Flight path meeting early warning method based on fuzzy comprehensive evaluation

Abstract

The invention discloses a track encounter early warning method based on fuzzy comprehensive evaluation. The method comprises the following steps: acquiring track information detected by a radar to a ship; selecting two to-be-judged tracks, and predicting the two to-be-judged tracks by adopting an LSTM network; according to the extrapolated flight path, determining the latest meeting probability and the latest meeting time; determining a membership function and an evaluation level of the recent meeting probability; determining a membership function and an evaluation level of the latest meeting time; determining the weight of the latest meeting probability and the weight of the latest meeting time by using an analytic hierarchy process; and constructing a risk assessment model by using a fuzzy set theory. According to the method, the calculation complexity is reduced, the error is reduced, the practicability and the accuracy are improved, and the problem of mixed batch and wrong batch of tracks is effectively warned.

Description

Technical Field

[0001] The invention belongs to the technical field of radar data processing, and in particular relates to a track encounter warning method based on fuzzy comprehensive evaluation. Background Art

[0002] The complexity of radar detection environment and the diversification of detection tasks have led to the development of radar detection technology towards digitalization, software, and intelligence, with an increasingly high degree of automation, and various technical indicators and radar systems are becoming more and more advanced, but they still need to be operated by specific personnel. Radars encounter various dense and busy air situations, complex electromagnetic environments, clutter environments, and various active and passive interference phenomena, which put forward higher and higher requirements for radar operators.

[0003] In recent years, the application of automation equipment and information technology in the field of ships has become more and more extensive. The installed capacity of electrical automation equipment for ship navigation and communication technology has continued to increase. Emerging navigation aids such as the Automatic Identification System (AIS system) and navigation radar have become the most widely used safety navigation aids. Traditional warning algorithms are mainly aimed at ship collision avoidance problems, but not at the problem of mixed batches and wrong batches of radar tracks. The AIS system, composed of shore-based (base station) facilities and ship-borne equipment, is a new type of digital navigation aid system and equipment that integrates network technology, modern communication technology, computer technology, and electronic information display technology. Most of the existing algorithms are based on AIS data, and do not consider the quantitative impact of radar measurement errors on the risk of track encounter warnings. Summary of the invention

[0004] The present invention aims to provide a track encounter warning method based on fuzzy comprehensive evaluation, which can reduce the computational complexity, reduce the error, improve the practicality and accuracy, and realize the effective warning of track mixed batch and wrong batch problem.

[0005] In order to achieve the purpose of the present invention, the present invention provides a track encounter warning method based on fuzzy comprehensive evaluation, comprising the following steps:

[0006] Step 1: Obtain the track information of the ship detected by the radar, wherein the track information includes the track position, speed, direction, and timestamp information;

[0007] Step 2: Select the first track and the second track To be judged, the track information is predicted using the LSTM network;

[0008] Step 3: Interpolate the predicted track information to determine the nearest encounter probability and the nearest encounter time TCPA;

[0009] Step 4: The nearest encounter probability and the most recent encounter time TCPA are selected as a factor set;

[0010] Step 5: Determine the nearest encounter probability Membership function and evaluation level ;

[0011] Step 6: Determine the membership function and evaluation level of the TCPA of the nearest encounter time ;

[0012] Step 7: Use the analytic hierarchy process to determine the probability weight of the nearest encounter and the weight of the time of the most recent encounter ;

[0013] Step 8: Based on the membership function and the weight, a hazard assessment model is constructed using fuzzy set theory to obtain the hazard degree of the track encounter. .

[0014] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned track encounter warning method based on fuzzy comprehensive evaluation when executing the program.

[0015] A non-transitory computer-readable storage medium, characterized in that the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the above-mentioned track encounter warning method based on fuzzy comprehensive evaluation.

[0016] A computer program product includes computer program instructions. When the computer program instructions are executed on a computer, the computer is enabled to execute the above-mentioned track encounter warning method based on fuzzy comprehensive evaluation.

[0017] Compared with the prior art, the significant progress of the present invention is that: the present invention aims at the target encounter alarm problem in the radar track, selects the two factors of the nearest encounter probability and the nearest encounter time to form a set of track encounter risk influencing factors, constructs a membership function, uses the hierarchical analysis method to determine the weights of the influencing factors, and finally constructs a collision risk assessment model based on the fuzzy set theory, thereby reducing the calculation complexity, reducing the error, and improving the practicality and accuracy, and can effectively warn of the problem of mixed and wrong batches of tracks.

