Direction warning system for mobile ADCP measurement vessels based on sensor data fusion

Abstract

This invention discloses a navigational ADCP measurement vessel orientation warning system based on sensor data fusion, belonging to the field of hydrological measurement technology. It includes: a sensor data acquisition module, which collects target vessel sensor data in real time based on electronic compass and gyroscope sensors; a data processing module, which uses a Kalman filter fusion algorithm to establish an attitude estimation state-space model, recursively predicts the target vessel's heading angle through state equations, and combines observation equations with real-time sensor data information to obtain the optimal estimated value of the heading angle, acquiring the real-time heading angle after data fusion; a lighting control module, which sets a heading angle threshold range to determine the real-time heading state of the vessel; if the real-time heading angle is continuously and stably within a preset angle range for more than 1 second, a corresponding color control signal is dynamically generated; and a fault self-checking module, which defines and executes a periodic self-checking and online calibration process for the electronic compass and gyroscope sensors. This invention improves the maintainability of the system.

Description

Technical Field

[0001] This invention relates to the field of hydrological measurement technology, specifically to a directional warning system for a mobile ADCP measurement vessel based on sensor data fusion. Background Technology

[0002] Existing navigational ADCP measurement vessel direction warning systems typically rely on a single electronic compass or a simple combination of sensors for heading determination. Their drawbacks include: single sensors are susceptible to interference from environmental magnetic fields or their own accumulated errors, leading to jumps or drifts in heading data; the stability and reliability of the output real-time heading angle are insufficient, making it difficult to meet the requirements of high-precision continuous measurement; traditional systems generally lack effective online fault diagnosis and adaptive fault-tolerance mechanisms. When sensors malfunction or fail, the system often fails directly or outputs incorrect warnings, failing to guarantee the continuity of measurement operations and increasing navigation safety risks. Summary of the Invention

[0003] To address the aforementioned technical issues, a navigational ADCP measurement vessel direction warning system based on sensor data fusion is provided. This technical solution resolves the problems of not being able to guarantee the continuity of measurement operations and navigation safety risks.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The navigational ADCP measurement vessel's direction warning system based on sensor data fusion includes:

[0006] Sensor data acquisition module, data processing module, lighting control module, fault self-diagnosis module;

[0007] Among them, the data processing module is electrically connected to the sensor data acquisition module, the lighting control module is electrically connected to the data processing module, and the fault self-test module is electrically connected to the sensor data acquisition module, the data processing module, and the lighting control module.

[0008] The sensor data acquisition module, based on electronic compass and gyroscope sensors, acquires sensor data from the target ship in real time.

[0009] The data processing module, based on the target ship's sensor data, uses a Kalman filter fusion algorithm to establish an attitude estimation state space model, recursively predicts the target ship's heading angle through the state equation, and combines the observation equation with real-time sensor data information to obtain the optimal estimate of the target ship's heading angle, and acquires the real-time heading angle after electronic compass-gyroscope data fusion.

[0010] The lighting control module determines the real-time heading angle based on the real-time heading angle after the fusion of electronic compass and gyroscope data. If the real-time heading angle is continuously and stably within a certain preset angle range for more than 1 second, a corresponding color control signal is dynamically generated and dynamically displayed on the ship's visual warning panel.

[0011] The fault self-test module defines and executes a periodic self-test and online calibration process for the electronic compass and gyroscope sensors, monitors the data quality and status of the electronic compass and gyroscope sensors in real time, and automatically triggers an adaptive emergency mechanism if an abnormality or fault is detected in the sensor.

[0012] Preferably, the sensor data acquisition module specifically includes:

[0013] Connect to the RMC100 electronic compass via RS422 interface, set the sampling frequency to 10-50Hz, read the target ship's magnetic north heading angle data in real time, and analyze the target ship's magnetic north heading angle data according to the IEC-61162-1 standard, normalizing the target ship's magnetic north heading angle data to a value within the range of 0°~360°.

[0014] 3. The navigational ADCP measurement vessel direction warning system based on sensor data fusion according to claim 2, characterized in that the sensor data acquisition module further includes:

[0015] For the acquired magnetic north heading angle data of the target ship, the ellipsoid fitting algorithm is used to fit the data points in the three-dimensional space of the acquired magnetic north heading angle data of the target ship using the least squares method to obtain the best ellipsoid model for compensation, eliminating the hard iron and soft iron error of the RMC100 electronic compass, and converting the compensated magnetic north heading angle data of the target ship into an absolute heading angle.

