CN116861241A - Space debris collision probability prediction method, system, terminal and storage medium based on artificial intelligence - Google Patents
Space debris collision probability prediction method, system, terminal and storage medium based on artificial intelligence Download PDFInfo
- Publication number
- CN116861241A CN116861241A CN202310848212.4A CN202310848212A CN116861241A CN 116861241 A CN116861241 A CN 116861241A CN 202310848212 A CN202310848212 A CN 202310848212A CN 116861241 A CN116861241 A CN 116861241A
- Authority
- CN
- China
- Prior art keywords
- information
- space debris
- data
- model
- collision probability
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 29
- 230000006978 adaptation Effects 0.000 claims abstract description 36
- 238000012549 training Methods 0.000 claims abstract description 18
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 36
- 238000012545 processing Methods 0.000 claims description 18
- 238000004140 cleaning Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 5
- 238000013139 quantization Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 abstract 2
- 239000012634 fragment Substances 0.000 description 12
- 230000008569 process Effects 0.000 description 7
- 230000001133 acceleration Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 238000009499 grossing Methods 0.000 description 3
- 239000013598 vector Substances 0.000 description 3
- 241000282414 Homo sapiens Species 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 2
- 238000012217 deletion Methods 0.000 description 2
- 230000037430 deletion Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000005484 gravity Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 101100259947 Homo sapiens TBATA gene Proteins 0.000 description 1
- 208000025174 PANDAS Diseases 0.000 description 1
- 208000021155 Paediatric autoimmune neuropsychiatric disorders associated with streptococcal infection Diseases 0.000 description 1
- 240000000220 Panda oleosa Species 0.000 description 1
- 235000016496 Panda oleosa Nutrition 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000009960 carding Methods 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/10—Information and communication technologies [ICT] supporting adaptation to climate change, e.g. for weather forecasting or climate simulation
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Operations Research (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to a space debris collision probability prediction method, a space debris collision probability prediction system, a space debris collision probability prediction terminal and a space debris collision probability prediction storage medium based on artificial intelligence, and belongs to the technical field of space safety monitoring; the space debris collision probability prediction method based on the artificial intelligence comprises the steps of taking a historical data set and a model to be trained, wherein the historical data set comprises satellite orbit historical data and space debris motion trail historical data; constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained; training the model to be trained according to the model adaptation data to obtain a collision probability prediction model; and carrying out collision prediction on the space debris by using the collision probability prediction model to obtain a prediction result. The application realizes the detection of the future space collision event and has important significance for improving the detection efficiency.
Description
Technical Field
The application relates to the technical field of space safety monitoring, in particular to a space debris collision probability prediction method, a space debris collision probability prediction system, a space debris collision probability prediction terminal and a space debris collision probability prediction storage medium based on artificial intelligence.
Background
With the continuous exploration and development of human beings in the space field, the number of space debris in space is also increasing; these space debris pose a hazard to the safety of space craft and satellites; therefore, prediction of spatial debris is important; the traditional orbit dynamics cannot meet the prediction of the increasingly large number of space fragments, the time consumption for predicting the space fragments by using the orbit dynamics is long, and the calculated amount is large; there is a need for a new way to achieve predictions of large and fast amounts of spatial debris.
Disclosure of Invention
The application provides a space debris collision probability prediction method, a system, a terminal and a storage medium based on artificial intelligence, which have the characteristic of improving the accuracy of space debris collision probability prediction.
The application aims to provide a space debris collision probability prediction method based on artificial intelligence.
The first object of the present application is achieved by the following technical solutions:
an artificial intelligence based space debris collision probability prediction method comprises the following steps:
the method comprises the steps of calling a historical data set and a model to be trained, wherein the historical data set comprises satellite orbit historical data and space debris motion trail historical data;
constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained;
training the model to be trained according to the model adaptation data to obtain a collision probability prediction model;
and carrying out collision prediction on the space debris by using the collision probability prediction model to obtain a prediction result.
By adopting the technical scheme, the data in the historical data set are quantized, the satellite orbit historical data and the space debris motion trail historical data are quantized into data which can be used and trained by a model, then the data are used for training the model to be trained, and finally the new space debris is subjected to collision prediction by using a collision probability prediction model obtained by training; by the method, the probability prediction can be carried out on the space debris by using the model, the collision probability prediction result of the space debris can be obtained only by inputting new space debris related data into the model, the consumed time is short, the efficiency is high, and the accuracy of the prediction result is improved.
The present application may be further configured in a preferred example to: before retrieving the model to be trained, the model to be trained is constructed using the ID3 algorithm.
The present application may be further configured in a preferred example to: the space debris movement track historical data comprises a plurality of component information, specifically space debris size and shape information, space debris track parameter information, space debris quantity and density information and space debris collision probability historical record information; the satellite orbit historical data comprises a plurality of component information, specifically satellite orbit parameter information and natural factor information; the step of constructing a model to be trained based on the ID3 algorithm and quantifying the historical data set to obtain model adaptation data comprises the following steps:
quantifying the space debris size and shape information into space debris length information, space debris width information and space debris height information;
quantizing the space debris orbit parameter information into space debris orbit height information and space debris orbit inclination angle information;
quantizing the space debris quantity and density information into space debris quantity information and space debris density information;
quantifying the spatial debris collision probability history information into collision probability level information, the collision probability level information comprising low, medium and high levels;
quantizing the satellite orbit parameter information into satellite orbit height information and satellite orbit parameter information;
quantizing the natural factor information into atmospheric density information and earth gravity information;
and integrating the quantized information to obtain the model adaptation data.