[0018] In order to more clearly illustrate the functional characteristics and structural parameters of the present invention, further description is given below in conjunction with the accompanying drawings and specific implementation methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0020] Figure 1 It is a track encounter warning flow chart of the present invention;

[0021] Figure 2 It is a graph of the closest encounter probability membership function curve of the present invention;

[0022] Figure 3 It is a closest encounter time membership function curve diagram of the present invention. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments; based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] The present invention is a track encounter warning method based on fuzzy comprehensive evaluation, combined with Figure 1 , including the following steps:

[0025] Step 1: Obtain the track information of the ship detected by the radar, wherein the track information includes the track position, speed, direction, and timestamp information;

[0026] Step 2: Select the first track and the second track To be judged, the track information is predicted using the LSTM network;

[0027] 2-1. Select track data ;in, is the longitude information, is the latitude information, For high information, For speed information, It is the heading information;

[0028] 2-2. The track data of m consecutive moments ~ Input the LSTM network and train it to output the M+1th track data;

[0029] 2-3. Determine the error between the M+1th track data obtained by detection and the M+1th data output by training, iteratively correct the LSTM network parameters, and train to obtain the LSTM prediction model;

[0030] 2-4. Predict the Q+1th track data based on the QMth to Qth track data, and then predict the Q+2th track data based on the Q-M+1th to Q+1th track data. When the detection track data covers the predicted data, use the latest detection track data.

[0031] Step 3: Interpolate the predicted track information to determine the nearest encounter probability and the closest encounter time TCPA; after interpolating and time-aligning the predicted track information, find the closest encounter probability , get the first track position point and the second track position point ;in, is the radar’s measured distance to the target, is the radar’s measurement variance of the target, is the measured elevation angle of the radar to the target; the closest encounter probability The most recent encounter time TCPA is obtained by corresponding to the timestamp.

[0032] Step 4: The nearest encounter probability and the most recent encounter time TCPA are selected as a factor set;

[0033] Step 5: Determine the nearest encounter probability Membership function and evaluation level ;

[0034] The membership function and evaluation level for:

[0035] 5-1. The first track position point and the second track position point Convert to rectangular coordinate system ,get and ;in, , , First Track The coordinate values ​​of the x-axis, y-axis, and z-axis in the radar rectangular coordinate system, , , The second track The coordinate values ​​of the x-axis, y-axis, and z-axis in the radar rectangular coordinate system;

[0036] 5-2. Constructing Statistics ; is the radar measurement covariance matrix, the test statistic follows a three-degree-of-freedom chi-square distribution, and its probability density function is , determine the minimum encounter probability ;

[0037] 5-3. Establish an evaluation set and evaluate the minimum encounter probability Perform level division and set probability values :

[0038] When the probability of encountering ≤ When , it is a very small risk S;

[0039] when ≤ ≤ When, it is a less dangerous MS;

[0040] when ≤ ≤ When M is considered as medium risk;

[0041] when ≤ ≤ When , it is a relatively high risk MB;

[0042] When the probability of encountering ≥ When the risk is high, it is considered as extremely dangerous B;

[0043] 5-4. The empirical formula for constructing the collision probability membership function is:

[0044] 5-5. Combination Figure 2 ,set up , , p 3 =0.5, p 4 =0.35, p 5 =0.65, p 6 =0.5, p 7 =0.8, p 8 =0.8, and obtain the membership value according to the membership function and the minimum encounter probability.

[0045] Step 6: Determine the membership function and evaluation level of the TCPA of the nearest encounter time ;

[0046] The empirical formula for constructing the TCPA of the closest encounter time is:

[0047] Set the time value ,set up , , , , , , , ;

[0048] When we meet recently When the flight path encounters a great danger B;

[0049] when When the flight path is , it will encounter MB that is of great danger;

[0050] when When the track encounters medium danger M;

[0051] when When the flight path is 100 km, it will encounter less dangerous MS;

[0052] when When , the track will encounter extremely small danger S.

[0053] Step 7: Use the analytic hierarchy process to determine the probability weight of the nearest encounter and the weight of the time of the most recent encounter ;

[0054] 7-1. Constructing a judgment matrix: Compare the influence of the parameters in the factor set on the track encounter, and obtain a judgment matrix based on the quantitative evaluation results of the parameter comparison according to the expert scoring method. In the example of the present invention, the judgment matrix is

[0055] ;

[0056] 7-2. After the consistency check of the judgment matrix, the maximum eigenvector is solved according to the judgment matrix, and normalized to obtain the weight of the track encounter risk assessment model: .

[0057] Step 8: Based on the membership function and the weight, a hazard assessment model is constructed using fuzzy set theory to obtain the hazard degree of the track encounter. :

[0058] ;

[0059] when When the track is not in danger, When , the track must have the greatest risk of encountering danger. hour, The larger it is, the more dangerous the track will be.