[0016] The hard iron error is a fixed error that manifests as a shift in different directions; the soft iron error varies with direction and is compensated by fitting a set of known target ship magnetic north heading angle data points.

[0017] Preferably, the sensor data acquisition module also includes:

[0018] Connecting to the BMI088 gyroscope via the SPI interface, the sampling frequency is set to 200-1000Hz to read the three-axis angular velocity data of the BMI088 gyroscope in real time. Time integration is performed on the read three-axis angular velocity data to calculate the change in the BMI088 gyroscope's three-axis angular velocity data relative to the heading angle, which is used as the heading increment of the BMI088 gyroscope's three-axis angular velocity data. High-frequency noise is suppressed by low-pass filtering, and zero bias error is removed to obtain the compensated heading increment of the BMI088 gyroscope's three-axis angular velocity data.

[0019] Preferably, the data processing module specifically includes:

[0020] Define the current magnetic north heading angle of the target ship and the state information of the three-axis angular velocity of the BMI088 gyroscope as it changes over time;

[0021] Using the three-axis angular velocity of the BMI088 gyroscope as input, a heading angle state equation is established, and the heading prediction of the target ship at the next moment is used as output.

[0022] The specific expression for the heading angle state equation is as follows:

[0023]

[0024] Where x is the target ship's heading angle state vector, k is the time step, and F is the time interval. At the time Dynamic changes, For the target ship at the time At the time For heading prediction, B is the three-axis angular velocity input matrix of the BMI088 gyroscope. For the BMI088 gyroscope at time The triaxial angular velocity;

[0025] Using the target ship's absolute heading angle as input, an observation equation is established, and the predicted value of the target ship's absolute heading angle is used as output;

[0026] The specific expression for the observation equation is as follows:

[0027]

[0028] in, For the target ship at the time Absolute heading angle observation prediction value, The observation matrix maps the target ship's absolute heading angle state vector to the observation space. To observe noise.

[0029] Preferably, the data processing module also includes:

[0030] By integrating the state equation and observation equation, an attitude estimation state space model is established. The Kalman filter algorithm is used to process the model, which is divided into two stages: prediction and update. In the prediction stage, based on the state equation and combined with the state vector and output of the BMI088 gyroscope at the previous moment, the target ship's heading angle state vector at the current moment is predicted. In the update stage, the target ship's absolute heading angle data is introduced and compared with the actual observed target ship absolute heading angle in combination with the observation matrix to remove observation noise. The predicted value of the target ship's heading angle state vector at the current moment is weighted and fused with the observed value of the target ship's absolute heading angle to obtain the optimal estimate of the target ship's heading angle. The real-time heading angle after the electronic compass-gyroscope data fusion is obtained.

[0031] Preferably, the lighting control module specifically includes:

[0032] Based on the real-time heading angle after data fusion from the electronic compass and gyroscope, a specific range of heading angle is set. If the real-time heading angle is in the range of 0° to 45°, it corresponds to the state of facing away from the operator and triggers the green light. If the real-time heading angle is in the range of 45° to 135°, it corresponds to the state of turning and triggers the yellow light. If the real-time heading angle is in the range of 135° to 180°, it corresponds to the state of returning to measurement and triggers the red light.

[0033] A state hold counter is set up for anti-shake logic judgment. In each processing cycle, the interval to which the current heading angle belongs is detected. If it is the same as the interval to which the previous cycle belongs, the state hold counter is incremented. If they are different, the counter is reset to zero.

[0034] Preferably, the lighting control module also includes:

[0035] For the state-holding counter incrementing, if the real-time heading angle remains continuously and stably within a certain range for more than 1 second, a corresponding color control signal is dynamically generated.

[0036] Based on the interval judgment result, a corresponding PWM control signal is generated to drive the three-color LED indicator. If the duty cycle of the green channel PWM is 100%, it indicates that the green light is flashing. If the duty cycles of the red and green channels PWM are both 50%, it indicates that the yellow light is flashing. If the duty cycle of the red channel PWM is 100%, it indicates that the red light is flashing.

[0037] By combining ambient light sensors, the ambient light intensity data of the target ship is acquired in real time, and a mapping relationship between ambient light intensity and PWM duty cycle is established. The brightness of LED lights is dynamically adjusted, increasing the brightness in dimly lit environments and decreasing the brightness in brightly lit environments.

