Self-adaptive stability control method for underwater sonar platform

Abstract

The invention relates to the technical field of adaptive control, in particular to an underwater sonar platform adaptive stability control method, which comprises the following steps of acquiring an angular velocity differential feature set, marking a disturbance direction, generating a reverse excitation signal, outputting a compensation rate instruction and completing rolling closed-loop control recording. According to the method, a disturbance trend is dynamically extracted by constructing a rolling angular velocity change feature set, a direction change rule is identified in combination with differential sequence symbol labeling, classification judgment is performed on a reverse relation between a disturbance direction and torque feedback, labeling is performed in combination with a rate response and a position state, and reverse compensation operation adaptive to the disturbance direction is excited. A feedback closed loop among response delay, a deviation relation and a compensation rate is dynamically established, real-time response of control parameters and adaptive matching of a compensation strategy are realized, the attitude adjustment efficiency and rolling control precision of a sonar platform are further improved, and the stability of a sonar detection direction and the environmental adaptability are ensured.

Description

Technical Field

[0001] This invention relates to the field of adaptive control technology, and in particular to an adaptive stabilization control method for an underwater sonar platform. Background Technology

[0002] The field of adaptive control technology mainly involves real-time monitoring of system status and dynamic adjustment of control strategies based on monitoring results to achieve optimal control of system performance in uncertain or changing environments. This includes parameter self-tuning control, adaptive gain adjustment, robust control, and model reference adaptive control. Through online parameter estimation and control law update mechanisms, the control system can maintain stable operation and possess good dynamic performance under unknown disturbances or model uncertainties. Traditional anti-roll stabilization control methods for underwater sonar platforms refer to the use of mechanical stabilization structures for physical vibration isolation or fixed-gain proportional-integral-derivative (PID) controllers to compensate for roll angular velocity when the sonar platform is subjected to roll disturbances. These methods rely on inertial measurement units to collect roll angle data, set fixed thresholds to determine the strength of the disturbance, and then use electric actuators to adjust the platform's attitude to offset the deviation caused by roll. However, these methods lack online adjustment capabilities during the control parameter setting phase, and the compensation mechanism relies on a fixed model matching strategy.

[0003] Existing technologies for anti-roll stabilization control of underwater sonar platforms rely on mechanical vibration isolation or fixed-gain controllers for compensation and adjustment. These technologies lack the ability to identify disturbance characteristics in real time, and the control parameters cannot be dynamically adjusted according to environmental changes after being set. This results in a lag in the system's response when encountering changes in the frequency and amplitude of disturbances, making it unable to maintain a stable attitude adjustment rhythm. In addition, the fixed-threshold method for judging the strength of disturbances lacks sensitivity to the evolution of trends, making it difficult for the platform to activate the compensation mechanism in a timely manner in the early stages of disturbances. The electric actuators rely solely on the model's preset response, ignoring the correlation between the trend of disturbance direction changes and the feedback signal. Ultimately, this leads to inaccurate roll compensation and lag in attitude adjustment, affecting the sonar's directional stability and detection accuracy. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an adaptive stabilization control method for underwater sonar platforms.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: an adaptive stabilization control method for an underwater sonar platform, comprising the following steps: S1: Obtain the original change data of X-axis angular velocity in the gyroscope component of the sonar platform, construct a one-dimensional data sequence of angular velocity change over time, perform differential processing on the change amplitude of angular velocity values ​​between two adjacent frames, establish a differential interval under a uniform sampling duration, and obtain the roll angular velocity change feature set. S2: Based on the roll angular velocity change feature set, read the positive and negative numerical labels between each group of difference partitions, accumulate and mark the number range of the continuous change sequence in the same direction, mark the continuous increasing and decreasing trend in a single direction, determine the disturbance direction, and obtain the roll disturbance trend direction label. S3: Based on the disturbance direction in the roll disturbance trend direction label, combined with the current torque direction feedback value of the magnetic coupling component output shaft, classify the disturbance direction as being in the same direction or opposite direction, determine whether the current disturbance has entered an alternating change state, and if so, generate a disturbance direction opposite excitation signal. S4: Based on the disturbance direction reverse excitation signal, select the default compensation speed when the direction is reversed, drive the output target rate command of the pitch and roll linkage component and record the response delay time, mark the lateral deviation position relationship between the sonar platform and the cabin contact point, and obtain the magnetic coupling structure rate response command annotation data. S5: Based on the data labeled with the rate response command of the magnetic coupling structure, and combined with the change in the angle between the current roll axis and the inertial coordinate system of the sonar platform gyroscope component, it is determined that the platform structure carried by the sonar assembly has completed the roll direction control closed-loop adjustment, and the underwater sonar platform adaptive stability control record is obtained.

[0006] As a further aspect of the present invention, the classification of the direction of disturbance as being in the same direction or opposite direction specifically means that if the torque direction feedback has the opposite sign to the disturbance direction, it is determined to be in the opposite direction; otherwise, it is in the same direction. Specifically, determining whether the current disturbance has entered an alternating change state involves extracting the longitudinal axis feedback displacement change rate sequence of the stable platform when the directions are opposite, determining whether the change rates in two consecutive time periods are in the same direction, and if they are not in the same direction, marking it as a disturbance cancellation trend. If both of the above conditions are met, it is determined that the current disturbance has entered an alternating change state.

[0007] As a further embodiment of the present invention, the roll angular velocity variation feature set includes fluctuation amplitude range, change frequency label, differential direction mark, and feature time series; the roll disturbance trend direction label includes disturbance symbol sequence, trend intensity level, duration mark, and inertial reference direction association parameter; the disturbance direction reverse excitation signal includes direction vector state, disturbance type identifier, torque vector polarity, and alternating change confirmation flag; the magnetic coupling structure rate response command annotation data includes response rate value, compensation delay time, lateral position deviation category, and linkage excitation mode; and the underwater sonar platform adaptive stabilization control record includes disturbance response history, closed-loop control completion status, and angle adjustment trajectory.

