Integrated navigation comprehensive correction method, device and equipment based on intelligent damping switching

Abstract

In order to improve the navigation precision of deep and far sea global navigation of an underwater vehicle, the invention provides an integrated navigation comprehensive correction method, device and equipment based on intelligent damping switching, and in the navigation process of the underwater vehicle, a current fault-tolerant damping navigation positioning result is output based on a damping algorithm; and when the satellite signal can be obtained, different correction methods are selected to correct the navigation positioning result of the current fault-tolerant damping, and the corrected navigation positioning result is output. The short-term precision of navigation is improved through damping algorithm switching, the long-term precision of navigation is improved through a comprehensive correction algorithm, the advantages of damping switching and intelligent comprehensive correction are combined, damping switching and comprehensive correction are carried out according to complex sea conditions and the floating times of the underwater vehicle, and the short-term precision and the long-term precision of navigation positioning are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of integrated navigation, and in particular, to an integrated navigation comprehensive correction method, device and equipment based on intelligent damping switching. Background Art

[0002] Under the background of global navigation in the deep and far sea, the damping algorithm is the main algorithm for inertial and Doppler (DVL) integrated navigation. When the velocity measurement accuracy of Doppler (DVL) deteriorates, the damping algorithm needs to be switched. The damping switching algorithm solves the short-term accuracy of navigation, while the long-term accuracy needs to be improved through an integrated correction algorithm. To maintain the stealth of the submersible, in actual missions, the submersible may surface to obtain satellite signals only once a day or even several days.

[0003] Currently, the integrated correction algorithms include two-point calibration and three-point calibration. Traditional integrated correction algorithms only study the application of one of them in integrated navigation, and do not fully consider the factor that the number of times of obtaining satellite signals cannot be determined in the whole mission. Therefore, the accuracy of the current integrated correction algorithms is limited. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention provides an integrated navigation comprehensive correction method, device and equipment based on intelligent damping switching.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows: On the one hand, the present invention provides an integrated navigation comprehensive correction method based on intelligent damping switching, including: During the navigation of the submersible, output the navigation and positioning result of the current fault-tolerant damping based on the damping algorithm; When satellite signals can be obtained, select different correction methods to correct the navigation and positioning result of the current fault-tolerant damping, and output the corrected navigation and positioning result, including: When the satellite reference point is obtained for the first time, only perform position reset, and the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after position reset; For the next obtained satellite reference point, first judge whether the current satellite reference point is a valid point according to the failure point criterion. If the current satellite reference point is a failure point, perform position reset. The currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after position reset, which improves the short-term positioning effect. If the currently obtained satellite reference point is a valid point, the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after position reset, which improves the short-term positioning effect. If two consecutive satellite reference points are obtained as valid points currently, further estimate and compensate the equivalent azimuth gyro drift on the premise of position reset, perform two-point calibration to correct the current navigation and positioning result of the fault-tolerant damping, and output the navigation and positioning result after two-point calibration, which improves the long-term positioning accuracy. If three consecutive satellite reference points are obtained as valid points currently, further estimate and compensate the equivalent azimuth gyro drift and horizontal gyro drift on the premise of position reset, perform three-point calibration to correct the current navigation and positioning result of the fault-tolerant damping, and output the navigation and positioning result after three-point calibration, which improves the long-term positioning accuracy.

[0006] Furthermore, judge the satellite reference point according to the failure point criterion. The method is as follows: Calculate the matrix corresponding to the current satellite reference point based on the satellite observation matrix and the state transition matrix :

[0007] where represents the state transition matrix from the time when the previous satellite reference point was obtained to the time when the current satellite reference point is obtained , respectively represent time, the satellite observation matrix at time; Calculate the condition number of the matrix corresponding to the current satellite reference point ; Judge whether the current satellite reference point is a failure point according to the condition number of the matrix corresponding to the current satellite reference point.

[0008] Furthermore, when the condition number of the matrix corresponding to the current satellite reference point is greater than the set threshold, it is considered that the condition number is too large, and the matrix corresponding to the current satellite reference point is prone to singularity, the gyro drift result is abnormal, and the current satellite reference point is a failure point. Otherwise, the current satellite reference point is a valid point.