[0038] Preferably, the fault self-diagnosis module specifically includes:

[0039] Set a timer to trigger a self-test process every 30 seconds for the RMC100 electronic compass. The self-test reads the current geomagnetic field strength data and determines whether the geomagnetic field strength is within a reasonable range. If the data exceeds the reasonable range, the abnormality is recorded and an early warning mechanism is triggered, which controls the indicator light to turn on red.

[0040] Perform a self-test on the BMI088 gyroscope, read the output value of the BMI088 gyroscope, calculate the zero bias, determine whether the zero bias exceeds the set threshold, and if it exceeds the threshold, record the abnormality and trigger the early warning mechanism.

[0041] If any abnormality is detected during the self-test of the RMC100 electronic compass and BMI088 gyroscope, the abnormality timestamp and abnormality type will be recorded immediately, and the data quality and status of the electronic compass and gyroscope sensors will be monitored in real time.

[0042] Preferably, the fault self-diagnosis module also includes:

[0043] When the equipment is started, the user is required to point the equipment in a known direction, record the sensor reading at this time, set it as the reference value, set the state transition equation and observation equation, obtain the relationship between the sensor reading and the actual heading, and each time a new target ship absolute heading angle observation value is obtained, calculate the residual between the target ship absolute heading angle observation value and the actual value, perform filtering estimation based on the residual, update the sensor parameters, and reduce its deviation.

[0044] If any sensor malfunction is detected, the system immediately enters emergency mode and automatically switches to single-sensor operation mode. If a malfunction is detected in the lighting module, the system automatically activates the backup audible alarm device, which emits a continuous alarm sound via a buzzer. All fault events, including sensor and lighting module malfunctions, are uploaded to the host computer in real time via a 2.4GHz radio, enabling remote recording and monitoring of fault events.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] This invention proposes a heading warning method for a mobile ADCP survey vessel based on sensor data fusion. By deeply fusing electronic compass and gyroscope data through a Kalman filter algorithm, it effectively overcomes the insufficient accuracy and instantaneous interference of a single sensor, obtaining continuous, stable, and high-precision real-time heading angle estimation, significantly improving the reliability of heading judgment. Based on the fused heading angle, a dynamic light warning mechanism realizes intuitive, stable, and visual early warning of the vessel's heading status, enhancing navigation safety. The integrated intelligent fault self-checking and adaptive emergency module can automatically switch the working mode and activate backup warning when the sensor is abnormal, ensuring the continuous and robust operation of the system under complex conditions. Overall, it constitutes a high-precision, highly reliable mobile ADCP survey vessel heading warning system with fault self-healing capabilities. Attached Figure Description

[0047] Figure 1 This is a framework diagram of a mobile ADCP measurement vessel direction warning system based on sensor data fusion. Detailed Implementation

[0048] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0049] Reference Figure 1 As shown, the navigational ADCP measurement vessel direction warning system based on sensor data fusion includes:

[0050] Sensor data acquisition module, data processing module, lighting control module, fault self-diagnosis module;

[0051] Among them, the data processing module is electrically connected to the sensor data acquisition module, the lighting control module is electrically connected to the data processing module, and the fault self-test module is electrically connected to the sensor data acquisition module, the data processing module, and the lighting control module.

[0052] The sensor data acquisition module, based on electronic compass and gyroscope sensors, acquires sensor data from the target ship in real time.

[0053] The data processing module, based on the target ship's sensor data, uses a Kalman filter fusion algorithm to establish an attitude estimation state space model, recursively predicts the target ship's heading angle through the state equation, and combines the observation equation with real-time sensor data information to obtain the optimal estimate of the target ship's heading angle, and acquires the real-time heading angle after electronic compass-gyroscope data fusion.

[0054] The lighting control module determines the real-time heading angle based on the real-time heading angle after the fusion of electronic compass and gyroscope data. If the real-time heading angle is continuously and stably within a certain preset angle range for more than 1 second, a corresponding color control signal is dynamically generated and dynamically displayed on the ship's visual warning panel.

[0055] The fault self-test module defines and executes a periodic self-test and online calibration process for the electronic compass and gyroscope sensors, monitors the data quality and status of the electronic compass and gyroscope sensors in real time, and automatically triggers an adaptive emergency mechanism if an abnormality or fault is detected in the sensor.