The present application may be further configured in a preferred example to: the training the model to be trained according to the model adaptation data to obtain a collision probability prediction model comprises the following steps:
calculating according to a preset information gain calculation formula and model adaptation data to obtain information gain data of each component information;
taking component information represented by the largest data in the information gain data as a division basis;
determining category information of the data set according to the division basis;
the collision probability historical record information in the data set of each category is called;
and obtaining a collision probability prediction result according to the collision probability historical record information and a preset collision probability calculation formula.
The present application may be further configured in a preferred example to: the preset information gain calculation formula comprises an overall entropy calculation formula, a component entropy calculation formula and a component information gain calculation formula.
The present application may be further configured in a preferred example to: the calculating the information gain data of each component information according to the preset information gain calculation formula and the model adaptation data comprises the following steps:
obtaining overall entropy information according to the collision probability historical record information and the overall entropy calculation formula;
obtaining component entropy information according to the quantized information of each component information and the component entropy calculation formula;
and obtaining information gain data of each component information according to the overall entropy information, the component entropy information and the component information gain calculation formula.
The present application may be further configured in a preferred example to: before the historical data set is called, data cleaning, data denoising and data interpolation processing are carried out on the historical data in the historical data set.
The application aims at providing a space debris collision probability prediction system based on artificial intelligence.
The second object of the present application is achieved by the following technical solutions:
an artificial intelligence based spatial debris collision probability prediction system comprising:
the system comprises a calling module, a training module and a training module, wherein the calling module is used for calling a historical data set and a model to be trained, and the historical data set comprises satellite orbit historical data and space debris motion trail historical data;
the quantization module is used for constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained;
the training module is used for training the model to be trained according to the model adaptation data to obtain a collision probability prediction model;
and the prediction module is used for carrying out collision prediction on the space fragments by using the collision probability prediction model to obtain a prediction result.
The application aims at providing a terminal.
The third object of the present application is achieved by the following technical solutions:
a terminal comprising a memory and a processor, the memory having stored thereon computer program instructions capable of being loaded and executed by the processor for the artificial intelligence based method for predicting probability of collision of spatial fragments.
A fourth object of the present application is to provide a computer medium capable of storing a corresponding program.
The fourth object of the present application is achieved by the following technical solutions:
a computer readable storage medium storing a computer program capable of being loaded by a processor and executing any one of the artificial intelligence based spatial debris collision probability prediction methods described above.
Drawings
Fig. 1 is a schematic flow chart of a space debris collision probability prediction method based on artificial intelligence in an embodiment of the application.
FIG. 2 is a schematic structural diagram of an artificial intelligence based spatial debris collision probability prediction system in an embodiment of the application.
Reference numerals illustrate: 1. a calling module; 2. a quantization module; 3. a training module; 4. and a prediction module.
Detailed Description
The present embodiment is only for explanation of the present application and is not to be construed as limiting the present application, and modifications to the present embodiment, which may not creatively contribute to the present application as required, are within the scope of the claims of the present application as far as they are protected by patent law.
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present application more apparent, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments of the present application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Embodiments of the application are described in further detail below with reference to the drawings.
With the development of space science and technology, the track of human beings in space is increased; the consequent impact is that the number of space debris in space is also increasing; spatial debris refers to the product of human spatial activity. The space environment pollution-free rocket comprises a rocket body and a satellite body for completing a task, a jet of the rocket, a throwing object in the process of executing a space task, fragments generated by collision between space objects and the like, and is a main pollution source of the space environment; the space debris can threaten the safety of space vehicles and satellites, and the threat brought by the increasingly more space debris is also increased, so that the collision prediction of the space debris is very important, and corresponding measures can be taken through the collision prediction result of the space debris so as to realize the avoidance of the space debris.
In the traditional space debris collision prediction research, an orbit dynamics method is generally adopted; this approach, while more reliable, is time consuming and requires significant computing resources; as the number of space debris increases, the speed of the space debris is also fast, and for these reasons, conventional orbit dynamics has failed to meet the collision prediction research requirements of the space debris.
At present, the artificial intelligence technology can improve the quality and speed of information processing, and can provide important support for a space situation awareness judging system to meet the requirements; therefore, artificial intelligence is adopted to predict the collision probability of the space debris; the ID3 decision tree adopted by the application is a machine learning method, can classify and predict data, and has higher precision and interpretability.
The application provides an artificial intelligence-based space debris collision probability prediction method, and the main flow of the method is described as follows.
As shown in fig. 1:
step S101: the historical dataset and the model to be trained are invoked.