[0008] As a further aspect of the present invention, the specific steps for obtaining the roll angular velocity change feature set are as follows: S111: Obtain the raw change data of X-axis angular velocity in the gyroscope component on the sonar platform, set the sampling frequency of the gyroscope sensor, arrange all raw angular velocity data within a fixed sampling period into a time series, construct a one-dimensional data sequence of angular velocity based on the time series index, and generate an angular velocity time series data frame. S112: Based on the angular velocity values ​​of two adjacent frames in the angular velocity time series data frame, extract the angular velocity value pairs corresponding to adjacent index positions respectively, perform difference calculation on the value pairs, arrange the calculation results in time order to generate a difference data column, limit the difference data column to each difference interval within a unified sampling time, extract the difference set in each interval to perform difference amplitude aggregation, and obtain a difference amplitude aggregation value column. S113: Based on the time index position corresponding to the differential amplitude aggregate value column, extract the differential amplitude variation range within each differential interval, integrate and construct the variation range of all intervals, and configure the construction result as a feature matrix representing the roll change trend to establish a roll angular velocity variation feature set.

[0009] As a further aspect of the present invention, the specific steps for obtaining the roll disturbance trend direction label are as follows: S211: Based on the roll angular velocity change feature set, sequentially read the angular velocity difference feature vector in each differential interval, and classify positive and negative according to the sign value of the mean term, construct an ordered sequence composed of sign labels, judge the sign consistency of continuous intervals segment by segment, record the number of continuous frames of each continuous positive or negative interval, form a sign consistency identification matrix, and obtain the unidirectional continuous segment statistics. S212: Based on the statistics of the same-direction continuous segment, extract the positive and negative signs and the duration of each segment record, filter the group with the most continuous changes in the same direction, map the corresponding time interval identifier to the side direction reference axis under the platform structure coordinate system, combine the average trend value of angular velocity in each segment to confirm the direction polarity, calculate and obtain the disturbance direction indication value, filter the direction polarity corresponding to the value higher than the disturbance direction judgment benchmark value, determine the corresponding time segment as the unidirectional disturbance trend interval, and establish the disturbance direction identifier sequence; S213: Based on the time period and corresponding directional label marked in the disturbance direction identification sequence, aggregate and encode the directional labels of all continuous disturbance trend segments to construct a logical sequence of disturbance direction changes on the time axis, and perform cumulative statistics based on the directional characteristics of the disturbance labels to obtain the percentage of disturbance direction duration and the maximum duration direction interval number, and establish a roll disturbance trend direction label.

[0010] As a further aspect of the present invention, the formula for calculating the disturbance direction indication value is as follows: ; in, Indicates the first Segment disturbance direction indication value, For the first Section No. The difference in angular velocity, The directional coefficient represents the current side inertial direction of the platform, and its value is ±1. For the first Number of samples in the segment This is a sign function; the output ±1 indicates the direction of the difference.

[0011] As a further aspect of the present invention, the specific steps for obtaining the disturbance direction reverse excitation signal are as follows: S311: Based on the disturbance direction recorded in the roll disturbance trend direction label, read the torque direction feedback value of the magnetic coupling component output shaft related to the disturbance direction, and determine whether the torque direction feedback and the disturbance direction sign are consistent. If the two signs are opposite, it is considered that the disturbance direction and the torque direction are opposite. If the two signs are consistent, it is considered that the directions are consistent, and a disturbance direction consistency determination result is generated. S312: Based on the determination result of the consistency of the disturbance direction, if the directions are opposite, extract the feedback displacement change rate sequence of the platform longitudinal axis, check whether the change rate in two consecutive time periods is in the same direction. If the change rate signs are opposite in the two time periods, it is marked as the disturbance cancellation trend is established. If the change rate signs are consistent, it is determined that the disturbance trend continues, and the disturbance cancellation trend determination result is obtained. S313: Based on the disturbance cancellation trend determination result, under the conditions that the disturbance direction is reversed and the disturbance cancellation trend is established, determine whether the current disturbance has entered an alternating change state, and simultaneously generate a disturbance direction reverse excitation signal.

[0012] As a further aspect of the present invention, the specific steps for obtaining the magnetic coupling structure rate response command annotation data are as follows: S411: Based on the reverse excitation signal of the disturbance direction, obtain the current speed calibration value of the magnetic coupling component output terminal, perform corresponding mapping in combination with the motor torque and speed control table, match it with the default compensation speed and generate a corresponding speed command, and output the default compensation speed command. S412: According to the default compensation speed command, drive the micro motor assembly of the pitch and roll linkage component to output the target speed command, monitor the response of the motor assembly in real time, record the delay time in the response process, the delay time is the time difference between the motor assembly receiving the speed command and the actual execution, reflecting any delay or lag phenomenon that occurs in the control process, and obtain response delay time data. S413: Based on the response delay time data, the lateral deviation position relationship between the sonar platform and the cabin contact point is used to perform state labeling according to the relationship between the lateral deviation and the disturbance direction, and magnetic coupling structure rate response command labeling data is generated.

[0013] As a further aspect of the present invention, the specific steps for obtaining the adaptive stabilization control record of the underwater sonar platform are as follows: S511: Obtain the labeled data of the rate response command of the magnetic coupling structure, determine the reverse compensation rate by real-time monitoring the lateral deviation and disturbance direction reverse excitation signal data of the contact point between the platform and the cabin, drive the control response, and generate the default compensation speed command. S512: Execute the default compensation speed command, collect the angle change data between the platform's roll axis and the inertial coordinate system in real time through the gyroscope component of the sonar platform, monitor the trend of angle change, determine that the angular velocity will not continue to increase after the disturbance, confirm whether the roll control has completed the closed-loop adjustment, and obtain the roll direction control closed-loop adjustment data. S513: The obtained roll direction control closed-loop adjustment data is compared in real time, and the platform's connection displacement is analyzed to determine whether the platform's attitude is stable. All errors are within the preset range, and an adaptive stability control record for the underwater sonar platform is generated.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, a disturbance trend is dynamically extracted by constructing a feature set of roll angular velocity changes. Combined with differential sequence symbol labeling to identify the direction change law, trend labeling is achieved in the early stage of disturbance, effectively improving the foresight and accuracy of disturbance identification. By classifying and judging the inverse relationship between disturbance direction and torque feedback, and labeling with rate response and position status, the reverse compensation operation adapted to the disturbance direction is accurately triggered, driving the platform linkage actuator to output the target rate command. During the control process, a feedback closed loop between response delay, deviation relationship and compensation rate is dynamically established, realizing real-time response of control parameters and adaptive matching of compensation strategy, thereby improving the attitude adjustment efficiency and roll control accuracy of the sonar platform, and ensuring the stability of sonar detection direction and environmental adaptability. Attached Figure Description