[0056] The sensor data acquisition module specifically includes:

[0057] Connect to the RMC100 electronic compass via RS422 interface, set the sampling frequency to 10-50Hz, read the target ship's magnetic north heading angle data in real time, and analyze the target ship's magnetic north heading angle data according to the IEC-61162-1 standard, normalizing the target ship's magnetic north heading angle data to a value within the range of 0°~360°.

[0058] The sensor data acquisition module also includes:

[0059] For the acquired magnetic north heading angle data of the target ship, the ellipsoid fitting algorithm is used to fit the data points in the three-dimensional space of the acquired magnetic north heading angle data of the target ship using the least squares method to obtain the best ellipsoid model for compensation, eliminating the hard iron and soft iron error of the RMC100 electronic compass, and converting the compensated magnetic north heading angle data of the target ship into an absolute heading angle.

[0060] The hard iron error is a fixed error that manifests as a shift in different directions; the soft iron error varies with direction and is compensated by fitting a set of known target ship magnetic north heading angle data points.

[0061] The sensor data acquisition module also includes:

[0062] Connecting to the BMI088 gyroscope via the SPI interface, the sampling frequency is set to 200-1000Hz to read the three-axis angular velocity data of the BMI088 gyroscope in real time. Time integration is performed on the read three-axis angular velocity data to calculate the change in the BMI088 gyroscope's three-axis angular velocity data relative to the heading angle, which is used as the heading increment of the BMI088 gyroscope's three-axis angular velocity data. High-frequency noise is suppressed by low-pass filtering, and zero bias error is removed to obtain the compensated heading increment of the BMI088 gyroscope's three-axis angular velocity data.

[0063] When using it, combine it with the modules mentioned above:

[0064] In existing technologies, when acquiring heading information, sensor data acquisition modules typically rely on a single sensor, the electronic compass, to directly output the heading angle. However, this is susceptible to errors caused by hard and soft iron, leading to decreased accuracy, and lacks systematic ellipsoidal fitting compensation. If gyroscope data is used independently for heading calculation, it will diverge over time due to accumulated integral errors and zero-bias drift, lacking an effective real-time filtering and error correction mechanism, ultimately resulting in poor heading data stability and insufficient long-term reliability. This step employs a multi-source data fusion method using the RMC100 electronic compass and the BMI088 gyroscope, combined with an ellipsoidal fitting algorithm to compensate for the hard and soft iron errors of the compass in real time. The gyroscope's angular velocity data is processed through integration, low-pass filtering, and zero-bias correction, improving the measurement accuracy and anti-interference capability of the heading angle. Furthermore, data complementarity suppresses the inherent defects of a single sensor, enhancing the system's reliability and long-term stability in dynamic environments.

[0065] The data processing module specifically includes:

[0066] Define the current magnetic north heading angle of the target ship and the state information of the three-axis angular velocity of the BMI088 gyroscope as it changes over time;

[0067] Using the three-axis angular velocity of the BMI088 gyroscope as input, a heading angle state equation is established, and the heading prediction of the target ship at the next moment is used as output.

[0068] The specific expression for the heading angle state equation is as follows:

[0069]

[0070] Where x is the target ship's heading angle state vector, k is the time step, and F is the time interval. At the time Dynamic changes, For the target ship at the time At the time For heading prediction, B is the three-axis angular velocity input matrix of the BMI088 gyroscope. For the BMI088 gyroscope at time The triaxial angular velocity;

[0071] Using the target ship's absolute heading angle as input, an observation equation is established, and the predicted value of the target ship's absolute heading angle is used as output;

[0072] The specific expression for the observation equation is as follows:

[0073]

[0074] in, For the target ship at the time Absolute heading angle observation prediction value, The observation matrix maps the target ship's absolute heading angle state vector to the observation space. To observe noise.

[0075] The data processing module also includes:

[0076] By integrating the state equation and observation equation, an attitude estimation state space model is established. The Kalman filter algorithm is used to process the model, which is divided into two stages: prediction and update. In the prediction stage, based on the state equation and combined with the state vector and output of the BMI088 gyroscope at the previous moment, the target ship's heading angle state vector at the current moment is predicted. In the update stage, the target ship's absolute heading angle data is introduced and compared with the actual observed target ship absolute heading angle in combination with the observation matrix to remove observation noise. The predicted value of the target ship's heading angle state vector at the current moment is weighted and fused with the observed value of the target ship's absolute heading angle to obtain the optimal estimate of the target ship's heading angle. The real-time heading angle after the electronic compass-gyroscope data fusion is obtained.