In the embodiment of the application, the historical data set comprises satellite orbit historical data and space debris motion trail historical data; specifically, the historical data can be obtained through satellites, ground telescopes or other equipment, and then the historical data is integrated to obtain a historical data set; before the historical data set is called, data cleaning, data denoising and data interpolation are required to be carried out on the historical data in the historical data set.
The specific format and content of the satellite orbit history data and the spatial debris movement trajectory history data varies from source to source and from scene of use.
The TLE two-line root number format is published by the north american aerospace national defense command department (North American aerospace defence command, NORAD). The satellite trajectory is described by two lines of TLE orbit parameter data, each line consisting of 69 characters including characters. For ease of distinction, the web site that provides the TLE data will typically add the name of the object to the two rows of parameters.
The 3 rows of structures of the satellite ephemeris are that the first row of data is the satellite name; the next two rows store satellite related data, 69 characters each, including 0-9, a-Z (uppercase), spaces, dots, and signs. The format description of rows 2 and 3 is shown in the following table.
Line 2 description of the formats
Line 3 description of the formats
For another example, JSON format is a lightweight data exchange format, commonly used to transfer and store structured data; the following is an example, JSON format containing fragment motion trajectory data: { "id": "123456", "type": "debris", "position": { "x":1000, "y":2000, "z":3000}, "velocity": { x ":10," y ":20," z ":30}," aceration "{" x ":1," y ":2," z ":3}," time stamp ": 2022-01T 12:00:000.000Z" }; this JSON object contains the ID, type, location, velocity, acceleration and timestamp information of the fragment; the position, the speed and the acceleration are three-dimensional vectors, and respectively represent the motion states of the fragments on three axes of x, y and z; the time stamp is in an ISO 8601 format and represents time information of the fragment motion trail data.
For the historical data, the acquisition paths are different and the formats are different, so that the historical data is required to be subjected to data processing to obtain a historical data set, and then the historical data set is used for calculation; the data processing modes in the embodiment of the application comprise data cleaning, denoising, interpolation and the like; by means of the data carding, more accurate satellite orbit data and space debris motion trail data can be obtained.
Satellite orbit and motion information typically contains orbit parameters, orbit states, and time stamps; the orbit parameters comprise parameters such as orbit inclination angle, ascent and intersection point right ascent, eccentricity, near-place amplitude angle and the like; the orbit state comprises information such as the position, the speed, the acceleration and the like of the satellite on the orbit; the time stamp is the time of recording data acquisition; the spatial debris motion trajectory typically contains position information, velocity information, acceleration information, and a time stamp; wherein the position information comprises position coordinates of the space debris in the three-dimensional space; the velocity information comprises velocity vectors of the spatial debris in three-dimensional space; the acceleration information comprises acceleration vectors of the space debris in the three-dimensional space; the time stamp is the time at which the data acquisition was recorded.
After the satellite orbit data and the space debris motion trail data are determined, cleaning is firstly carried out, wherein the data cleaning refers to processing invalid information such as errors, deletions, repetition, independence and the like in the data so as to improve the data quality; the data cleaning method comprises the steps of removing repeated data, and sequencing and de-duplicating the data to avoid the influence of the repeated data on analysis results; processing the missing value, wherein the missing value can be processed by adopting methods such as interpolation, deletion or replacement; processing the abnormal value, and smoothing, deleting or replacing the abnormal value to process; for example, for a set of track data, if duplicate data or missing values are present, the data cleansing can be performed using the pandas library in Python; duplicate data is removed using the drop_duplicate () method, and missing values are deleted using the drop () method.
After the data cleaning process, performing data denoising, namely processing noise in the data to improve the data quality; the data denoising method comprises the steps of smoothing and filtering, and smoothing the data by adopting methods such as moving average, median filtering and the like; wavelet transformation, denoising data through wavelet transformation, and reserving main characteristics of the data; denoising the data by a statistical method based on denoising of the statistical method, for example, processing by using statistics such as mean value, standard deviation and the like; for example, for a set of motion data, if noise is present, denoising may be performed using a numpy library in Python; the data is smoothed using a median filtering method.
Finally, data interpolation is needed to be carried out on the data, wherein the data interpolation refers to filling of missing data so as to improve the data quality; the data interpolation method comprises the steps of linearly interpolating, and interpolating missing data through a linear function; interpolation of Lagrange, namely interpolating the missing data through a Lagrange polynomial; spline interpolation, namely interpolating missing data through a spline function; for example, for a set of track data, if there are missing values, interpolation can be performed using the scipy library in Python; linear interpolation is performed using the interpolation 1d () method.
The processing of the historical data is realized by the mode, and a historical data set is obtained; the historical data set is stored in the server, and the historical data set is convenient to use only by directly calling the historical data set when the historical data set is needed to be used; and the historical data in the historical data set are subjected to data processing, so that the data quality in the set can be ensured.
It can be understood that in the embodiment of the application, the acquired satellite orbit data and space debris motion trail data are subjected to data processing by adopting a data processing method, and then the processed data are integrated to obtain a historical data set; by the method, the historical data can be conveniently used later, and meanwhile, the data processing efficiency and the data processing quality are improved.
Step S102: and constructing a model to be trained based on an ID3 algorithm, and quantifying the historical data set to obtain model adaptation data.