[0015] Figure 1 This is a flowchart of the main steps of the present invention; Figure 2 This is a flowchart of the process for obtaining the roll angular velocity variation feature set in this invention; Figure 3 This is a flowchart of the process for obtaining the roll disturbance trend direction label in this invention; Figure 4 This is a flowchart of the process for obtaining the reverse excitation signal in the perturbation direction according to the present invention; Figure 5 This is a flowchart of the process for obtaining the labeled data of the rate response command of the magnetic coupling structure of the present invention; Figure 6 This is a flowchart of the adaptive stabilization control record acquisition process for the underwater sonar platform of the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] Please see Figure 1 An adaptive stabilization control method for an underwater sonar platform includes the following steps: S1: Obtain the original change data of X-axis angular velocity in the gyroscope component on the sonar platform, set the acquisition frequency to 100Hz, construct a one-dimensional data sequence of angular velocity change over time, perform differential processing based on the change amplitude of angular velocity values ​​between two adjacent frames in the one-dimensional data sequence, establish a differential interval under a uniform sampling duration, and obtain the roll angular velocity change feature set. S2: Based on the roll angular velocity change feature set, read the positive and negative value labels between each group of difference partitions, determine the maximum number of consecutive identical signs, combine the platform side inertial reference direction, accumulate and mark the number range of the continuous change sequence in the same direction, mark the continuous increasing and decreasing trend in a single direction, determine the disturbance direction, and obtain the roll disturbance trend direction label. S3: Based on the disturbance direction recorded in the roll disturbance trend direction label, and combined with the current torque direction feedback value of the magnetic coupling component output shaft, classify and judge whether it is in the same direction or opposite direction with the disturbance direction. If the torque direction feedback and the disturbance direction have opposite signs, it is determined that the directions are opposite; otherwise, the directions are the same. Under the condition of opposite directions, extract the longitudinal axis feedback displacement change rate sequence of the stable platform, and determine whether the change rate in two consecutive time periods is in the same direction. If they are not in the same direction, it is marked that the disturbance cancellation trend is established. When both of the above two conditions are met, it is determined that the current disturbance has entered the alternating change state, and a disturbance direction reverse excitation signal is generated. S4: Based on the excitation signal in the opposite direction of the disturbance, the current rate calibration value of the output end of the magnetic coupling component is mapped to the motor torque and speed control table. The default compensation speed when the direction is reversed is selected. The micro motor component of the pitch and roll linkage component is driven to output the target rate command and record the response delay time. The status is marked by the lateral deviation position relationship between the sonar platform and the contact point of the cabin, and the rate response command annotation data of the magnetic coupling structure is obtained. S5: Based on the data labeled by the rate response command of the magnetic coupling structure, and combined with the change in the angle between the current roll axis and the inertial coordinate system of the gyroscope component on the sonar platform's stabilizing base, it is determined that the platform structure supported by the sonar assembly has completed the closed-loop adjustment of the roll direction control, and the adaptive stabilization control record of the underwater sonar platform is obtained.

[0019] The conditions for completing the closed-loop adjustment of the roll direction control are as follows: The magnetic coupling output direction is consistent with the compensation direction: ensuring that the actuator responds in the opposite direction to the disturbance. The trend of gyroscope X-axis angular velocity variation tends to be stable: the perturbation angular velocity does not show a continuous increasing trend; The connection displacement is controlled within the error range: the attitude stability meets the control closure condition.

[0020] The roll angular velocity variation feature set includes fluctuation amplitude range, change frequency label, differential direction mark, and feature time series. The roll disturbance trend direction label includes disturbance symbol sequence, trend intensity level, duration mark, and inertial reference direction associated parameters. The disturbance direction reverse excitation signal includes direction vector state, disturbance type identifier, torque vector polarity, and alternating change confirmation flag. The magnetic coupling structure rate response command annotation data includes response rate value, compensation delay time, lateral position deviation category, and linkage excitation mode. The underwater sonar platform adaptive stabilization control record includes disturbance response history, closed-loop control completion status, and angle adjustment trajectory.

[0021] Please see Figure 2 The specific steps of S1 are as follows: S111: Obtain the raw change data of X-axis angular velocity in the gyroscope component on the sonar platform, set the sampling frequency of the gyroscope sensor to 100Hz, arrange all raw angular velocity data within a fixed sampling period into a time series, construct a one-dimensional data sequence of angular velocity based on the time series index, and generate an angular velocity time series data frame. To obtain raw X-axis angular velocity data from the gyroscope component on the sonar platform, the sensor monitoring system must first be activated during normal navigation of the sonar platform, ensuring that the gyroscope module has completed self-test and initialization configuration. Then, the sampling frequency is set to 100Hz, meaning a set of angular velocity values ​​is collected every 10 milliseconds. With a sampling period of 10 seconds, a total of 1000 data points are collected. These raw data are then bound to sampling time indices; for example, the first sample is bound to a time of 0.00 seconds, the 1000th sample to a time of 9.99 seconds, forming a complete time series array. If a set of actual collected values ​​is [0.01, 0.02, 0.00, -0.01, -0.03] rad / s, the corresponding time index is... [0.00, 0.01, 0.02, 0.03, 0.04] seconds. Based on this, all angular velocity data frames are organized into time series data frames according to timestamp order and transformed into a one-dimensional array structure. In practical applications, such as when shipborne equipment continuously collects navigation attitude information in complex sea conditions, this structure can ensure the dynamic evolution of the X-axis roll state recorded over time. Subsequently, the sampled data in the time series is processed into equally spaced segments, dividing 1000 sets of data into 10 segments, each segment containing 100 sets of samples, forming 10 sub-matrices. Each matrix corresponds to the angular velocity recording sequence of a time period. This structure can realize the grouping basis for subsequent difference calculation and interval feature aggregation, and finally obtain the angular velocity time series data frame.