[0077] When using it, combine it with the modules mentioned above:

[0078] Traditional target ship heading angle prediction methods based on BMI088 gyroscopes typically suffer from the following problems: when relying solely on the integration of three-axis angular velocity to calculate the heading, the inherent zero-bias noise and cumulative integration error of the gyroscope easily lead to significant drift in the heading prediction over time; in scenarios with drastic dynamic changes or long-term operation, the lack of integration with external absolute heading observations makes it difficult to correct system errors, resulting in decreased heading estimation accuracy and stability; the lack of effective modeling and real-time compensation mechanisms for observation noise makes it prone to observation bias under external interference, affecting the reliability of heading estimation. This step utilizes the high-precision positioning of the gyroscope... The frequency dynamic response characteristics provide short-term heading change predictions. By periodically correcting integral drift through absolute heading angle observations, the long-term stability of heading estimation is improved. By weighted fusion of predicted and observed values, the optimal balance between gyroscope data and external heading data is achieved, which not only suppresses noise interference but also enhances the robustness of the system in dynamic environments. The introduction of observation noise modeling and real-time filtering update mechanisms effectively reduces the impact of external disturbances such as magnetic field anomalies on heading observations, outputting more accurate and reliable real-time fused heading angles, which are suitable for navigation and control scenarios with high heading accuracy requirements.

[0079] The lighting control module specifically includes:

[0080] Based on the real-time heading angle after data fusion from the electronic compass and gyroscope, a specific range of heading angle is set. If the real-time heading angle is in the range of 0° to 45°, it corresponds to the state of facing away from the operator and triggers the green light. If the real-time heading angle is in the range of 45° to 135°, it corresponds to the state of turning and triggers the yellow light. If the real-time heading angle is in the range of 135° to 180°, it corresponds to the state of returning to measurement and triggers the red light.

[0081] A state hold counter is set up for anti-shake logic judgment. In each processing cycle, the interval to which the current heading angle belongs is detected. If it is the same as the interval to which the previous cycle belongs, the state hold counter is incremented. If they are different, the counter is reset to zero.

[0082] The lighting control module also includes:

[0083] For the state-holding counter incrementing, if the real-time heading angle remains continuously and stably within a certain range for more than 1 second, a corresponding color control signal is dynamically generated.

[0084] Based on the interval judgment result, a corresponding PWM control signal is generated to drive the three-color LED indicator. If the duty cycle of the green channel PWM is 100%, it indicates that the green light is flashing. If the duty cycles of the red and green channels PWM are both 50%, it indicates that the yellow light is flashing. If the duty cycle of the red channel PWM is 100%, it indicates that the red light is flashing.

[0085] By combining ambient light sensors, the ambient light intensity data of the target ship is acquired in real time, and a mapping relationship between ambient light intensity and PWM duty cycle is established. The brightness of LED lights is dynamically adjusted, increasing the brightness in dimly lit environments and decreasing the brightness in brightly lit environments.

[0086] When using it, combine it with the modules mentioned above:

[0087] The shortcomings of existing technologies lie in the fact that traditional lighting control modules typically rely on data from a single sensor for heading determination, lacking multi-sensor data fusion. This results in heading angle measurements being susceptible to interference and exhibiting poor stability. During interval switching, false triggering due to instantaneous fluctuations is common, and the lack of an effective anti-shake mechanism can cause frequent light flickering or unnecessary state jumps. Furthermore, lighting brightness adjustment often uses fixed PWM output, failing to dynamically adapt to ambient light intensity, leading to insufficient visibility in strong or weak light environments and affecting the reliability of the indication. The beneficial effects of this approach are: by fusing data from the electronic compass and gyroscope, the accuracy and stability of the real-time heading angle are improved; by setting specific heading angle intervals and a state-holding counter, anti-shake logic is implemented, effectively avoiding false triggering due to instantaneous fluctuations and ensuring continuous and reliable lighting state switching; by dynamically adjusting the PWM duty cycle using an ambient light sensor, the LED lighting brightness can adapt to changes in ambient light intensity, enhancing visibility and energy efficiency under different lighting conditions; the overall design achieves intelligent and stable indication of lighting status and adaptive brightness control, improving system robustness and user experience.