[0009] Further, during the navigation of the submersible, the current measurement residual of the Doppler tester is obtained in real time and input into a pre-trained damping switching threshold prediction model based on an artificial neural network to predict the damping switching threshold online in real time, so as to achieve damping switching.

[0010] Further, a training method for a damping switching threshold prediction model based on an artificial neural network includes: Add different classical noises to the Doppler tester, including but not limited to step noise, periodic oscillation noise, ramp noise, and ramp sine noise, and perform chi-square detection on the navigation data of the Doppler tester after adding different classical noises; Use the K-means method to classify the chi-square detection results of the navigation data of the Doppler tester after adding different classical noises, and obtain the clustering centers under different classical noises as the expected chi-square detection thresholds under the corresponding classical noises; Take the measurement residuals of the Doppler tester corresponding to the navigation data of the Doppler tester after adding different classical noises as input data, and the expected chi-square detection thresholds under different classical noises as labels, and train the constructed artificial neural network to obtain a trained damping switching threshold prediction model based on the artificial neural network.

[0011] On the other hand, a combined navigation comprehensive correction device based on intelligent damping switching is provided, including: A first module for outputting the navigation and positioning result of the current fault-tolerant damping based on a damping algorithm during the navigation of the submersible; A second module for selecting different correction methods to correct the navigation and positioning result of the current fault-tolerant damping and outputting the corrected navigation and positioning result when a satellite signal can be obtained, including: When the satellite reference point is obtained for the first time, only position reset is performed, and the currently output navigation and positioning result is the navigation and positioning result of the current fault-tolerant damping; For the next obtained satellite reference point, first determine whether the current satellite reference point is a valid point according to the failure point criterion. If the current satellite reference point is a failure point, perform position reset. The currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after position reset, which improves the short-term positioning effect. If the currently obtained satellite reference point is a valid point, the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after position reset, which improves the short-term positioning effect. If two consecutive satellite reference points are obtained as valid points currently, on the premise of position reset, further estimate and compensate the equivalent azimuth gyro drift, and perform two-point calibration to correct the current navigation and positioning result of the fault-tolerant damping, and output the navigation and positioning result after two-point calibration, which improves the long-term positioning accuracy. If three consecutive satellite reference points are obtained as valid points currently, on the premise of position reset, further estimate and compensate the equivalent azimuth gyro drift and the horizontal gyro drift, and perform three-point calibration to correct the current navigation and positioning result of the fault-tolerant damping, and output the navigation and positioning result after three-point calibration, which improves the long-term positioning accuracy.

[0012] On the other hand, the present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned combined navigation comprehensive correction method based on intelligent damping switching are realized.

[0013] On the other hand, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned combined navigation comprehensive correction method based on intelligent damping switching are realized.

[0014] On the other hand, the present invention provides a computer program product. The computer program product is stored on a computer-readable storage medium and includes computer instructions. When the computer instructions are run by a processor, the computer device realizes the steps of the above-mentioned combined navigation comprehensive correction method based on intelligent damping switching.

[0015] Compared with the prior art, the technical effects of the present invention are as follows: To maintain the concealment of the submersible, in actual missions, it may take one day or even several days to surface to obtain satellite signals once. The comprehensive correction algorithm has two-point calibration and three-point calibration. The traditional comprehensive correction algorithm only studies the application of one of them in combined navigation and does not make the most efficient use of satellite signals for correction. In order to improve the utilization rate of satellite reference points, the present invention proposes a comprehensive correction algorithm for typical mission scenarios and decides whether to perform position reset according to the failure point criterion. The present invention improves the utilization rate of satellite reference points and improves the long-term accuracy of navigation and positioning.

[0016] The inertial navigation damping algorithm is a prerequisite for comprehensive calibration, which is divided into undamped, internal damping, and external damping states. The damping algorithm switching improves the short-term accuracy of navigation, and the comprehensive calibration algorithm improves the long-term accuracy of navigation. In the face of complex sea conditions in the deep and far seas, by combining the advantages of damping switching and intelligent comprehensive calibration, according to the complex sea conditions and the number of times the submersible floats, damping switching and comprehensive calibration are carried out, improving the short-term and long-term accuracy of navigation and positioning. Brief Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.

