CN114036835A - Virtual simulation method and system for training skill of on-orbit filling operation - Google Patents
Virtual simulation method and system for training skill of on-orbit filling operation Download PDFInfo
- Publication number
- CN114036835A CN114036835A CN202111309798.4A CN202111309798A CN114036835A CN 114036835 A CN114036835 A CN 114036835A CN 202111309798 A CN202111309798 A CN 202111309798A CN 114036835 A CN114036835 A CN 114036835A
- Authority
- CN
- China
- Prior art keywords
- training
- task
- model
- simulation
- orbit
- 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.)
- Withdrawn
Links
- 238000012549 training Methods 0.000 title claims abstract description 99
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000011156 evaluation Methods 0.000 claims abstract description 29
- 238000004088 simulation Methods 0.000 claims abstract description 27
- 238000012360 testing method Methods 0.000 claims abstract description 20
- 238000013508 migration Methods 0.000 claims abstract description 6
- 230000005012 migration Effects 0.000 claims abstract description 6
- 238000012795 verification Methods 0.000 claims abstract description 5
- 230000005284 excitation Effects 0.000 claims abstract description 4
- 238000005457 optimization Methods 0.000 claims abstract description 4
- 238000013439 planning Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 230000006870 function Effects 0.000 claims description 5
- 238000007726 management method Methods 0.000 claims description 5
- 230000006399 behavior Effects 0.000 claims description 3
- 238000013210 evaluation model Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 239000003380 propellant Substances 0.000 claims description 3
- 230000002787 reinforcement Effects 0.000 claims description 3
- 238000012384 transportation and delivery Methods 0.000 claims description 2
- 230000037452 priming Effects 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a virtual simulation method for training skill of on-orbit filling operation, which comprises the following steps: step 1, building a simulation environment the same as a real physical filling environment; step 2, issuing an on-orbit filling training task; step 3, performing related operation training through an operation interface of system simulation; step 4, constructing an efficiency evaluation network, evaluating the effectiveness of the model output in the step 3, and outputting an optimal training model; orderly superposing all operation skill optimal training models, then carrying out parameter fitting optimization, introducing an excitation function, constructing a whole on-orbit filling task model, carrying out training verification and evaluation until the parameters of the task model are converged, and outputting the task model; and (3) migrating the output task model to an on-orbit filling test platform for testing by a virtual-real migration method, outputting the optimal task training model if the test result is qualified, and returning to the step 2 if the test result is unqualified. The invention can be directly applied to the implementation of actual tasks.
Description
Technical Field
The invention belongs to the technical field of space, and particularly relates to an on-orbit filling operation skill training virtual simulation method and system.
Background
With the development of the on-orbit service with more and higher demands, the on-orbit refueling service becomes an important development direction and is a lead and basis for driving other on-orbit service technologies. The problems of high difficulty, high cost and the like of directly implementing the on-orbit filling operation task at present highlight the important value of the on-orbit filling operation skill training virtual simulation technology. In the face of a space environment with multiple interference sources and large uncertainty, the on-orbit filling by utilizing the virtual simulation technology has stronger flexibility and robustness, and the existing new technology can be well utilized to simulate and train an autonomous filling task, and the new technology can be well tried and found in multiple aspects, so that an optimal model training result and a task implementation scheme are obtained. At present, aiming at an on-orbit filling operation skill virtual simulation system, a complete and mature training evaluation simulation system which has a virtual-real migration function and is used for independently training and evaluating a result for each operation skill is not provided for carrying out an on-orbit filling task. And, each verification project is basically performed depending on the space station. If the operation is performed separately and dispersedly, the time is greatly consumed and multiple space station resources are occupied, so a complete and efficient virtual simulation system and method for training the on-orbit filling operation skills are urgently needed.
Disclosure of Invention
Compared with manual intervention and judgment aiming at the state of the whole task, the autonomy and robustness are improved, meanwhile, the system has strong virtual and real migration capability, and the virtual simulation training model result can be directly used for implementing the actual task.