It is understood that the model adaptation data herein refers to data that can be used and trained by the model to be trained; in the embodiment of the application, the model is constructed based on the ID3 algorithm, so that the data in the historical data set is quantized into the data which can be matched with the model for use, and the subsequent operation is convenient.
Specifically, the space debris movement track history data comprises a plurality of component information, specifically space debris size and shape information, space debris track parameter information, space debris quantity and density information and space debris collision probability history record information; the satellite orbit history data contains a plurality of component information, specifically satellite orbit parameter information and natural factor information.
Quantifying the space debris size and shape information into space debris length information, space debris width information and space debris height information; quantizing the space debris orbit parameter information into space debris orbit height information and space debris orbit inclination angle information; quantizing the space debris quantity and density information into space debris quantity information and space debris density information; quantifying the spatial debris collision probability history information into collision probability level information, the collision probability level information comprising low, medium and high levels; quantizing the satellite orbit parameter information into satellite orbit height information and satellite orbit parameter information; quantizing the natural factor information into atmospheric density information and earth gravity information; and integrating the quantized information to obtain the model adaptation data.
It can be understood that determining a plurality of component information contained in the space debris motion trail historical data and the satellite orbit historical data according to the space debris motion trail historical data, and then specifically quantifying the component information; the quantization process is to convert the component information into a numerical variable and a classification variable; for example, the size and shape of the space debris is quantized into numerical variables such as length, width, and height; and for the space debris collision probability history information, the space debris collision probability history information is quantized into classification variables, namely low, medium and high three grades.
It should be noted that, in the above examples, it is mentioned that both the satellite orbit data and the space debris data include the number information, that is, in the process of quantifying the history data, the operation is performed for the satellite or the space debris data corresponding to each number; the satellite orbit data or the space debris data is one-to-one corresponding to a specific number even after a series of processing.
Step S103: and training the model to be trained according to the model adaptation data to obtain a collision probability prediction model.
After the model adaptation data is obtained, the model adaptation data can be utilized to train the model to be trained to obtain a collision probability model; specifically, calculating information gain data of each component information according to a preset information gain calculation formula and model adaptation data; the component information represented by the largest data in the information gain data is regarded as a division basis; determining category information of the data set according to the division basis; the collision probability historical record information in the data set of each category is called; and obtaining a collision probability prediction model according to the collision probability historical record information and a preset collision probability calculation formula.
The preset information gain calculation formula comprises a total entropy calculation formula, a component entropy calculation formula and a component information gain calculation formula; the overall entropy calculation formula is H (S) = - Σ (p_i) log2 (p_i), where p_i is the probability of each class; the component entropy calculation formula is H (S, a) =Σ (p_i) H (s_i), where s_i is a subset under feature a; the component information gain calculation formula is IG (S, a) =h (S) -H (S, a).
Calculating according to three formulas included in a preset information gain calculation formula and model adaptation data; obtaining overall entropy information according to the collision probability historical record information and an overall entropy calculation formula; obtaining component entropy information according to the quantized information of each component information and a component entropy calculation formula; and obtaining information gain data of each component information according to the overall entropy information, the component entropy information and the component information gain calculation formula.
In order to explain the above calculation process in detail, an example will be described below.
| Numbering device | Space debris length (m) | Space debris width (m) | Space debris height (m) |
| 1 | 1 | 0.5 | 0.5 |
| 2 | 1.5 | 1 | 1 |
| 3 | 0.5 | 0.5 | 0.5 |
| 4 | 2 | 1 | 1 |
| 5 | 1 | 1 | 0.5 |
| Numbering device | Chip track height (km) | Inclination angle (°) of chip track | Satellite orbit altitude (km) | Satellite orbit inclination angle (°) |
| 1 | 800 | 10 | 800 | 5 |
| 2 | 1000 | 20 | 1000 | 10 |
| 3 | 600 | 5 | 600 | 2 |
| 4 | 1200 | 30 | 1200 | 15 |
| 5 | 900 | 15 | 900 | 8 |
Here a set of data sets, containing 5 samples.
For the samples, firstly, calculating the overall entropy of the data set by using an overall entropy formula; wherein, the sample number is 5, and comprises 2 high collision probabilities, 1 medium collision probability and 2 low collision probabilities; it is thereby possible to obtain a ratio of high collision probability of 2/5, a ratio of medium collision probability of 1/5, and a ratio of low collision probability of 2/5 for the data set; calculated according to the general entropy formula, - (2/5) log2 (2/5) - (1/5) log2 (1/5) - (2/5) log2 (2/5) about 1.522; the overall entropy of the dataset is 1.522.
After the overall entropy of the data set is obtained, the component entropy and the information gain of each component information need to be calculated.
For the size and shape information of the space debris, when calculating the feature, calculating according to the quantized space debris length information, space debris width information and space debris height information; first, the conditional entropy of the length is calculated: for space debris with a length of less than 2m, there are 2 high collision probabilities, 1 low collision probability, so its entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for space debris with a length of 2m or more, there are 0 high collision probabilities, 1 medium collision probability and 1 low collision probability, so the entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the length is: (3/5) 0.918+ (2/5) 1 ≡0.951.