[0018] Figure 1 It is a flowchart of a combined navigation comprehensive calibration method based on intelligent damping switching provided in an embodiment; Figure 2 It is a schematic diagram of selecting different calibration methods according to the situation of the acquired satellite reference points in an embodiment; Figure 3 It is a schematic diagram of the navigation trajectory of a submersible in an embodiment; Figure 4 It is a comparison chart of the positioning errors of three navigation and positioning methods in Experiment 1; Figure 5 It is a comparison chart of the positioning errors of three navigation and positioning methods in Experiment 2; Figure 6 It is a comparison chart of the positioning errors of three navigation and positioning methods in Experiment 3. Detailed Embodiments

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0020] Referring to Figure 1 , in one embodiment, a combined navigation comprehensive calibration method based on intelligent damping switching is provided, including: During the navigation of the submersible, based on the damping algorithm, output the navigation and positioning result of the current fault-tolerant damping; When satellite signals can be acquired, select different calibration methods to calibrate the navigation and positioning result of the current fault-tolerant damping, and output the calibrated navigation and positioning result.

[0021] The Doppler velocity log (DVL) is the basis for the inertial Doppler integrated navigation of a submersible. When a speed measurement failure occurs in the DVL, damping needs to be switched. When there is no failure in the DVL, the measurement residual follows a zero-mean Gaussian white noise distribution. Its mean and covariance can be estimated through mathematical statistics methods. At the same time, the navigation system can measure and record the measurement residual. When a speed measurement failure occurs in the DVL, the mean and covariance of the measurement error deviate from those obtained by measurement. Therefore, whether to switch the damping is usually determined by the mean and covariance of the measurement error.

[0022] When the submersible is navigating normally, the DVL has high speed measurement accuracy, and the external damping algorithm is usually adopted. When the submarine makes a maneuvering turn, there is a lateral drift velocity on the submarine hull, and there is a lateral velocity error in the DVL speed measurement, which affects the navigation solution accuracy of the external damping algorithm, and the damping algorithm needs to be switched. When the submarine encounters violent fluctuations in ocean currents during navigation, the instantaneous speed measurement error of the DVL will be relatively severe, and the damping algorithm also needs to be switched. When the submarine makes ascending and descending movements, affected by the salinity, temperature and water depth of the seawater, the change in the speed of sound varies, and the damping algorithm also needs to be switched. In addition, the DVL belongs to an active sonar and is easily detected by anti-submarine sonars during operation.

[0023] When there is no anti-submarine sonar in the submarine navigation area, the DVL can be normally powered on and operate. The damping algorithm of INS / DVL integrated navigation is adopted. When the DVL has high speed measurement accuracy, the external damping algorithm is adopted; when the DVL has low speed measurement accuracy, it is judged according to the submarine maneuvering state. When the submarine has small maneuvers (approximate uniform linear motion), the internal damping is the best algorithm. When the submarine has large maneuvers, the pure inertial navigation solution without damping is the best algorithm. When the submarine passes through the anti-submarine sonar array area, in order to maintain the concealment of the submarine, the DVL needs to be turned off, and the damping algorithm is selected according to the submarine maneuvering state. When the submarine has small maneuvers (such as approximate uniform linear motion), the internal damping is the best algorithm. When the submarine has large maneuvers, the pure inertial navigation solution without damping is the best algorithm.

[0024] The K-means method is a classic clustering algorithm that can divide data into multiple clusters, and the data within each cluster is as similar as possible. In the present invention, the K-means method is used to divide the measurement residuals in various classical noise situations into two categories, namely the normal speed measurement of the DVL and the abnormal speed measurement of the DVL. The clustering center of the abnormal category is regarded as the threshold for damping switching in this noise situation. When the recorded measurement residual reaches this threshold, damping switching can be carried out.

[0025] In another embodiment, during the navigation of the submersible vehicle, the measurement residual of the current Doppler tester is obtained in real time and input into a pre-trained damping switching threshold prediction model based on an artificial neural network to predict the damping switching threshold online in real time, thereby realizing damping switching. Among them, the training method of the damping switching threshold prediction model based on an artificial neural network includes: Add different classical noises to the Doppler tester, including but not limited to step noise, periodic oscillation noise, ramp noise, and ramp sine noise, and perform chi-square detection on the navigation data of the Doppler tester after adding different classical noises; Use the K-means method to classify the chi-square detection results of the navigation data of the Doppler tester after adding different classical noises, and obtain the clustering centers under different classical noises as the expected chi-square detection thresholds under the corresponding classical noises; Take the measurement residual of the Doppler tester corresponding to the navigation data of the Doppler tester after adding different classical noises as the input data, and the expected chi-square detection thresholds under different classical noises as the labels, and train the constructed artificial neural network to obtain a trained damping switching threshold prediction model based on the artificial neural network.