The invention adopts a technical scheme for solving the technical problems in the prior art, which is as follows: an on-orbit filling operation skill training virtual simulation method comprises the following steps:
step 1, building a simulation environment the same as a real physical filling environment, simulating track parameters, service star and target star information and spatial natural environment information of on-orbit filling, acquiring target star information and interaction data of a target star and the target star through the service star, and acquiring behavior data of the filling environment and the target star by the service star;
step 2, analyzing the current training task requirement according to the simulation environment simulation information and the acquired data, performing task planning, issuing a command, and issuing an on-orbit filling training task according to the command;
step 3, performing related operation training through an operation interface of system simulation, decomposing the on-orbit filling task into a plurality of operable operation links according to procedures, constructing an ordered operation training network, establishing corresponding operation skill training models aiming at different operation links, and performing operation skill training by respectively adopting a reinforcement learning method; during training, according to received instructions, making corresponding operation commands, performing each training according to the sequence of situation awareness, command decision, task training and effect evaluation, adopting data collected by the service stars in the step 1, training the system according to specific operation skills, calling an environment, selecting tools and an inference rule set, selecting n tools from a tool box according to the skill requirements, and performing single equipment operation training by the system when n is 1; when n is larger than 1, the system performs various equipment operation training; the system simulates and deduces various on-orbit filling operation skill training until the parameters of each operation skill training model are converged and then outputs the model and a simulation deduction result;
step 4, constructing an efficiency evaluation network, analyzing and comparing the on-orbit filling equipment consumption, the target state change and the change of the overall space environment situation before and after the operation skill training according to the simulation deduction result of the step 3, evaluating the effectiveness of the output model of the step 3 according to the evaluation model, and outputting an optimal training model; orderly superposing all operation skill optimal training models, then carrying out parameter fitting optimization, introducing an excitation function, constructing a whole on-orbit filling task model, carrying out training verification and evaluation until the parameters of the task model are converged, and outputting the task model; migrating the output task model to an on-orbit filling test platform for testing by a virtual-real migration method, and if the test result is qualified, outputting the optimal task training model and ending the task; and if the test is unqualified, outputting an unqualified operation link, returning to the step 2, planning and training the task again aiming at the unqualified link until the test result is qualified, outputting a final integral task training model, and ending the task.
The invention adopts another technical scheme for solving the technical problems in the prior art, which is as follows: the on-orbit filling operation skill training virtual simulation system adopting the method comprises a space target and environment virtual simulation subsystem, a service satellite virtual simulation subsystem, an operation skill training management and evaluation subsystem and a guidance control and efficiency evaluation subsystem.
The service satellite virtual simulation subsystem comprises a terminal tool box, a propellant conveying system, a smart mechanical arm and an autonomous real-time relative navigation system.
The invention has the advantages and positive effects that: the method of dividing the whole task into multi-operation skill training evaluation and combining with an artificial intelligence algorithm is adopted, so that unified expression and learning training of various operation skill strategy networks are realized, and compared with manual intervention and judgment aiming at the state of the whole task, the autonomy and the robustness are improved; the method supports real-time task training control, data recording and playback, and outputs three-dimensional battlefield situation display, training models and evaluation data; and joint test with third-party equipment is supported.
Drawings
FIG. 1 is a flowchart illustrating the operation of the virtual simulation method of the present invention;
FIG. 2 is a flowchart of step 3 of the virtual simulation method of the present invention;
FIG. 3 is an architecture diagram of the virtual simulation system of the present invention;
FIG. 4 is a block diagram of a virtual simulation system according to the present invention.
Detailed Description
In order to further understand the contents, features and effects of the present invention, the following embodiments are illustrated and described in detail with reference to the accompanying drawings:
referring to fig. 1 to 2, a virtual simulation method for training skill of on-orbit filling operation includes the following steps:
step 1, in an environment and filling simulation stage, according to environmental elements such as optics and electromagnetism faced by an on-orbit filling service satellite and the maneuvering characteristics of the on-orbit filling service satellite, a simulation environment identical to a real physical filling environment is built, orbit parameters, service satellite and target satellite information and space natural environment information of on-orbit filling are simulated, target satellite information and interactive data of the target satellite and the target satellite are collected through the service satellite, and the service satellite is used for collecting filling environment and behavior data of the target satellite.
And 2, in a command control stage, analyzing the current training task requirement according to the simulation environment simulation information and the acquired data, planning a task, issuing a command, and issuing an on-orbit filling training task according to the command.
Step 3, in a decision deduction stage, performing related operation training through an operation interface of system simulation, decomposing an on-orbit filling task into a plurality of operational links according to procedures, constructing an ordered operation training network, establishing corresponding operation skill training models aiming at different operational links, and performing operation skill training by respectively adopting a reinforcement learning method; during training, according to received instructions, making corresponding operation commands, performing each training according to the sequence of situation awareness, command decision, task training and effect evaluation, adopting data collected by the service stars in the step 1, training the system according to specific operation skills, calling an environment, selecting tools and an inference rule set, selecting n tools from a tool box according to the skill requirements, and performing single equipment operation training by the system when n is 1; when n is larger than 1, the system performs various equipment operation training; and (3) simulating and deducing various on-orbit filling operation skill training by the system until the parameters of each operation skill training model are converged, and outputting the model and a simulation deduction result.