[0088] The fault self-diagnosis module specifically includes:

[0089] Set a timer to trigger a self-test process every 30 seconds for the RMC100 electronic compass. The self-test reads the current geomagnetic field strength data and determines whether the geomagnetic field strength is within a reasonable range. If the data exceeds the reasonable range, the abnormality is recorded and an early warning mechanism is triggered, which controls the indicator light to turn on red.

[0090] The reasonable range is, for example, 20-60 μT;

[0091] Perform a self-test on the BMI088 gyroscope, read the output value of the BMI088 gyroscope, calculate the zero bias, determine whether the zero bias exceeds the set threshold, and if it exceeds the threshold, record the abnormality and trigger the early warning mechanism.

[0092] The threshold is defined as greater than 0.5° / s.

[0093] If any abnormality is detected during the self-test of the RMC100 electronic compass and BMI088 gyroscope, the abnormality timestamp and abnormality type will be recorded immediately, and the data quality and status of the electronic compass and gyroscope sensors will be monitored in real time.

[0094] The fault self-diagnosis module also includes:

[0095] When the equipment is started, the user is required to point the equipment in a known direction, record the sensor reading at this time, set it as the reference value, set the state transition equation and observation equation, obtain the relationship between the sensor reading and the actual heading, and each time a new target ship absolute heading angle observation value is obtained, calculate the residual between the target ship absolute heading angle observation value and the actual value, perform filtering estimation based on the residual, update the sensor parameters, and reduce its deviation.

[0096] If any sensor malfunction is detected, the system immediately enters emergency mode and automatically switches to single-sensor operation mode. If a malfunction is detected in the lighting module, the system automatically activates the backup audible alarm device, which emits a continuous alarm sound via a buzzer. All fault events, including sensor and lighting module malfunctions, are uploaded to the host computer in real time via a 2.4GHz radio, enabling remote recording and monitoring of fault events.

[0097] When using it, combine it with the modules mentioned above:

[0098] Traditional fault-failure switching mechanisms are mostly simple master-slave switching, failing to fully utilize data from healthy sensors to estimate and compensate for faulty sensor parameters online, thus reducing the system's robustness and accuracy after a fault. This step combines timed self-checks with real-time data quality monitoring to achieve continuous monitoring of sensor status and reduce detection blind spots. A filtering estimation algorithm based on benchmark values ​​and residual analysis is introduced, which can update sensor parameters online, adaptively reduce deviations, and improve the long-term stability and reliability of heading angle measurement. The fault emergency mode has intelligent switching and multiple redundancy warning capabilities, ensuring that the core functions of the system are not interrupted under a single fault. At the same time, remote real-time recording and monitoring of fault events are achieved through wireless communication, improving the system's maintainability and overall security.

[0099] The specific implementation method is as follows:

[0100] Taking a work vessel performing a mobile ADCP flow measurement in a section of the Yangtze River as an example, the measurement vessel needs to repeatedly navigate along a preset cross section. The direction of the bow frequently switches between three states: "downstream with the back of the operator" (safety), "turning to cross the cross section" (caution), and "upstream with the operator" (warning). The operator is located in the cockpit at the stern and needs to clearly and stably know the orientation of the bow in order to complete the measurement task safely and efficiently.

[0101] Before the measurement task begins, the operator points the bow of the ship to due north (known direction) and activates the warning system. After the system is powered on, the fault self-test module first guides the rapid calibration: record that the electronic compass reading is 0° (north) and the gyroscope angular velocity is zero. Set this as the reference and the sensor data acquisition module starts working.

[0102] The electronic compass (RMC100) outputs the heading angle at a frequency of 20Hz through the RS422 interface. The system immediately uses the stored ellipsoid fitting coefficients (obtained from the calibration on the survey vessel before leaving the factory) to compensate the original data in real time, eliminating magnetic interference caused by ferromagnetic materials such as the ship's engine and ADCP bracket, and outputs an absolute heading angle accurate to 0.1° (358.5° after compensation).

[0103] The gyroscope (BMI088) outputs high-precision three-axis angular velocity at a frequency of 500Hz through the SPI interface. After low-pass filtering to remove high-frequency vibration noise, zero drift is deducted in real time (the self-test calculation shows that the current zero bias is 0.02° / s). Then, the angular velocity is integrated to calculate the short-time, high-dynamic-response heading change.