[0026] The traditional comprehensive correction algorithm needs to be carried out under the navigation conditions of equal latitude and low speed, and its practicability is greatly limited. The comprehensive correction algorithm adopted by the present invention includes two-point correction and three-point correction. The rotation modulation inertial navigation can suppress the gyro drift in the horizontal direction. The main error influencing factor is the remaining equivalent azimuth gyro drift. The two-point correction method can estimate the equivalent azimuth gyro drift, but cannot estimate the gyro drift in the horizontal direction. The three-point correction method can estimate the gyro drift in the horizontal direction, and then compensate the gyro constant drift of the system. When there is a satellite signal, matrix operations will be performed during correction. When the condition number of the matrix is too large, the matrix operation is prone to singularity, and the calculated gyro drift result will be abnormal. Such a correction point is called a failure point. The comprehensive correction needs to avoid the failure point. When encountering a failure point, the gyro drift is not calculated, and only the position is reset. According to the order of obtaining the satellite reference points, if the first point is a failure point, the position is reset. If two effective points appear continuously, two-point correction is performed. If three effective points appear continuously, three-point correction is performed. The position reset can replace the inertial navigation positioning result with the satellite reference point, and the positioning error instantly becomes zero, improving the short-term positioning accuracy by means of external information. On the premise of position reset, the comprehensive correction further estimates the gyro drift error, substitutes the gyro drift error into the inertial navigation error model to calculate the positioning error caused by the gyro drift error, and performs positioning error compensation at the inertial navigation positioning output end to reduce the positioning error caused by the internal factor of gyro drift and improve the long-term positioning accuracy.

[0027] In the present invention, the two-point calibration and three-point calibration can be implemented by using the two-point calibration and three-point calibration methods commonly used in the prior art in this field. Preferably, the two-point calibration method adopted is a two-point calibration method based on a normal vector position model, and the three-point calibration method adopted is a three-point calibration method based on a normal vector position model. The two-point calibration method based on a normal vector position model does not require heading reference information and does not require navigation under the conditions of equal latitude and small maneuvers, improving the applicability of the submersible navigation in complex environments. The two-point calibration method based on a normal vector position model can estimate the equivalent azimuth gyro drift, but cannot estimate the horizontal gyro drift. The three-point calibration method based on a normal vector position model can not only estimate the equivalent azimuth gyro drift, but also estimate the horizontal gyro drift, and then compensate the system for the gyro constant drift and correct the position error.

[0028] Specifically, when satellite signals can be obtained, different calibration methods are selected to calibrate the navigation and positioning results of the current fault-tolerant damping according to the order of the obtained satellite reference points and the validity of the satellite reference points, including: When the satellite reference point is obtained for the first time, a position reset is performed, and the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after the position reset; For the next obtained satellite reference points, first judge whether the current satellite reference point is a valid point according to the failure point criterion. If the current satellite reference point is a failure point, a position reset is performed, and the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after the position reset, improving the short-term positioning effect; if the currently obtained one satellite reference point is a valid point, the currently output navigation and positioning result is the navigation and positioning result of the fault-tolerant damping after the position reset, improving the short-term positioning effect; if the currently continuously obtained two satellite reference points are valid points, on the premise of position reset, further estimate and compensate the equivalent azimuth gyro drift, perform two-point calibration to calibrate the navigation and positioning results of the current fault-tolerant damping, and output the navigation and positioning results after two-point calibration, improving the long-term positioning accuracy; if the currently continuously obtained three satellite reference points are valid points, on the premise of position reset, further estimate and compensate the equivalent azimuth gyro drift and horizontal gyro drift, perform three-point calibration to calibrate the navigation and positioning results of the current fault-tolerant damping, and output the navigation and positioning results after three-point calibration, improving the long-term positioning accuracy.