Step 4, in the efficiency evaluation stage, an efficiency evaluation network is constructed, the on-orbit filling equipment consumption, the target state change and the change of the overall space environment situation before and after the operation skill training are analyzed and compared according to the simulation deduction result in the step 3, the effectiveness of the output model in the step 3 is evaluated according to the evaluation model, and the optimal training model is output; orderly superposing all operation skill optimal training models, then carrying out parameter fitting optimization, introducing an excitation function, constructing a whole on-orbit filling task model, carrying out training verification and evaluation until the parameters of the task model are converged, and outputting the task model; migrating the obtained task model to an on-orbit filling test platform for testing by a virtual-real migration method, outputting an optimal task training model if a test result is qualified, and ending the task; and if the test is unqualified, outputting an unqualified operation link, returning to the step 2, planning and training the task again aiming at the unqualified link until the test result is qualified, outputting a final integral task training model, and ending the task.
Referring to fig. 3 to 4, an on-orbit filling operation skill training virtual simulation system adopting the method includes a space target and environment virtual simulation subsystem, a service satellite virtual simulation subsystem, an operation skill training management and evaluation subsystem and a guidance and performance evaluation subsystem.
The space target and environment simulation subsystem is responsible for the simulation of the space target and the environment and reproducing the operation process of the space target;
the service satellite virtual simulation subsystem is divided into 4 virtual simulation subsystems according to the main composition of a service satellite, and each type of simulation subsystem is responsible for simulation, parameterization customization and state feedback of the virtual simulation subsystem; the service satellite is provided with two mechanical arms, and interaction data of the service satellite, the target satellite and the virtual environment are collected through the sensors so as to present an on-orbit filling working state and a task progress state.
The operation skill training management and evaluation subsystem is responsible for task planning and management of on-orbit filling operation skill training, simulation of command scheduling, parameterization customization and system efficiency evaluation, and selects equipment resources, planning tasks, execution tasks and single operation skill training evaluation tasks which can meet requirements based on different on-orbit filling operation skills according to attributes of the different operation skills; the system intrinsic ability indexes such as space coverage, operation time interval, operation duration and the like can be evaluated, and the system dynamic ability indexes such as tool use accuracy, target accuracy, operation skill completion effect and the like can also be evaluated; task evaluation efficiency can be output based on user-defined areas, targets, time, information systems and precision requirements, and a single operation skill training model and evaluation efficiency output are supported.
The guidance control and efficiency evaluation subsystem provides functions required by operation guidance control, comprises user setting, authority allocation, seat configuration and task setting, and has comprehensive efficiency evaluation capability on-orbit filling tasks. When the training of a single skill is finished, the whole task is trained and evaluated according to the filling task flow, and the influence of continuous operation skills on the whole filling task is tested; and the whole filling task training model and the output of the efficiency evaluation result are supported.
In this embodiment, the service satellite virtual simulation subsystem includes a terminal toolbox, a propellant delivery system, a smart robotic arm, and an autonomous, real-time relative navigation system.
Although the preferred embodiments of the present invention have been described above with reference to the accompanying drawings, the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and those skilled in the art can make many modifications without departing from the spirit and scope of the present invention as defined in the appended claims.
Claims (3)
1. A virtual simulation method for training skill of on-orbit filling operation is characterized by comprising the following steps:
step 1, building a simulation environment the same as a real physical filling environment, simulating track parameters, service star and target star information and spatial natural environment information of on-orbit filling, acquiring target star information and interaction data of a target star and the target star through the service star, and acquiring behavior data of the filling environment and the target star by the service star;
step 2, analyzing the current training task requirement according to the simulation environment simulation information and the acquired data, performing task planning, issuing a command, and issuing an on-orbit filling training task according to the command;
step 3, performing related operation training through an operation interface of system simulation, decomposing the on-orbit filling task into a plurality of operable operation links according to procedures, constructing an ordered operation training network, establishing corresponding operation skill training models aiming at different operation links, and performing operation skill training by respectively adopting a reinforcement learning method; during training, according to received instructions, making corresponding operation commands, performing each training according to the sequence of situation awareness, command decision, task training and effect evaluation, adopting data collected by the service stars in the step 1, training the system according to specific operation skills, calling an environment, selecting tools and an inference rule set, selecting n tools from a tool box according to the skill requirements, and performing single equipment operation training by the system when n is 1; when n is larger than 1, the system performs various equipment operation training; the system simulates and deduces various on-orbit filling operation skill training until the parameters of each operation skill training model are converged and then outputs the model and a simulation deduction result;
step 4, constructing an efficiency evaluation network, analyzing and comparing the on-orbit filling equipment consumption, the target state change and the change of the overall space environment situation before and after the operation skill training according to the simulation deduction result of the step 3, evaluating the effectiveness of the output model of the step 3 according to the evaluation model, and outputting an optimal training model; orderly superposing all operation skill optimal training models, then carrying out parameter fitting optimization, introducing an excitation function, constructing a whole on-orbit filling task model, carrying out training verification and evaluation until the parameters of the task model are converged, and outputting the task model; migrating the output task model to an on-orbit filling test platform for testing by a virtual-real migration method, and if the test result is qualified, outputting the optimal task training model and ending the task; and if the test is unqualified, outputting an unqualified operation link, returning to the step 2, planning and training the task again aiming at the unqualified link until the test result is qualified, outputting a final integral task training model, and ending the task.