Then, the conditional entropy of the width is calculated: for space debris with a width of less than 1m, there are 2 high collision probabilities, 1 low collision probability, so its entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for space debris with a width of 1m or more, there are 0 high collision probability, 1 medium collision probability and 1 low collision probability, so the entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the width is: (3/5) 0.918+ (2/5) 1 ≡0.951.
Finally, the conditional entropy of the height is calculated: for space debris with a height of less than 0.5m, there are 2 high collision probabilities, 1 low collision probability, so its entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for space debris with a height of 0.5m or more, there are 0 high collision probability, 1 medium collision probability, 1 low collision probability, so its entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the height is: (3/5) 0.918+ (2/5) 1 ≡0.951.
The conditional entropy of the three components of length, width and height is synthesized, and the conditional entropy of the size and shape of the space debris is obtained as follows: 0.951+0.951+0.951= 2.853; thus, the information gain for space debris size and shape is: 1.522-2.853 = -1.331.
Aiming at the orbit parameters of the space debris, when calculating the characteristics, respectively calculating the orbit height and the orbit inclination angle of the quantized space debris; first, the conditional entropy of the track height is calculated: for space debris with a track height of 900km or more, there are 2 medium collision probabilities and 1 high collision probability, so the entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for space debris with track height less than 900km, there are 1 high collision probability, 0 medium collision probability and 1 low collision probability, so the entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the track height is: (3/5) 0.918+ (2/5) 1 ≡0.951.
Then, the conditional entropy of the track tilt angle is calculated: for space debris with the track inclination angle larger than 10 degrees, there are 2 medium collision probabilities and 1 high collision probability, so the entropy is as follows: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for space debris with the track inclination angle less than or equal to 10 degrees, the high collision probability of 1, the medium collision probability of 0 and the low collision probability of 1 exist, so that the entropy is as follows: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the track tilt angle is: (3/5) 0.918+ (2/5) 1 ≡0.951.
The conditional entropy of the two components is synthesized, and the conditional entropy of the orbit parameters of the space debris is obtained as follows: 0.951+0.951= 1.902; therefore, the information gain of the orbit parameters of the space debris is: 1.522-1.902-0.38.
For the orbit parameters of the satellite, when calculating the characteristic, respectively calculating the orbit height and the orbit inclination angle of the satellite according to the quantized characteristic; first, the conditional entropy of the track height is calculated: for satellites with orbit heights greater than 800km, there are 2 medium collision probabilities, 1 high collision probability, so its entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for satellites with orbital heights of 800km or less, there are 1 high collision probability, 0 medium collision probability, 1 low collision probability, so its entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the track height is: (3/5) 0.918+ (2/5) 1 ≡0.951.
Then, the conditional entropy of the track tilt angle is calculated: for satellites with orbital inclinations greater than 5 °, there are 1 high probability of collision, 2 medium probability of collision, so its entropy is: - (2/3) log2 (2/3) - (1/3) log2 (1/3) ≡0.918; for satellites with orbital inclinations less than or equal to 5 °, there are 1 high collision probability, 0 medium collision probability, 1 low collision probability, so its entropy is: - (1/2) log2 (1/2) ≡1; therefore, the conditional entropy of the track tilt angle is: (3/5) 0.918+ (2/5) 1 ≡0.951.
The conditional entropy of the two components is synthesized, and the conditional entropy of the orbit parameters of the satellite is obtained as follows: 0.951+0.951= 1.902; thus, the information gain of the orbit parameters of the satellite is: 1.522-1.902-0.38.
Aiming at the characteristic of the atmospheric density, only one numerical variable exists, and the atmospheric density is divided into high density and low density directly according to the numerical variable; specifically, the density of the air is 1.5e-12kg/m or less 3 Has 2 high collision probabilities and 2 medium collision probabilities, so that the entropy is as follows: - (2/4) log2 (2/4) =1.000; for atmospheric densities greater than 1.5e-12kg/m 3 The space debris of (2) has 0 high collision probability, 0 medium collision probability and 1 low collision probability, so the entropy is as follows: - (1/1) log2 (1/1) ≡0; thus, the atmospheric density conditional entropy is: (4/5) 1.000+ (1/5) 0-0.951; the information gain of the atmospheric density is: 1.522-0.951. Apprxeq.0.571.
Aiming at the characteristic of the density of the fragments, two numerical variables exist, and the fragments are divided into high density and low density directly according to the values; specifically, for space debris with a debris density of 120 or less, there are 1 high collision probability, 1 medium collision probability, 1 low collision probability, so its entropy is: - (1/3) log2 (1/3) ≡1.585; for space debris with the number of fragments greater than 120, there are 1 high collision probability, 1 medium collision probability and 0 low collision probability, so the entropy is: - (1/2) log2 (1/2) ≡1.000; thus, the conditional entropy of the patch density is: - (3/5) log2 (3/5) - (2/5) log2 (2/5) ≡0.971; information gain for calculating patch density: 1.522-0.971=0.551.