[0029] Refer to Figure 2 In an embodiment of the present invention, the method for judging the satellite reference point according to the failure point criterion is as follows: Calculate the matrix corresponding to the current satellite reference point based on the satellite observation matrix and the state transition matrix : ;

[0030] Where Indicates from the previous obtained satellite reference point The state transition matrix from the moment to the currently obtained satellite reference point at that moment, respectively represent the moment, the satellite observation matrix at the moment; Calculate the condition number of the matrix corresponding to the current satellite reference point ; ; Based on the condition number of the matrix corresponding to the current satellite reference point to determine whether the current satellite reference point is a failure point. When the condition number of the matrix corresponding to the current satellite reference point is greater than the set threshold, it is considered that the condition number is too large, and the matrix corresponding to the current satellite reference point is prone to singularity, the gyro drift result appears abnormal, and the current satellite reference point is a failure point. Otherwise, the current satellite reference point is a valid point. ;

[0031] ; Among them: ; is the angular velocity of the Earth's rotation, is the time difference between two observations. From the moment when the previous satellite reference point was obtained to the moment when the current satellite reference point is obtained, for each , it corresponds to the time difference from the k -1 moment to the k moment, k The value range of is

[0032] ; ; ; Among them is the unit vector from the center of the virtual sphere to the carrier position, including x , y , z the components in three directions , among which is the latitude, is the longitude, is the transverse latitude, is the transverse longitude.

[0033] ; Among them is the time constant related to the Markov process.

[0034] Regarding the condition number For the corresponding set threshold, those skilled in the art can set it according to the actual situation and experience. Without loss of generality, the condition number can be obtained from historical data the normal value range of, obtain the condition number the mean of the normal value range of. When the condition number is more than 100 times the mean of the normal value, it is considered that the condition number is too large, the matrix operation is prone to singularity, and the calculated gyro drift result is abnormal. This calibration point is a failure point; the comprehensive calibration should avoid the failure point, and only perform position reset instead of gyro drift calculation when encountering the failure point.

[0035] The damping algorithm switching improves the short-term accuracy of navigation, and the comprehensive calibration algorithm improves the long-term accuracy of navigation. Traditional methods study the two separately. Facing the complex sea conditions in the deep and far sea, the present invention combines the advantages of the two and proposes a comprehensive calibration method based on intelligent damping switching, which intelligently performs damping switching and comprehensive calibration according to the typical task scenarios of complex sea conditions and the number of times the submersible floats, improving the navigation accuracy and the adaptability of the algorithm.

[0036] To verify the damping switching and comprehensive calibration algorithms, the trajectory of a typical task scenario is adopted. During the navigation process, the submersible successively experiences three typical task scenarios: large maneuver of the submersible, severe fluctuation of ocean currents, and encounter with anti-submarine sonar, and the switching of the fault-tolerant damping algorithm is required; in addition, the submersible floats to the water surface three times at 35h, 55h, and 80h to receive satellite signals, and the satellite signals can be used for comprehensive calibration.

[0037] Refer to Figure 3, which is a schematic diagram of the submersible's navigation trajectory. The submersible starts sailing from a depth of -300m at 90° east longitude and 45° north latitude. The course angle is 45° east of north. It remains stationary for 6h for initial alignment, accelerates uniformly to 5m / s in 20s, and then sails in a straight line at a constant speed of 5m / s. It starts to surface at 30h, reaches the water surface after 310s to receive satellite signals for 80s, and then dives to a depth of -300m after 310s and continues to sail in a straight line at a constant speed of 5m / s. It starts to maneuver at 40h, turns right at 0.5° / s for 180s, continues to sail at a constant speed for 1h, turns left at 0.5° / s for 360s, sails at a constant speed for 2h, turns right at 0.5° / s for 360s, sails at a constant speed for 1h, turns left at 0.5° / s for 180s, and sails at a constant speed until 70h. It starts to surface at 70h, reaches the water surface after 310s to receive satellite signals for 80s, and then dives to a depth of -300m after 310s and continues to sail in a straight line at a constant speed of 5m / s until 80h. At 80h, the DVL is turned off, and it surfaces uniformly to a depth of -40m in 270s, maintains the depth unchanged, continues to sail forward in a straight line at a constant speed of 5m / s for 12h, turns right at 1° / s for 90s, sails forward at a constant speed for 900s, turns left at 1° / s for 90s. It dives to a depth of -300 in 278s and continues to sail forward at a constant speed of 5m / s until 100h. It starts to surface at 100h, reaches the water surface after 310s to receive satellite signals for 80s, and then dives to a depth of -300m after 310s and continues to sail in a straight line at a constant speed of 5m / s until it reaches the end point at 140h.