2. An in-orbit priming operation skill training virtual simulation system adopting the method of claim 1, which is characterized by comprising a space target and environment virtual simulation subsystem, a service satellite virtual simulation subsystem, an operation skill training management and evaluation subsystem and a guidance and performance evaluation subsystem.
3. An in-orbit fueling operation skill training virtual simulation system according to claim 2, wherein the service satellite virtual simulation subsystem comprises a tip kit, a propellant delivery system, a smart robotic arm, and an autonomous, real-time relative navigation system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111309798.4A CN114036835A (en) | 2021-11-07 | 2021-11-07 | Virtual simulation method and system for training skill of on-orbit filling operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111309798.4A CN114036835A (en) | 2021-11-07 | 2021-11-07 | Virtual simulation method and system for training skill of on-orbit filling operation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114036835A true CN114036835A (en) | 2022-02-11 |
Family
ID=80136562
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111309798.4A Withdrawn CN114036835A (en) | 2021-11-07 | 2021-11-07 | Virtual simulation method and system for training skill of on-orbit filling operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114036835A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115240504A (en) * | 2022-09-21 | 2022-10-25 | 中海油能源发展股份有限公司采油服务分公司 | LNG filling simulation training system between ships |
-
2021
- 2021-11-07 CN CN202111309798.4A patent/CN114036835A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115240504A (en) * | 2022-09-21 | 2022-10-25 | 中海油能源发展股份有限公司采油服务分公司 | LNG filling simulation training system between ships |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104573182B (en) | A kind of design method for aircraft Multimode Control System | |
| CN112487668A (en) | Near-physical simulation integrated debugging method and system based on digital twin | |
| CN114415630B (en) | Comprehensive test platform and method for airplane management system | |
| CN112051750B (en) | Digital twin system of four-rotor unmanned aerial vehicle | |
| CN111931371A (en) | Application mode design method of multi-satellite cooperative ground verification system | |
| CN109765802B (en) | Virtual DCS simulation system of large phase modulator | |
| CN110929422B (en) | Robot cluster simulation method and device | |
| CN114036835A (en) | Virtual simulation method and system for training skill of on-orbit filling operation | |
| Xia et al. | Virtual comissioning of manufacturing system intelligent control | |
| Wiendahl et al. | Modeling and simulation of assembly systems | |
| Verma et al. | Survey of command execution systems for NASA spacecraft and robots | |
| CN115221799B (en) | Multi-branch-based unmanned cluster super-real-time deduction system and method | |
| CN116620327A (en) | Lane changing decision method for realizing automatic driving high-speed scene based on PPO and Lattice | |
| CN110502838A (en) | Spare parts management strategy optimization model based on emulation | |
| Majstorović et al. | Industry 4.0 in Serbia-state of development | |
| Jaafari et al. | A simulation model for life cycle project management | |
| CN110675724A (en) | Design method of robot gluing teaching system | |
| Ottjes et al. | Design and control of multi-AGV systems: reuse of simulation software | |
| CN117195599B (en) | Cloud-edge cooperative satellite service migration verification method and device | |
| Freund et al. | Application of automatic action planning for several work cells to the German ETS-VII space robotics experiments | |
| Schneider et al. | Integrating genetic algorithms and interactive simulations for airbase logistics planning | |
| Bocchino et al. | Industry Best Practices in Robotics Software Engineering | |
| CN117313439B (en) | Multi-scene multi-machine type unmanned aerial vehicle simulation system | |
| RU2816639C1 (en) | Method for creating controllers for controlling walking robots based on reinforcement learning | |
| Pisanich et al. | Mission simulation facility: Simulation support for autonomy development |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220211 |
|
| WW01 | Invention patent application withdrawn after publication |