[0022] S112: Based on the angular velocity values ​​of two adjacent frames in the angular velocity time series data frame, extract the angular velocity value pairs corresponding to adjacent index positions respectively, perform difference calculation on the value pairs, arrange the calculation results in time order to generate a difference data column, limit the difference data column to each difference interval within a uniform sampling time, extract the difference set in each interval and aggregate the difference amplitude to obtain the difference amplitude aggregated value column. Based on the data segments already divided in the angular velocity time series data frame, the angular velocity values ​​of any two adjacent frames are extracted and the difference is processed, i.e., a value is set... The angular velocity value at time is , The angular velocity value at time is The difference is ,For example , ,but The difference calculation is performed on 100 samples in each segment, resulting in 99 difference values, forming a difference data column. Each segment can be constructed into a difference array such as [-0.01, 0.03, -0.02, 0.01, ...]. Then, the absolute value, mean, and range are calculated on the difference arrays. The formula for the absolute value and mean of the difference is... Taking the difference sequence [-0.01, 0.03, -0.02, 0.01] as an example, The range calculation formula is: The result is The difference statistical features of each segment are combined into vector pairs [0.0175, 0.05]. The above calculation process is repeated for each group of segments to construct 9 groups of vectors, resulting in a total of 9×2-dimensional data structure. A difference feature column arranged in time order is constructed, which is denoted as the difference amplitude aggregation value column.

[0023] Table 1 Characteristics of the Differential Interval of Roll Angular Velocity Table 1 lists the specific values ​​of the mean and range calculated based on each difference interval, recording the statistical characteristics of the change in roll angular velocity.

[0024] S113: Based on the time index position corresponding to the differential amplitude aggregate value column, extract the differential amplitude variation range within each differential interval, integrate and construct the variation range of all intervals, and configure the construction result as a feature matrix representing the roll change trend to establish a roll angular velocity change feature set. Based on the difference feature vectors of each interval in the aggregated difference magnitude value column, a two-dimensional feature matrix needs to be constructed. Each pair of mean and range values ​​is combined into a two-dimensional vector in chronological order. For example, the second interval is [0.015, 0.05]. A total of 9 pairs constitute a 9×2 feature matrix. Next, an operation to identify the inter-interval change trend is performed on this feature matrix. The Euclidean distance function is used to calculate the degree of change between two adjacent vectors. The Euclidean distance formula is... ,in The mean, For the range, taking intervals 6 and 7 as examples, the vector changes from [0.016, 0.03] to [0.019, 0.06]. Substituting these values ​​into the calculation yields... The result is compared with the preset fluctuation identification benchmark value of 0.025 rad / s. If the Euclidean distance is greater than the benchmark value, it is marked as a fluctuation range. The setting value of 0.025 is set to 1.5 times the mean standard deviation based on the ship's roll dynamic response amplitude in the experimental scenario. It belongs to the abnormal judgment boundary that conforms to the sea state change. Finally, the identified feature vectors are marked as roll feature points, and the complete feature matrix formed by them is the roll angular velocity change feature set.

[0025] Please see Figure 3The specific steps of S2 are as follows: S211: Based on the roll angular velocity change feature set, sequentially read the angular velocity difference feature vector in each differential interval, and classify positive and negative according to the sign value of the mean term, construct an ordered sequence composed of sign labels, judge the sign consistency of continuous intervals segment by segment, record the number of continuous frames of each continuous positive or negative interval, form a sign consistency identification matrix, and obtain the statistic of the same direction continuous segment. Based on the differential interval feature vector constructed from the roll angular velocity variation feature set, the mean angular velocity term in each interval needs to be read one by one, and its numerical sign needs to be determined as positive or negative. The sign determination method is: if the mean is greater than zero, it is recorded as +1, and if it is less than zero, it is recorded as -1. The signs of all segments are sequentially arranged into a label sequence, in the form of [+1, +1, +1, -1, -1, +1]. This sequence is used to analyze the continuous trend of direction. Then, a continuous sign consistency check is performed on the label sequence. Specifically, adjacent labels are compared for equality. If they are consecutively equal, they are grouped into the same group, and the start and end values ​​are recorded. The bundle index and the frame length of the group are used. For example, in the above sequence, the first group has indices 0-2 (3 frames), the second group has indices 3-4 (2 frames), and the third group has index 5 (1 frame). All group results are recorded in tabular form, with each row including "starting frame index, ending frame index, number of frames, and symbol". A symbol consistency identification matrix is ​​constructed. Then, the segment with the largest number of frames in all groups is located. This segment is the segment with the strongest unidirectional persistence. The interval number, frame length, and symbol label of the segment are used as input data for subsequent direction recognition operations. Finally, the unidirectional persistence segment statistics are obtained.

[0026] S212: Based on the statistics of continuous segments in the same direction, extract the positive and negative signs and duration of each segment. Filter the group with the most frames showing continuous change in the same direction, map the corresponding time interval identifier to the side direction reference axis in the platform structure coordinate system, and confirm the direction polarity by combining the average trend value of angular velocity within each segment, using the formula: ; The perturbation direction indication value is obtained through calculation. Directional polarities corresponding to values ​​higher than the perturbation direction determination benchmark are selected, and the corresponding time intervals are determined as unidirectional perturbation trend intervals, thus establishing a perturbation direction identification sequence. Indicates the first Segment disturbance direction indication value, For the first Section No. The difference in angular velocity, The directional coefficient represents the current side inertial direction of the platform, and its value is ±1. For the first Number of samples in the segment This is a sign function; the output ±1 indicates the direction of the difference. Based on the statistics of continuous segments in the same direction, combined with the segment number, sign direction, and frame length data, the sample segment with the most consecutive frames is extracted. The angular velocity difference array in this segment is combined with the directional inertial reference axis in the side direction for calculation. First, the sign direction is determined. The value is +1 or -1, and it is multiplied one by one with each difference value point. The absolute value of the product is then taken and summed. The result is divided by the product of the square root of the number of samples in that segment and the absolute value of the sum of the sign function, thus obtaining the disturbance direction indication value.

[0027] The benchmark value for determining the direction of disturbance is set at 0.005, based on the reliability and stability requirements of disturbance trends in actual usage scenarios. This benchmark value was determined experimentally and gradually optimized in practical applications. Its setting considers the ratio of disturbance amplitude to background noise during normal platform navigation. Specifically, the benchmark value of 0.005 is an empirical value obtained by measuring and analyzing the changes in the platform's roll rate under multiple sea states. This value can effectively distinguish between small background changes and obvious disturbance trends, ensuring accurate detection of important directional changes even when the disturbance signal is weak. This benchmark value may be adjusted depending on the platform's operating conditions (such as changes in speed and sea state), but it is generally kept within a sufficiently low range to ensure accurate differentiation between the presence and absence of disturbances and avoid missing potential disturbance directions.