[0038] (I) Simulation steps To verify the damping switching algorithm, known common noise is added to the Doppler tester, and then its navigation data is processed to obtain the chi-square detection function, which is found to correspond to the fault time. After adding various classical noises to the Doppler tester, chi-square detection is performed for each case. Then the K-means function is used for classification, and the larger clustering center is used as the expected threshold for that case.

[0039] Then, the measurement residuals of the Doppler tester in each case are used as input data, and the corresponding threshold discriminated by the K-means method is used as the label. With the number of hidden layers selected as 8, the above-obtained data is used for network training. And a set of data is extracted for prediction verification.

[0040] To achieve the purpose of real-time predicting the damping switching threshold, finally, the trained neural network is embedded into the integrated navigation program to realize online judgment of the threshold and damping switching, and the offline judgment threshold and online judgment threshold are respectively compared with the expected threshold to observe its stability and accuracy in online use.

[0041] As shown in Table 1, which is a comparison table of threshold judgment for artificial neural networks. When using a neural network to judge the threshold, whether it is offline judgment or online judgment, the effect is good, and the error is within an acceptable range, basically achieving the expected purpose. There are differences between the results of offline threshold judgment and online threshold judgment. It is analyzed that the training data is not enough, resulting in possible differences in the neural networks built each time. After testing, as the training data increases, the neural network becomes more stable, the difference in threshold judgment will continue to decrease, and the prediction effect will also become better and better.

[0042] Table 1 Comparison Table of Threshold Judgment for Artificial Neural Networks

[0043] Next, the submersible floats to the water surface three times at 35h, 55h, and 80h to receive satellite signals, and all are judged to be valid points.

[0044] Refer to Figure 4 , Figure 4 is the comparison chart of positioning errors of three navigation and positioning methods in Experiment 1. Figure 4 In it, "pure inertial navigation" represents navigation and positioning based on pure inertial navigation, "traditional method" represents navigation and positioning using the traditional damping switching algorithm, and "method of this paper" represents damping switching combined with two-point correction for positioning. From Figure 4 it can be seen that the method of damping switching combined with two-point correction for positioning is superior to "pure inertial navigation" and "traditional method". Using the method of damping switching combined with two-point correction for positioning, two-point correction for positioning is carried out at three locations of 35h, 55h, and 80h during the submarine navigation process. The maximum positioning error from 55h to 80h is 2.46 n mile; after 80h, the maximum positioning error and RMS of the method of damping switching combined with two-point correction for positioning are 3.78 n mile and 2.28 n mile respectively.

[0045] Refer to Figure 5 , Figure 5 is the comparison chart of positioning errors of three navigation and positioning methods in Experiment 2. Figure 5 In it, "pure inertial navigation" represents navigation and positioning based on pure inertial navigation, "traditional method" represents navigation and positioning using the traditional damping switching algorithm, and "method of this paper" represents damping switching combined with three-point correction for positioning. From Figure 5 it can be seen that the method of damping switching combined with three-point correction for positioning is superior to "pure inertial navigation" and "traditional method". Using damping switching combined with three-point correction for positioning, three-point correction for positioning is carried out at three locations of 35h, 55h, and 80h during the submarine navigation process. The maximum positioning error from 55h to 80h is 2.59 n mile; after 80h, the maximum positioning error and RMS of the three-point correction algorithm for positioning are 1.00 n mile and 0.70 n mile respectively.

[0046] In summary, the navigation positioning accuracy of damping switching combined with two-point calibration is better than that of damping switching combined with three-point calibration at 55-80h. After 80h, the navigation positioning accuracy of damping switching combined with two-point calibration is worse than that of damping switching combined with three-point calibration. Therefore, in the complex sea conditions of the deep and far sea, combining the advantages of both, the present invention proposes a combined navigation comprehensive calibration method based on intelligent damping switching, which performs two-point calibration when two-point calibration is satisfied and three-point calibration when three-point calibration is satisfied.