From the above calculation results, we can obtain the information gain for each feature as follows:
1) Size and shape of space debris: -1.331;
2) Orbit parameters of space debris: -0.38;
3) Orbit parameters of satellites: -0.38;
4) Atmospheric density: 0.571;
5) Density of chips: 0.551;
therefore, the air density can be selected as the division basis, and the data set can be better divided because the information gain is the largest.
Step S104: and carrying out collision prediction on the space debris by using the collision probability prediction model to obtain a prediction result.
After the collision probability prediction model is obtained, the model can be used to predict the collision of the space debris.
The following is illustrative:
according to the data, carrying out collision probability prediction through a collision probability prediction model; dividing the data set into the data set with the atmospheric density being less than or equal to 1.5e -12 kg/m 3 And greater than 1.5e -12 kg/m 3 Two classes; since the atmospheric density of the dataset is 1.6e -12 kg/m 3 The data set is thus divided into more than 1.6e -12 kg/m 3 Is a category of (2); in this category, there are 0 high collision probabilities, 0 medium collision probabilities, 1 low collision probability, so the prediction result is a low collision probability.
The application also provides a space debris collision probability prediction system based on artificial intelligence, as shown in fig. 2, which comprises a calling module 1, a training module and a prediction module, wherein the calling module is used for calling a historical data set and a model to be trained, and the historical data set comprises satellite orbit historical data and space debris motion trail historical data; the quantization module 2 is used for constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained; the training module 3 is used for training the model to be trained according to the model adaptation data to obtain a collision probability prediction model; and the prediction module 4 is used for carrying out collision prediction on the space debris by using the collision probability prediction model to obtain a prediction result.
In order to better execute the program of the method, the application also provides a terminal, which comprises a memory and a processor.
Wherein the memory may be used to store instructions, programs, code, sets of codes, or sets of instructions. The memory may include a stored program area and a stored data area, wherein the stored program area may store instructions for implementing an operating system, instructions for at least one function, instructions for implementing the above-described artificial intelligence-based space debris collision probability prediction method, and the like; the storage data area may store data and the like involved in the above-described artificial intelligence-based spatial debris collision probability prediction method.
The processor may include one or more processing cores. The processor performs the various functions of the application and processes the data by executing or executing instructions, programs, code sets, or instruction sets stored in memory, calling data stored in memory. The processor may be at least one of an application specific integrated circuit, a digital signal processor, a digital signal processing device, a programmable logic device, a field programmable gate array, a central processing unit, a controller, a microcontroller, and a microprocessor. It will be appreciated that the electronics for implementing the above-described processor functions may be other for different devices, and embodiments of the present application are not particularly limited.
The present application also provides a computer-readable storage medium, for example, comprising: a U-disk, a removable hard disk, a Read Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes. The computer readable storage medium stores a computer program that can be loaded by a processor and that performs the artificial intelligence based method of predicting probability of collision of a spatial debris.
The above description is only illustrative of the preferred embodiments of the present application and of the principles of the technology employed. It will be appreciated by persons skilled in the art that the scope of the disclosure referred to in the present application is not limited to the specific combinations of technical features described above, but also covers other technical features which may be formed by any combination of the technical features described above or their equivalents without departing from the spirit of the disclosure. Such as the above-mentioned features and the technical features disclosed in the present application (but not limited to) having similar functions are replaced with each other.
Claims (10)
1. The space debris collision probability prediction method based on artificial intelligence is characterized by comprising the following steps of:
retrieving a historical data set, wherein the historical data set comprises satellite orbit historical data and space debris motion trail historical data;
constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained;
training the model to be trained according to the model adaptation data to obtain a collision probability prediction model;
and carrying out collision prediction on the space debris by using the collision probability prediction model to obtain a prediction result.
2. The artificial intelligence based space debris collision probability prediction method according to claim 1, wherein the model to be trained is constructed using an ID3 algorithm before the model to be trained is called.
3. The method for predicting the probability of a space debris collision based on artificial intelligence according to claim 1, wherein the space debris motion trajectory history data comprises a plurality of component information, in particular space debris size and shape information, space debris trajectory parameter information, space debris number and density information and space debris collision probability history information; the satellite orbit historical data comprises a plurality of component information, specifically satellite orbit parameter information and natural factor information; the step of constructing a model to be trained based on the ID3 algorithm and quantifying the historical data set to obtain model adaptation data comprises the following steps:
quantifying the space debris size and shape information into space debris length information, space debris width information and space debris height information;
quantizing the space debris orbit parameter information into space debris orbit height information and space debris orbit inclination angle information;
quantizing the space debris quantity and density information into space debris quantity information and space debris density information;
quantifying the spatial debris collision probability history information into collision probability level information, wherein the collision probability level information comprises high, medium and low;
quantizing the satellite orbit parameter information into satellite orbit height information and satellite orbit parameter information;
quantifying the natural factor information into atmospheric density information;
and integrating the quantized information to obtain the model adaptation data.
4. The method for predicting the collision probability of space debris based on artificial intelligence according to claim 3, wherein the training the model to be trained according to the model adaptation data to obtain the collision probability prediction model comprises:
calculating according to a preset information gain calculation formula and model adaptation data to obtain information gain data of each component information;
taking component information represented by the largest data in the information gain data as a division basis;
determining category information of the data set according to the division basis;
the collision probability historical record information in the data set of each category is called;
and obtaining a collision probability prediction model according to the collision probability historical record information and a preset collision probability calculation formula.