[0028] For example, for section number 2, the input parameter is... , =[-0.015, -0.018, -0.021, -0.018], Lateral heading coefficient Their signs are all negative. Substitute into the calculation: If its value is greater than the disturbance direction determination benchmark value of 0.005, it indicates that the directional disturbance in this segment is persistent, the directional label is valid, the disturbance direction is negative, and thus the disturbance direction identification sequence is obtained.

[0029] The formula's operational logic primarily measures the directional strength and persistence of disturbances in each segment, by comprehensively considering factors such as the magnitude of the angular velocity difference, sign consistency, and sample size. Specifically, the operational logic for each parameter in the formula is as follows: First, the numerator of the formula The weighted sum of the absolute values ​​of the angular velocity differences within each segment was calculated. Each angular velocity difference... This represents the change in angular velocity at each time point within the segment, multiplied by After calculating the (lateral directionality coefficient), the absolute value of the result reflects the absolute contribution of the variation amplitude at that point in time. Lateral Directional Coefficient The value is ±1, representing the alignment with the platform's side reference direction; therefore, the product... It can distinguish between positive and negative directions, corresponding to different disturbance trends. By calculating the absolute value of the weighted sum of all differences, it can quantify the total intensity of the disturbance within the entire segment.

[0030] denominator middle, This represents the number of sample points within a segment. Taking the square root is to standardize the sample size, ensuring fair comparison of segments with different sample sizes. (Sign function) Its function is to determine the difference in angular velocity. In the direction, if If it is positive, then The value is -1 if the difference is positive and -1 otherwise. By summing, we obtain the sum of all signs within the segment. If all the difference signs are the same (i.e., all positive or all negative), the absolute value of this value is the segment length; if the signs are not the same, the value is smaller.

[0031] Ultimately, the ratio of the numerator to the denominator... The indicator value represents the disturbance direction indices of a segment, quantifying the disturbance intensity and directional consistency within that segment. This indicator value allows us to determine whether the disturbance in that segment has significant persistence and directionality, thus enabling the selection of the dominant disturbance direction in subsequent processing.

[0032] The disturbance direction indication value indicates the strength and consistency of the disturbance direction of the roll angular velocity change within a specific time interval. Specifically, it measures the degree of combination between the amplitude (i.e., the difference) of the angular velocity change and the consistency of the direction within that interval. A higher indication value indicates that the angular velocity change within that interval not only has a larger amplitude but also maintains a consistent direction, forming a clear and persistent disturbance trend. Conversely, a lower indication value indicates that the disturbance within that interval is relatively weak or the direction changes more frequently, making it difficult to distinguish as a single disturbance trend. This indication value, by weighted summation of the absolute values ​​of the angular velocity differences and combined with the judgment of sign consistency within the interval, can effectively identify disturbance directions with persistence and stability. The disturbance direction indication value is a quantitative evaluation of the disturbance strength and directional consistency. Its level directly affects the subsequent identification and labeling of the disturbance trend direction. In practical applications, it helps to determine the directional characteristics of the platform under disturbance during motion and its duration, thus providing a basis for further dynamic analysis and control strategies.

[0033] S213: Based on the time period and corresponding directional label marked in the disturbance direction identification sequence, aggregate and encode the directional labels of all continuous disturbance trend segments to construct a logical sequence of disturbance direction changes on the time axis, and perform cumulative statistics based on the directional characteristics of the disturbance labels to obtain the percentage of disturbance direction duration and the maximum duration direction interval number, and establish the roll disturbance trend direction label. Based on the directional labels and corresponding time period numbers marked in the disturbance direction identification sequence, the segments whose directions have been confirmed are aggregated segment by segment. First, index extraction is performed on all segments marked as valid disturbance directions to obtain their start and end time ranges in the original time series, combined with the disturbance direction indices. The judgment logic for determining a result greater than the perturbation direction judgment benchmark value of 0.005 groups each valid segment according to its sign direction value. If multiple adjacent segments have the same directional label and their time intervals are continuous or the interval does not exceed one sampling period (e.g., 0.01 seconds), they can be regarded as the same perturbation trend block. Based on this, all continuous perturbation segments are merged into a trend interval to form an aggregated perturbation direction segment. For example, if segments 2 and 3 are both labeled with direction −1, and their end and start times are 1.99 seconds and 2.00 seconds respectively, they can be merged into a segment with a duration of 1.99–2.59 seconds. For continuous disturbance trend segments, this operation sequentially traverses all identified segments to construct a complete disturbance trend timeline. Then, it calculates the duration of segments with the same directionality in the timeline, accumulates the disturbance duration values ​​under each direction, and counts the frequency of occurrence to form a disturbance directionality distribution vector. At the same time, it extracts the trend segment with the longest cumulative time, assigns it a number, and assigns it a corresponding time index. For example, if the disturbance trend segment with direction −1 lasts for 3.0 seconds and the segment with direction +1 lasts for 1.5 seconds, then direction −1 is used as the main disturbance direction label, and its segment number is recorded as the trend identification output. Finally, the roll disturbance trend direction label is established.

[0034] Please see Figure 4 The specific steps of S3 are as follows: S311: Based on the disturbance direction recorded in the roll disturbance trend direction label, read the torque direction feedback value of the magnetic coupling component output shaft related to the disturbance direction, and determine whether the torque direction feedback and the disturbance direction sign are consistent. If the two signs are opposite, it is considered that the disturbance direction and the torque direction are opposite. If the two signs are consistent, it is considered that the directions are consistent, and a disturbance direction consistency judgment result is generated. Based on the disturbance direction recorded in the roll disturbance trend direction label, the system first obtains the torque direction feedback value of the magnetic coupling component output shaft related to the disturbance direction. This value represents the inertial force feedback of the current platform, and its range is typically ±5 Nm. If this value has the opposite sign to the disturbance direction, the disturbance direction is considered to be opposite to the torque direction. Therefore, the system first compares the torque feedback signal with the signs in the disturbance direction label one by one. If the signs are different (e.g., the torque direction is positive, and the disturbance direction is negative), they are determined to be opposite directions; if the signs are the same, they are determined to be the same direction. The system records this result and performs preprocessing for subsequent disturbance trend determination.

[0035] Table 2 Comparison of Disturbance Direction and Torque Feedback As shown in Table 2, the system checks the consistency of the signs of the torque direction feedback and the disturbance direction label in each sampling period, and generates the final direction consistency judgment result for subsequent processing.