[0047] Referring to Figure 6 , Figure 6 Figure 7 is a comparison chart of the positioning errors of three navigation positioning methods in Experiment 3. Figure 6 In it, "pure inertial navigation" represents navigation positioning based on pure inertial navigation, "traditional method" represents navigation positioning using the traditional damping switching algorithm, and "method of the present invention" represents the combined navigation comprehensive calibration method based on intelligent damping switching proposed by the present invention. It can be seen from Figure 6 that the combined navigation comprehensive calibration method based on intelligent damping switching is better than "pure inertial navigation" and "traditional method". Using the combined navigation comprehensive calibration method based on intelligent damping switching, the submersible performs two-point calibration at 55h and three-point calibration at 80h during the navigation process; the navigation positioning results output by two-point calibration are in the interval of 55h-80h, and the navigation positioning results output by three-point calibration are after 80h, with the smallest positioning error of both.

[0048] The combined navigation comprehensive calibration method based on intelligent damping switching proposed by the present invention combines the advantages of fault-tolerant damping switching two-point calibration and fault-tolerant damping switching three-point calibration, and has the highest positioning accuracy.

[0049] Facing the complex sea conditions of long-term navigation in the deep and far sea, compared with the traditional fixed threshold of damping switching, an artificial neural network is introduced in the present invention to select the threshold online in real time to achieve damping switching; the number of times of surfacing cannot be determined during the mission of the submersible, and the traditional method only realizes a single two-point calibration or three-point calibration. The present invention performs comprehensive calibration by making the most efficient use of valid points according to the number of valid points of the actual calibration points. The damping switching algorithm improves the short-term accuracy of navigation positioning, and the comprehensive calibration improves the long-term accuracy of navigation positioning. The present invention performs comprehensive calibration on the basis of intelligent damping switching for typical mission scenarios, combines the advantages of both intelligent damping switching and intelligent comprehensive calibration, and improves the short-term accuracy and long-term accuracy of navigation positioning.

[0050] On the other hand, the present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the combined navigation comprehensive correction method based on intelligent damping switching provided in any of the above embodiments are implemented. The computer device may be a server. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store sample data. The network interface of the computer device is used to communicate with an external terminal through a network connection.

[0051] On the other hand, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the combined navigation comprehensive correction method based on intelligent damping switching provided in any of the above embodiments are implemented.

[0052] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it may include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application may include non-volatile and / or volatile memories. Non-volatile memories may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0053] Matters not described in detail in the present invention are well-known technologies.

[0054] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0055] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation to the scope of the invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

[0056] The above is only the preferred embodiment of the present invention and is not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A combined navigation comprehensive correction method based on intelligent damping switching, characterized in that: include: During the navigation of the submersible, the current fault-tolerant damped navigation positioning result is output based on the damping algorithm; When satellite signals can be acquired, different correction methods are selected to correct the current fault-tolerant and damped navigation and positioning results, and the corrected navigation and positioning results are output, including: When the satellite reference point is first acquired, the position is reset. The navigation positioning result currently output is the navigation positioning result with fault tolerance and damping after the position is reset. For the satellite reference point obtained next, first determine whether the current satellite reference point is a valid point according to the failure point criterion. If the current satellite reference point is a failure point, reset the position, and the navigation positioning result currently output is the navigation positioning result of fault-tolerant damping after the position reset, thereby improving the short-term positioning effect; if one satellite reference point is currently obtained as a valid point, the navigation positioning result currently output is the navigation positioning result of fault-tolerant damping after the position reset, thereby improving the short-term positioning effect; if two satellite reference points are currently obtained continuously as valid points, further estimate and compensate for the equivalent azimuth gyro drift under the premise of position reset, perform two-point calibration to correct the current navigation positioning result of fault-tolerant damping, output the navigation positioning result after the two-point calibration, thereby improving the long-term positioning accuracy; if three satellite reference points are currently obtained continuously as valid points, further estimate and compensate for the equivalent azimuth gyro drift and horizontal gyro drift under the premise of position reset, perform three-point calibration to correct the current navigation positioning result of fault-tolerant damping, output the navigation positioning result after the three-point calibration, thereby improving the long-term positioning accuracy.