[0060] Table 1 The first track and the second track data

[0061]

[0062] Table 2 Track encounter danger

[0063]

[0064] According to Tables 1 and 2, the two tracks gradually approach each other, the speed remains stable, and the angles gradually approach each other, which makes the collision risk gradually increase.

[0065] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0066] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A track encounter warning method based on fuzzy comprehensive evaluation, characterized in that: The following steps are involved: Step 1: Obtain the track information of the ship detected by the radar, wherein the track information includes the track position, speed, direction, and timestamp information; Step 2: Select the first track and the second track To be judged, the track information is predicted using the LSTM network; Step 3: Interpolate the predicted track information to determine the nearest encounter probability and the nearest encounter time TCPA; Step 4: The nearest encounter probability and the most recent encounter time TCPA are selected as a factor set; Step 5: Determine the nearest encounter probability Membership function and evaluation level ; Step 6: Determine the membership function and evaluation level of the TCPA of the nearest encounter time ; Step 7: Use the analytic hierarchy process to determine the probability weight of the nearest encounter and the weight of the time of the most recent encounter ; Step 8: Based on the membership function and the weight, a hazard assessment model is constructed using fuzzy set theory to obtain the hazard degree of the track encounter. .

2. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: The step 2 comprises the following steps: 2-1. Select track data ;in, is the longitude information, is the latitude information, For high information, For speed information, It is the heading information; 2-2. The track data of m consecutive moments ~ Input the LSTM network and train it to output the M+1th track data; 2-3. Determine the error between the M+1th track data obtained by detection and the M+1th data output by training, iteratively correct the LSTM network parameters, and train to obtain the LSTM prediction model; 2-4. Predict the Q+1th track data based on the QMth to Qth track data, and then predict the Q+2th track data based on the Q-M+1th to Q+1th track data. When the detection track data covers the predicted data, the latest detection track data is used.

3. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: Determining the nearest encounter probability in step 3 TCPA: After interpolating and time-aligning the predicted track information, find the probability of closest encounter , get the first track position point and the second track position point ;in, is the radar’s measured distance to the target, is the radar’s measurement variance of the target, is the measured elevation angle of the radar to the target; the closest encounter probability The most recent encounter time TCPA is obtained by corresponding to the timestamp.

4. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: The membership function and evaluation level in step 5 for: 5-1. The first track position point and the second track position point Convert to rectangular coordinate system ,get and ;in, , , First Track The coordinate values ​​of the x-axis, y-axis, and z-axis in the radar rectangular coordinate system, , , The second track The coordinate values ​​of the x-axis, y-axis, and z-axis in the radar rectangular coordinate system; 5-2. Constructing Statistics ; is the radar measurement covariance matrix, the test statistic follows a three-degree-of-freedom chi-square distribution, and its probability density function is , determine the minimum encounter probability ; 5-3. Establish an evaluation set and evaluate the minimum encounter probability Perform level division and set probability values : When the probability of encountering ≤ When , it is a very small risk S; when ≤ ≤ When, it is a less dangerous MS; when ≤ ≤ When M is considered as medium risk; when ≤ ≤ When , it is a relatively high risk MB; When the probability of encountering ≥ When the risk is high, it is considered as extremely dangerous B; 5-4. According to the probability density function The significance level and probability value of Using the assignment method, the collision probability membership function is established; 5-5. Obtain a membership value according to the membership function and the minimum encounter probability.

5. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: In step 6, the assignment method is used to determine the membership function and evaluation level of the TCPA of the closest encounter time for: Set the time value ; When we meet recently When the flight path encounters a great danger B; when When the flight path is , it will encounter MB that is of great danger; when When the track encounters medium danger M; when When the flight path is 100 km, it will encounter MS with lesser danger; When we meet recently When , the track will encounter extremely small danger S.

6. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: The step 7 comprises the following steps: 7-1. Constructing a judgment matrix: Compare the influence of the parameters in the factor set on the track encounter, and obtain the judgment matrix based on the quantitative evaluation results of the parameter comparison according to the expert scoring method; 7-2. After the consistency check of the judgment matrix, the maximum eigenvector is solved and normalized to obtain the weights of the track encounter risk assessment model.

7. The method for early warning of track encounter based on fuzzy comprehensive evaluation according to claim 1, characterized in that: The flight path of step 8 may encounter a dangerous : ； when When the track is not in danger, When , the track must have the greatest risk of encountering danger. hour, The larger it is, the more dangerous the track will be.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the computer to execute any one of claims 1 to 7.

10. A computer program product comprising computer program instructions, characterized in that When the computer program instructions are executed on a computer, the computer is caused to perform the method according to any one of claims 1 to 7.

Images