5. The artificial intelligence based space debris collision probability prediction method according to claim 4, wherein the preset information gain calculation formula includes an overall entropy calculation formula, a component entropy calculation formula, and a component information gain calculation formula.
6. The method for predicting the probability of collision of a space debris based on artificial intelligence according to claim 5, wherein the calculating the information gain data of each component information according to the preset information gain calculation formula and the model adaptation data comprises:
obtaining overall entropy information according to the collision probability historical record information and the overall entropy calculation formula;
obtaining component entropy information according to the quantized information of each component information and the component entropy calculation formula;
and obtaining information gain data of each component information according to the overall entropy information, the component entropy information and the component information gain calculation formula.
7. The artificial intelligence based space debris collision probability prediction method according to claim 1, wherein before the historical data set is called, data cleaning, data denoising and data interpolation processing are performed on the historical data in the historical data set.
8. An artificial intelligence based space debris collision probability prediction system, comprising:
the invoking module (1) is used for invoking a historical data set and a model to be trained, wherein the historical data set comprises satellite orbit historical data and space debris motion trail historical data;
the quantization module (2) is used for constructing a model to be trained based on an ID3 algorithm, and quantizing the historical data set to obtain model adaptation data, wherein the model adaptation data refer to data which can be used and trained by the model to be trained;
the training module (3) is used for training the model to be trained according to the model adaptation data to obtain a collision probability prediction model;
and the prediction module (4) is used for carrying out collision prediction on the space debris by utilizing the collision probability prediction model to obtain a prediction result.
9. A terminal comprising a memory and a processor, the memory having stored thereon computer program instructions capable of being loaded by the processor and performing the method according to any of claims 1-7.
10. A computer readable storage medium, characterized in that a computer program is stored which can be loaded by a processor and which performs the method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310848212.4A CN116861241B (en) | 2023-07-11 | 2023-07-11 | Space debris collision probability prediction method and system based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310848212.4A CN116861241B (en) | 2023-07-11 | 2023-07-11 | Space debris collision probability prediction method and system based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116861241A true CN116861241A (en) | 2023-10-10 |
| CN116861241B CN116861241B (en) | 2024-02-20 |
Family
ID=88226453
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310848212.4A Active CN116861241B (en) | 2023-07-11 | 2023-07-11 | Space debris collision probability prediction method and system based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116861241B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120089517A (en) * | 2010-12-15 | 2012-08-13 | 한국항공우주연구원 | System and Method for Collision Risk Management between Satellite and Space Debris |
| US20150134295A1 (en) * | 2013-10-28 | 2015-05-14 | Korea Aerospace Research Institute | Csm-based analysis system for collision risk |
| CN112214927A (en) * | 2020-09-21 | 2021-01-12 | 中国工程物理研究院计算机应用研究所 | Machine learning-based ultrahigh-speed collision fragment cloud rapid simulation method |
| CN112623283A (en) * | 2020-12-30 | 2021-04-09 | 苏州三六零智能安全科技有限公司 | Space object abnormity detection method, device, equipment and storage medium |
| US20210261276A1 (en) * | 2020-02-20 | 2021-08-26 | Raytheon Company | Space object maneuver detection |
| US20220058922A1 (en) * | 2020-08-20 | 2022-02-24 | Rutgers, The State University Of New Jersey | Computer-based systems configured for space object orbital trajectory predictions and methods thereof |
| EP4016504A1 (en) * | 2020-12-15 | 2022-06-22 | OKAPI:Orbits GmbH | Method, computer program and hardware for determining a collision probability |
| CN114780915A (en) * | 2022-04-12 | 2022-07-22 | 北京航天驭星科技有限公司 | Method, device and equipment for determining data correctness of collision early warning service platform |
| CN114936478A (en) * | 2022-07-26 | 2022-08-23 | 数字太空(北京)智能技术研究院有限公司 | Space debris collision early warning simulation platform and simulation method |
| CN115285381A (en) * | 2022-10-09 | 2022-11-04 | 北京开运联合信息技术集团股份有限公司 | Collision early warning method and device for space debris |
| US20220371755A1 (en) * | 2019-09-26 | 2022-11-24 | Mitsubishi Electric Corporation | Collision avoidance assistance device, satellite constellation forming system, collision avoidance assistance method, computer readable medium, collision avoidance assistance system, and satellite constellation business device |
| FR3127616A1 (en) * | 2021-09-27 | 2023-03-31 | Thales | Aerial avoidance management process |