2. The integrated navigation comprehensive correction method based on intelligent damping switching according to claim 1 is characterized in that: The satellite reference point is judged according to the failure point criterion, the method is: Calculate the matrix corresponding to the current satellite reference point based on the satellite observation matrix and state transfer matrix : ； in Indicates the satellite reference point obtained from the previous From the time to the current satellite reference point The state transfer matrix at time , , Respectively time, Satellite observation matrix at time; Calculate the matrix corresponding to the current satellite reference point condition number ; According to the matrix corresponding to the current satellite reference point The condition number is used to determine whether the current satellite reference point is a failure point.

3. The integrated navigation comprehensive correction method based on intelligent damping switching according to claim 2 is characterized in that: The matrix corresponding to the current satellite reference point condition number If it is greater than the set threshold, the condition number is considered too large, and the matrix corresponding to the current satellite reference point It is easy to produce singularity, the gyro drift result is abnormal, the current satellite reference point is a failure point, otherwise the current satellite reference point is a valid point.

4. The integrated navigation comprehensive correction method based on intelligent damping switching according to claim 1, 2 or 3, characterized in that: During the navigation of the submersible, the measurement residual of the current Doppler tester is obtained in real time and input into a pre-trained damping switching threshold prediction model based on an artificial neural network, and the damping switching threshold is predicted online in real time to realize damping switching.

5. The integrated navigation comprehensive correction method based on intelligent damping switching according to claim 4 is characterized in that: Different classical noises are added to the Doppler tester, including step noise, periodic oscillation noise, ramp noise, and ramp sine noise, and chi-square test is performed on the navigation data of the Doppler tester after adding different classical noises; The K-means method is used to classify the chi-square detection results of the Doppler tester navigation data after adding different classical noises, and the cluster centers under different classical noises are obtained as the expected chi-square detection thresholds under the corresponding classical noises. The measurement residual of the Doppler tester corresponding to the navigation data of the Doppler tester after adding different classical noises is used as input data, and the expected chi-square detection threshold under different classical noises is used as the label. The constructed artificial neural network is trained to obtain a trained damping switching threshold prediction model based on the artificial neural network.

6. The integrated navigation comprehensive correction device based on intelligent damping switching is characterized in that: include: The first module is used to output the current fault-tolerant damped navigation positioning result based on the damping algorithm during the navigation of the submersible; The second module is used to select different correction methods to correct the current fault-tolerant and damped navigation and positioning results when satellite signals can be acquired, and output the corrected navigation and positioning results, including: When the satellite reference point is first acquired, only the position is reset. The navigation positioning result currently output is the fault-tolerant and damped navigation positioning result after the position is reset. For the satellite reference point obtained next, first determine whether the current satellite reference point is a valid point according to the failure point criterion. If the current satellite reference point is a failure point, reset the position, and the navigation positioning result currently output is the navigation positioning result of fault-tolerant damping after the position reset, thereby improving the short-term positioning effect; if one satellite reference point is currently obtained as a valid point, the navigation positioning result currently output is the navigation positioning result of fault-tolerant damping after the position reset, thereby improving the short-term positioning effect; if two satellite reference points are currently obtained continuously as valid points, further estimate and compensate for the equivalent azimuth gyro drift under the premise of position reset, perform two-point calibration to correct the current navigation positioning result of fault-tolerant damping, output the navigation positioning result after the two-point calibration, thereby improving the long-term positioning accuracy; if three satellite reference points are currently obtained continuously as valid points, further estimate and compensate for the equivalent azimuth gyro drift and horizontal gyro drift under the premise of position reset, perform three-point calibration to correct the current navigation positioning result of fault-tolerant damping, output the navigation positioning result after the three-point calibration, thereby improving the long-term positioning accuracy.

7. The integrated navigation comprehensive correction device based on intelligent damping switching according to claim 6 is characterized in that: In the second module, the satellite reference point is judged according to the failure point criterion, and the method is as follows: Calculate the matrix corresponding to the current satellite reference point based on the satellite observation matrix and state transfer matrix : ； in Indicates the satellite reference point obtained from the previous From the time to the current satellite reference point The state transfer matrix at time , Respectively time, Satellite observation matrix at time; Calculate the matrix corresponding to the current satellite reference point condition number ; According to the matrix corresponding to the current satellite reference point The condition number is used to determine whether the current satellite reference point is a failure point.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the combined navigation comprehensive correction method based on intelligent damping switching as claimed in claim 1 are implemented.

Images