[0036] S312: Based on the result of the consistency of the disturbance direction, if the directions are opposite, extract the feedback displacement change rate sequence of the platform's longitudinal axis, check whether the change rate in two consecutive time periods is in the same direction. If the change rate signs are opposite in the two time periods, it is marked as the disturbance cancellation trend is established. If the change rate signs are consistent, it is determined that the disturbance trend continues, and the result of the disturbance cancellation trend determination is obtained. In the case of opposite directions, the system further extracts the longitudinal axis feedback displacement rate sequence of the stable platform. This sequence reflects the platform's displacement dynamics in the longitudinal direction. For example, when the platform moves in the opposite direction in the longitudinal direction, the rate of change will increase or decrease significantly. The feedback displacement sequence is usually sampled every 0.01 seconds, with units of mm / s. The system extracts the rate of change for two consecutive time periods from this feedback sequence and compares their signs sequentially. If the signs of the rate of change in the two time periods are opposite (i.e., one time period is positive and the other is negative), it is considered that the platform's displacement has undergone reverse cancellation, that is, the platform's disturbance trend has been effectively suppressed, and the system will mark that time period as a period of disturbance cancellation trend.

[0037] Table 3. Sequence of longitudinal displacement rate of the stable platform As shown in Table 3, the system compares the signs of two adjacent time intervals in the rate of change sequence. If the signs are opposite, it indicates that the disturbance cancellation trend is valid. For example, between 0.02 seconds and 0.03 seconds, the displacement rate of change changes from -0.8 mm / s to +2.0 mm / s. The system recognizes the sign reversal and ultimately concludes that the disturbance cancellation trend is valid.

[0038] S313: Based on the disturbance cancellation trend determination result, under the conditions that the disturbance direction is reversed and the disturbance cancellation trend is established, determine whether the current disturbance has entered an alternating change state, and simultaneously generate a disturbance direction reverse excitation signal; When both the disturbance direction consistency determination result and the disturbance cancellation trend determination result meet the conditions, the system further determines whether the disturbance has entered an alternating change state, that is, whether the reverse and same direction changes of the disturbance alternate. The system will determine whether the disturbance has begun to alternate by comparing the change pattern of the disturbance trend in different time periods, especially the sign change of the platform's longitudinal feedback rate of change. If the disturbance direction in two consecutive time periods shows a periodic change (for example, the disturbance direction is negative, then becomes positive, and alternates repeatedly), it is determined that the disturbance has entered an alternating change state, and a disturbance direction reverse excitation signal is generated. This signal can be used as the activation condition of the subsequent control system to start the reverse compensation mechanism of the control system when the disturbance alternates, so as to ensure the stability of the platform.

[0039] Table 4. Determination of the direction of disturbance and the excitation signal in the opposite direction. As shown in Table 4, the system determines whether a disturbance has entered an alternating change state based on the periodic change of the disturbance direction (alternating between positive and negative changes). During the time interval of 0.03 seconds to 0.04 seconds, when the disturbance direction changes from positive to negative, the system records that the disturbance cancellation trend has been established and generates a disturbance direction reversal excitation signal. This signal will trigger the reverse compensation mechanism of the control system to ensure the stable operation of the platform.

[0040] Please see Figure 5 The specific steps of S4 are as follows: S411: Based on the reverse excitation signal of the disturbance direction, obtain the current speed calibration value of the magnetic coupling component output, perform corresponding mapping with the motor torque and speed control table, match with the default compensation speed and generate the corresponding speed command, and output the default compensation speed command. Based on the reverse excitation signal in the disturbance direction, the system first obtains the current rate calibration value from the magnetic coupling component. This value reflects the platform's current operating rate and is typically monitored in real-time by sensors on the magnetic coupling device, generating corresponding calibration data. At this point, the rate calibration value (e.g., 4.5 Nm / s) is mapped to the motor's torque-speed control table, which sets different speeds according to different rate values. The system selects the default compensation speed for reverse operation based on the data in this mapping table. For example, when the rate calibration value is 4.5 Nm / s, the control table returns a corresponding compensation speed of 15 rpm (revolutions per minute). The system uses this compensation speed as a control command to drive the micro-motor assembly of the pitch and roll linkage components. This process ensures that the motor assembly can provide the necessary reverse rate command based on the reverse excitation signal in the disturbance direction, ultimately generating the default compensation speed command.

[0041] Table 5. Examples of Torque-Speed ​​Control As shown in Table 5, the correspondence between the speed calibration value and the compensation speed is mapped through the motor speed control table, and the system selects the corresponding compensation speed according to the input speed value.

[0042] S412: Based on the default compensation speed command, the micro motor assembly driving the pitch and roll linkage component outputs the target speed command, monitors the response of the motor assembly in real time, records the delay time in the response process, the delay time is the time difference between the motor assembly receiving the speed command and the actual execution, reflects any delay or lag phenomenon that occurs in the control process, and obtains the response delay time data. Based on the default compensated speed command, the system sends a corresponding speed command (e.g., 15 rpm) to the micromotor assembly via the control motor driver. Upon receiving the command, the motor assembly begins execution and feeds back the actual speed via sensors. Simultaneously, the system records the real-time delay time during the motor assembly's response—the time difference between receiving the command and achieving the actual speed. For example, in actual testing, if the command is 15 rpm and the motor assembly reaches that speed after 0.03 seconds, the system records the response delay time as 0.03 seconds. This delay time helps evaluate the responsiveness of the motor control system and is used for subsequent system optimization. Ultimately, the response delay time data is recorded and used for subsequent control optimization.

[0043] Table 6 Motor response delay time data As shown in Table 6, the motor response delay time recorded by the system can reflect the delay that may occur in the real-time control process and provide data support for subsequent optimization.

[0044] S413: Based on response delay time data, the lateral deviation position relationship between the sonar platform and the cabin contact point is used to perform state labeling according to the relationship between the lateral deviation and the disturbance direction, and generate magnetic coupling structure rate response command labeling data. When a disturbance in the opposite direction is triggered by an excitation signal, the system monitors the contact point between the cabin and the platform in real time, paying particular attention to changes in the platform's lateral deviation position to infer whether the disturbance has caused a significant state change. For example, if the contact point between the platform and the cabin experiences a lateral deviation after the disturbance reverses (e.g., from 0.5mm to 1.2mm), the system marks this as a state change and associates this data with the rate response command, ultimately generating labeled data for the magnetically coupled structure rate response command. This labeled data provides a basis for subsequent analysis of the disturbance's impact on platform stability. By monitoring the deviations of multiple contact points in real time, the system can refine the analysis of the disturbance's impact on various parts of the platform, ensuring that the platform can adapt to dynamically changing external environments.