| CN116304924A (en) * | 2023-03-06 | 2023-06-23 | 北京航空航天大学 | Decision tree forest construction system for target satellite intention recognition and application method thereof |
-
2023
- 2023-07-11 CN CN202310848212.4A patent/CN116861241B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120089517A (en) * | 2010-12-15 | 2012-08-13 | 한국항공우주연구원 | System and Method for Collision Risk Management between Satellite and Space Debris |
| US20150134295A1 (en) * | 2013-10-28 | 2015-05-14 | Korea Aerospace Research Institute | Csm-based analysis system for collision risk |
| US20220371755A1 (en) * | 2019-09-26 | 2022-11-24 | Mitsubishi Electric Corporation | Collision avoidance assistance device, satellite constellation forming system, collision avoidance assistance method, computer readable medium, collision avoidance assistance system, and satellite constellation business device |
| US20210261276A1 (en) * | 2020-02-20 | 2021-08-26 | Raytheon Company | Space object maneuver detection |
| US20220058922A1 (en) * | 2020-08-20 | 2022-02-24 | Rutgers, The State University Of New Jersey | Computer-based systems configured for space object orbital trajectory predictions and methods thereof |
| CN112214927A (en) * | 2020-09-21 | 2021-01-12 | 中国工程物理研究院计算机应用研究所 | Machine learning-based ultrahigh-speed collision fragment cloud rapid simulation method |
| EP4016504A1 (en) * | 2020-12-15 | 2022-06-22 | OKAPI:Orbits GmbH | Method, computer program and hardware for determining a collision probability |
| CN112623283A (en) * | 2020-12-30 | 2021-04-09 | 苏州三六零智能安全科技有限公司 | Space object abnormity detection method, device, equipment and storage medium |
| FR3127616A1 (en) * | 2021-09-27 | 2023-03-31 | Thales | Aerial avoidance management process |
| CN114780915A (en) * | 2022-04-12 | 2022-07-22 | 北京航天驭星科技有限公司 | Method, device and equipment for determining data correctness of collision early warning service platform |
| CN114936478A (en) * | 2022-07-26 | 2022-08-23 | 数字太空(北京)智能技术研究院有限公司 | Space debris collision early warning simulation platform and simulation method |
| CN115285381A (en) * | 2022-10-09 | 2022-11-04 | 北京开运联合信息技术集团股份有限公司 | Collision early warning method and device for space debris |
| CN116304924A (en) * | 2023-03-06 | 2023-06-23 | 北京航空航天大学 | Decision tree forest construction system for target satellite intention recognition and application method thereof |
Non-Patent Citations (5)
| Title |
|---|
| MD. RIFTABIN KABIR ET AL.: "A satellite collision avoidance system based on General Regression Neural Network", 《2020 IEEE/ACM INTERNATIONAL CONFERENCE ON BIG DATA COMPUTING, APPLICATIONS AND TECHNOLOGIES (BDCAT)》, pages 154 - 160 * |
| ZICAI SHEN ETAL.: "Effects of space environments on space laser system for spacecraft", 《PROCEEDINGS OF SPIE》, pages 1 - 5 * |
| 冯博鑫等: "在轨拦截器空间碎片碰撞概率算法", 《四川兵工学报》, vol. 36, no. 4, pages 46 - 49 * |
| 曾安里等: "空间碎片碰撞预警分析与显示技术研究", 《空间碎片研究》, vol. 19, no. 4, pages 21 - 27 * |
| 杨大陶: "空间碎片预警并行计算系统的设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》, no. 3, pages 15 - 26 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116861241B (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111062282B (en) | Substation pointer instrument identification method based on improved YOLOV3 model | |
| CN108629231B (en) | Obstacle detection method, apparatus, device and storage medium | |
| CN109191922B (en) | Large-scale four-dimensional track dynamic prediction method and device | |
| CN108519094B (en) | Local path planning method and cloud processing terminal | |
| CN112526513B (en) | Millimeter wave radar environment map construction method and device based on clustering algorithm | |
| US11295517B2 (en) | Generating realistic point clouds | |
| CN110609881A (en) | Vehicle trajectory deviation detection method, system and storage medium | |
| CN111540193A (en) | Traffic data restoration method for generating countermeasure network based on graph convolution time sequence | |
| CN115145906B (en) | Preprocessing and completion method for structured data | |
| EP3657402A1 (en) | Pruning neural networks that include element-wise operations | |
| WO2023247827A1 (en) | Method and system for processing point-cloud data | |
| CN115759033A (en) | Method, device and equipment for processing track data | |
| CN116861241B (en) | Space debris collision probability prediction method and system based on artificial intelligence | |
| CN116495204A (en) | Rail transaction detection method and system based on transaction probability | |
| Chen et al. | An application of convolutional neural network to derive vessel movement patterns | |
| CN111832343A (en) | Eye tracking method and device and storage medium | |
| CN114137525A (en) | Multi-target detection method and system based on vehicle-mounted millimeter wave radar | |
| CN114792397A (en) | SAR image urban road extraction method, system and storage medium | |
| CN109612467B (en) | Track simplifying method and system and track visualization system | |
| CN111832548A (en) | Train positioning method | |
| CN116049253B (en) | Deep learning-based Internet of vehicles data modeling method and system | |
| CN117036966B (en) | Learning method, device, equipment and storage medium for point feature in map | |
| CN113554118B (en) | Quantum financial bill classification method, system, storage medium and terminal | |
| CN111338336B (en) | Automatic driving method and device | |
| CN115270920A (en) | Ship target classical trajectory generation method based on density space clustering |
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 |