocket, and g be the acceleration due to gravity. ω This is the Earth's rotational angular velocity. For the mass of the rocket, The cross product of the antisymmetric matrix representation of the Earth's rotational angular velocity. rocket burn rate , Earth's standard gravitational acceleration , This refers to the specific impulse of the rocket engine. For the range of thrust amplitude, It is a unit vector in the vertical direction. This indicates the maximum permissible angle between the rocket's longitudinal axis and the vertical direction. This means that the rocket's powered landing trajectory must be contained within a conical region with the landing point as its vertex. , , These are displacements in the x, y, and z directions, respectively. Indicates the semi-apex angle of the conical region. The maximum speed of the rocket, For the dry weight of the rocket, This represents the weight of the rocket at the end of the first stage. This means that the weight at the end of the first stage must not be less than the dry weight of the rocket. These represent the initial displacement, initial velocity, and initial weight of the rocket at the start of its first stage. The displacement constraints at the end of the first stage. This indicates the speed constraint at the end of the first phase.
5. The two-stage control method for reusable rocket-powered soft landing according to claim 4, characterized in that, The convex optimization problem P2 is established as follows: Introduce slack variables into the optimization problem P0. σ ( t ) replace || T c ( t )||2, transforming the non-convex constraint into a convex constraint, resulting in P1; Introducing variable substitution: , , And using Taylor series to pair Perform an approximate expansion: ； in, z 0( t z0(t) represents the minimum mass logarithmic trajectory corresponding to the maximum fuel consumption rate, therefore z0(t) is a lower bound of z(t); Add constraints: ； The final convex optimization problem P2 that needs to be solved is obtained.
6. A two-stage control system for reusable rocket-powered soft landing, characterized in that, include: The first-stage control module is used to establish the dynamic model for the first stage, introduce constraints for powered descent and landing, and control the rocket's altitude directly above the landing point. The position, in which the terminal velocity constraints of the reusable rocket at the end of the first stage include: the velocity in the x and z directions is 0; The first stage control module includes: The first optimization module is used to establish an optimization problem P0 that considers the fuel-saving aspect of reusable rockets. The second optimization module transforms the non-convex constraints in optimization problem P0 into convex constraints, and establishes optimization problem P1. The third optimization module introduces variable substitution in P1 to obtain the first-stage convex optimization problem P2; The second-stage control module is used to control the reusable rocket to move vertically to the landing point based on the end displacement and end velocity constraints of the first stage. It also considers the fuel-efficient operation of the reusable rocket to establish a convex optimization problem P3. The computation module is used to solve convex optimization problems P2 and P3 using convex optimization algorithms; The system includes an initial altitude calculation submodule, used to calculate the altitude at the end of the first stage based on the rocket's initial mass and maximum thrust. The initial altitude calculation submodule includes: The first calculation unit is used to obtain the maximum acceleration of the deceleration and landing phase based on the initial mass of the reusable rocket and the maximum thrust as the vertical force of the rocket. The second calculation unit is used to determine the rocket's descent speed constraint. This gives the upper bound of the rocket's vertical descent time. The third calculation unit is used to calculate the displacement of the vertical descent phase based on the calculated upper time limit and maximum acceleration of the rocket's powered landing phase. .
7. A two-stage control system for reusable rocket-powered soft landing according to claim 6, characterized in that, The control height of the first stage control module As the initial displacement constraints for the second stage, the convex optimization problem for the second stage is established.
8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the two-stage control method for reusable rocket-powered soft landing as described in any one of claims 1 to 5.