[0045] Table 7 Lateral Deviation Data of Platform Contact Points As shown in Table 7, by monitoring the changes in the position of the lateral deviation and the direction of the disturbance, the system can generate accurate labeled data of the rate response command of the magnetically coupled structure, which provides effective data support for subsequent disturbance trend control.

[0046] Please see Figure 6 The specific steps of S5 are as follows: S511: Obtain the labeled data of the magnetic coupling structure rate response command, determine the reverse compensation rate by real-time monitoring the lateral deviation of the contact point between the platform and the cabin and the reverse excitation signal data of the disturbance direction, drive the control response, and generate the default compensation speed command. Based on the rate response command annotation data of the magnetic coupling structure, the system first obtains the rate calibration value from the platform's magnetic coupling components and monitors disturbance signals through sensors to detect the deviation between the platform and the cabin contact point in real time. The system then maps the rate calibration value to the motor control table; the mapping result determines the compensation speed, which in turn adjusts the reverse response signal to drive the micro-motor component to execute the compensation speed. During system monitoring, the system adjusts the compensation speed and verifies the effectiveness of the rate command to ensure stable platform operation. For example, when the rate calibration value is 5.5 Nm / s, the system determines the compensation speed to be 20 rpm through control table mapping. The command data will be used for platform control response, further executing control system commands to ultimately obtain the default compensation speed command. Table 8. Mapping Table of Rate Calibration Values ​​and Compensated Speed As shown in Table 8, the system ensures that the platform can respond to roll direction control according to preset commands by mapping the rate calibration value to the compensation speed.

[0047] S512: Executes the default compensation speed command, collects the angle change data between the platform's roll axis and the inertial coordinate system in real time through the gyroscope component of the sonar platform, monitors the trend of angle change, determines that the angular velocity will not continue to increase after the disturbance, confirms whether the roll control has completed the closed-loop adjustment, and obtains the roll direction control closed-loop adjustment data. By combining the gyroscope components on the sonar platform's stabilizing base, the system collects data on the changes in the angle between the platform's roll axis and the inertial coordinate system, and monitors the trend of the gyroscope's X-axis angular velocity in real time. At this point, the system determines whether the angular velocity shows a continuous increasing trend. By comparing historical data with the current angular velocity change, it ensures that the impact of disturbances on the platform is within a controllable range. If the angular velocity fluctuation tends to stabilize (for example, the angular velocity change decreases from 0.15 rad / s to 0.02 rad / s), the system confirms that the roll control has completed closed-loop adjustment, indicating that platform stability has been controlled, and finally, the closed-loop adjustment data for roll direction control is obtained. Table 9. Gyroscope X-axis angular velocity variation data As shown in Table 9, by monitoring the changing trend of the X-axis angular velocity, the system can determine in real time whether the roll control has completed closed-loop adjustment.

[0048] S513: Real-time comparison of the acquired roll direction control closed-loop adjustment data, combined with the platform's connection displacement, analyzes whether the platform's attitude is stable, determines that all errors are within the preset range, and generates an adaptive stability control record for the underwater sonar platform. Based on the roll direction control closed-loop adjustment data, the system further collects the platform's connection displacement data and compares it with a preset error range to ensure the stability of the platform's attitude. During this process, the system determines whether the displacement error is within an acceptable range (e.g., displacement error ≤ 0.2mm). If the error meets the standard, it is marked as stable, and the system generates the final control record data. Ultimately, the system generates adaptive stabilization control records for the underwater sonar platform. These records provide a complete control process, ensuring that the platform's anti-roll performance meets the expected requirements. Table 10 Record of Roll Control Measures for Underwater Sonar Platforms As shown in Table 10, by monitoring displacement error, the system can confirm that the underwater sonar platform has completed anti-roll adaptive stabilization control.

[0049] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. An adaptive stabilization control method for an underwater sonar platform, characterized in that, Includes the following steps: S1: Obtain the original change data of X-axis angular velocity in the gyroscope component of the sonar platform, construct a one-dimensional data sequence of angular velocity change over time, perform differential processing on the change amplitude of angular velocity values ​​between two adjacent frames, establish a differential interval under a uniform sampling duration, and obtain the roll angular velocity change feature set. S2: Based on the roll angular velocity change feature set, read the positive and negative numerical labels between each group of difference partitions, accumulate and mark the number range of the continuous change sequence in the same direction, mark the continuous increasing and decreasing trend in a single direction, determine the disturbance direction, and obtain the roll disturbance trend direction label. S3: Based on the disturbance direction in the roll disturbance trend direction label, combined with the current torque direction feedback value of the magnetic coupling component output shaft, classify the disturbance direction as being in the same direction or opposite direction, determine whether the current disturbance has entered an alternating change state, and if so, generate a disturbance direction opposite excitation signal. S4: Based on the disturbance direction reverse excitation signal, select the default compensation speed when the direction is reversed, drive the output target rate command of the pitch and roll linkage component and record the response delay time, mark the lateral deviation position relationship between the sonar platform and the cabin contact point, and obtain the magnetic coupling structure rate response command annotation data. S5: Based on the data labeled with the rate response command of the magnetic coupling structure, and combined with the change in the angle between the current roll axis and the inertial coordinate system of the sonar platform gyroscope component, it is determined that the platform structure carried by the sonar assembly has completed the roll direction control closed-loop adjustment, and the underwater sonar platform adaptive stability control record is obtained.

2. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The classification of the direction of disturbance as being in the same direction or opposite direction is as follows: if the torque direction feedback has the opposite sign to the disturbance direction, it is determined to be in the opposite direction; otherwise, it is in the same direction. Specifically, determining whether the current disturbance has entered an alternating change state involves extracting the longitudinal axis feedback displacement change rate sequence of the stable platform when the directions are opposite, determining whether the change rates in two consecutive time periods are in the same direction, and if they are not in the same direction, marking it as a disturbance cancellation trend. If both of the above conditions are met, it is determined that the current disturbance has entered an alternating change state.

3. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The roll angular velocity variation feature set includes fluctuation amplitude range, change frequency label, differential direction marker, and feature time series. The roll disturbance trend direction label includes disturbance symbol sequence, trend intensity level, duration marker, and inertial reference direction associated parameters. The disturbance direction reverse excitation signal includes direction vector state, disturbance type identifier, torque vector polarity, and alternating change confirmation flag. The magnetic coupling structure rate response command annotation data includes response rate value, compensation delay time, lateral position deviation category, and linkage excitation mode. The underwater sonar platform adaptive stabilization control record includes disturbance response history, closed-loop control completion status, and angle adjustment trajectory.

4. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The specific steps for obtaining the roll angular velocity change feature set are as follows: S111: Obtain the raw change data of X-axis angular velocity in the gyroscope component on the sonar platform, set the sampling frequency of the gyroscope sensor, arrange all raw angular velocity data within a fixed sampling period into a time series, construct a one-dimensional data sequence of angular velocity based on the time series index, and generate an angular velocity time series data frame. S112: Based on the angular velocity values ​​of two adjacent frames in the angular velocity time series data frame, extract the angular velocity value pairs corresponding to adjacent index positions respectively, perform difference calculation on the value pairs, arrange the calculation results in time order to generate a difference data column, limit the difference data column to each difference interval within a unified sampling time, extract the difference set in each interval to perform difference amplitude aggregation, and obtain a difference amplitude aggregation value column. S113: Based on the time index position corresponding to the differential amplitude aggregate value column, extract the differential amplitude variation range within each differential interval, integrate and construct the variation range of all intervals, and configure the construction result as a feature matrix representing the roll change trend to establish a roll angular velocity variation feature set.

5. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The specific steps for obtaining the roll disturbance trend direction label are as follows: S211: Based on the roll angular velocity change feature set, sequentially read the angular velocity difference feature vector in each differential interval, and classify positive and negative according to the sign value of the mean term, construct an ordered sequence composed of sign labels, judge the sign consistency of continuous intervals segment by segment, record the number of continuous frames of each continuous positive or negative interval, form a sign consistency identification matrix, and obtain the unidirectional continuous segment statistics. S212: Based on the statistics of the same-direction continuous segment, extract the positive and negative signs and the duration of each segment record, filter the group with the most continuous changes in the same direction, map the corresponding time interval identifier to the side direction reference axis under the platform structure coordinate system, combine the average trend value of angular velocity in each segment to confirm the direction polarity, calculate and obtain the disturbance direction indication value, filter the direction polarity corresponding to the value higher than the disturbance direction judgment benchmark value, determine the corresponding time segment as the unidirectional disturbance trend interval, and establish the disturbance direction identifier sequence; S213: Based on the time period and corresponding directional label marked in the disturbance direction identification sequence, aggregate and encode the directional labels of all continuous disturbance trend segments to construct a logical sequence of disturbance direction changes on the time axis, and perform cumulative statistics based on the directional characteristics of the disturbance labels to obtain the percentage of disturbance direction duration and the maximum duration direction interval number, and establish a roll disturbance trend direction label.

6. The adaptive stabilization control method for an underwater sonar platform according to claim 5, characterized in that, The formula for calculating the disturbance direction indication value is: ； in, Indicates the first Segment disturbance direction indication value, For the first Section No. The difference in angular velocity, This represents the directional coefficient of the platform's current side inertial direction. For the first Number of samples in the segment It is a symbolic function.

7. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The specific steps for obtaining the reverse excitation signal in the perturbation direction are as follows: S311: Based on the disturbance direction recorded in the roll disturbance trend direction label, read the torque direction feedback value of the magnetic coupling component output shaft related to the disturbance direction, and determine whether the torque direction feedback and the disturbance direction sign are consistent. If the two signs are opposite, it is considered that the disturbance direction and the torque direction are opposite. If the two signs are consistent, it is considered that the directions are consistent, and a disturbance direction consistency determination result is generated. S312: Based on the determination result of the consistency of the disturbance direction, if the directions are opposite, extract the feedback displacement change rate sequence of the platform longitudinal axis, check whether the change rate in two consecutive time periods is in the same direction. If the change rate signs are opposite in the two time periods, it is marked as the disturbance cancellation trend is established. If the change rate signs are consistent, it is determined that the disturbance trend continues, and the disturbance cancellation trend determination result is obtained. S313: Based on the disturbance cancellation trend determination result, under the conditions that the disturbance direction is reversed and the disturbance cancellation trend is established, determine whether the current disturbance has entered an alternating change state, and simultaneously generate a disturbance direction reverse excitation signal.

8. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The specific steps for obtaining the labeled data of the rate response command of the magnetic coupling structure are as follows: S411: Based on the reverse excitation signal of the disturbance direction, obtain the current speed calibration value of the magnetic coupling component output terminal, perform corresponding mapping in combination with the motor torque and speed control table, match it with the default compensation speed and generate a corresponding speed command, and output the default compensation speed command. S412: According to the default compensation speed command, drive the micro motor assembly of the pitch and roll linkage component to output the target speed command, monitor the response of the motor assembly in real time, record the delay time in the response process, the delay time is the time difference between the motor assembly receiving the speed command and the actual execution, reflecting any delay or lag phenomenon that occurs in the control process, and obtain response delay time data. S413: Based on the response delay time data, the lateral deviation position relationship between the sonar platform and the cabin contact point is used to perform state labeling according to the relationship between the lateral deviation and the disturbance direction, and magnetic coupling structure rate response command labeling data is generated.

9. The adaptive stabilization control method for an underwater sonar platform according to claim 1, characterized in that, The specific steps for obtaining the adaptive stabilization control record of the underwater sonar platform are as follows: S511: Obtain the labeled data of the rate response command of the magnetic coupling structure, determine the reverse compensation rate by real-time monitoring the lateral deviation and disturbance direction reverse excitation signal data of the contact point between the platform and the cabin, drive the control response, and generate the default compensation speed command. S512: Execute the default compensation speed command, collect the angle change data between the platform's roll axis and the inertial coordinate system in real time through the gyroscope component of the sonar platform, monitor the trend of angle change, determine that the angular velocity will not continue to increase after the disturbance, confirm whether the roll control has completed the closed-loop adjustment, and obtain the roll direction control closed-loop adjustment data. S513: The obtained roll direction control closed-loop adjustment data is compared in real time, and the platform's connection displacement is analyzed to determine whether the platform's attitude is stable. All errors are within the preset range, and an adaptive stability control record for the underwater sonar platform is generated.

Images