[0104] The Kalman filter of the data processing module simultaneously receives the two data streams mentioned above. The filter's internal model is based on the optimal heading fused from the previous moment (initialized to 0°), combined with the instantaneous heading change integrated by the gyroscope (the ship turns slightly to the right, angular velocity +0.5° / s), predicting that the heading at the current moment should be 0.25°.

[0105] The filter incorporates the absolute heading (358.5°) observed by the electronic compass. Since the system is aware that the electronic compass may experience instantaneous magnetic jump noise in dynamic environments (passing through bridge steel structures), and the short-term predictions of the gyroscope are very reliable, the Kalman gain algorithm assigns a higher weight (70%) to the "predicted value" and a lower weight (30%) to the "observed value". After weighted fusion, the system outputs a final fused heading angle of 0.9°. This result not only smooths out the accidental jumps of the electronic compass, but also corrects for the possible long-term small drifts of the gyroscope with the absolute heading, achieving a stable, accurate (better than 1°) heading angle output with an update frequency of up to 100Hz.

[0106] The lighting control module receives a fused heading angle of 0.9°. According to the preset rule, 0°~45° is the safe range for "facing away from the operator" and corresponds to the green light. The module starts the anti-shake logic: continuously monitors 20 data cycles (about 1 second). The heading angle always fluctuates slightly between 0.5° and 1.2° and is stable within the range of 0°~45°. The status counter is incremented to 20. After the counter is full, the module determines that the status is valid and generates a green PWM control signal (green channel duty cycle 100%).

[0107] The LED light on the bow warning panel should light up green. The ambient light sensor detects that it is a cloudy evening with weak ambient light. The module automatically increases the PWM base frequency, increasing the LED brightness to 80% of its maximum value, ensuring that the indicator remains visible in dim environments. When the ship completes the measurement and needs to turn around, the light will automatically switch to flashing yellow after the heading angle enters the 45°~135° warning range and stabilizes for more than 1 second. When it enters the 135°~180° warning range, it will switch to solid red, visually informing the operator that the bow is facing them and caution is required.

[0108] Throughout the voyage, the fault self-check module automatically ran every 30 seconds. During one self-check, it was found that the electronic compass suddenly output a geomagnetic field strength of 65μT, exceeding the reasonable range of 20-60μT (possibly due to a temporary approach to a large steel cargo ship). The system immediately determined that the compass data was abnormal, recorded the "compass magnetic field anomaly" event internally, and sent a timestamped alarm message to the shore-based monitoring center via a 2.4GHz radio. The system immediately triggered an adaptive emergency mechanism: the data processing module automatically ignored this abnormal observation value and relied entirely on the highly reliable gyroscope data for course estimation; the lighting control module added a "data degradation" flag to the status judgment but maintained the current indicator light status. After about 15 seconds, the ship moved away from the interference source, the compass data returned to normal, and the system automatically and seamlessly switched back to the dual-sensor fusion mode. Throughout the process, the system did not mislead the operator, and the lighting indication remained continuous and stable, ensuring the continuity and safety of the measurement operation.

[0109] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A heading warning system for a ship under way using an ADCP sensor data fusion based measurement, characterized in that, include: Sensor data acquisition module, data processing module, lighting control module, fault self-diagnosis module; Among them, the data processing module is electrically connected to the sensor data acquisition module, the lighting control module is electrically connected to the data processing module, and the fault self-test module is electrically connected to the sensor data acquisition module, the data processing module, and the lighting control module. The sensor data acquisition module, based on electronic compass and gyroscope sensors, acquires sensor data from the target ship in real time. The data processing module, based on the target ship's sensor data, uses a Kalman filter fusion algorithm to establish an attitude estimation state space model, recursively predicts the target ship's heading angle through the state equation, and combines the observation equation with real-time sensor data information to obtain the optimal estimate of the target ship's heading angle, and acquires the real-time heading angle after electronic compass-gyroscope data fusion. The lighting control module determines the real-time heading angle based on the real-time heading angle after the fusion of electronic compass and gyroscope data. If the real-time heading angle is continuously and stably within a certain preset angle range for more than 1 second, a corresponding color control signal is dynamically generated and dynamically displayed on the ship's visual warning panel. The fault self-test module defines and executes a periodic self-test and online calibration process for the electronic compass and gyroscope sensors, monitors the data quality and status of the electronic compass and gyroscope sensors in real time, and automatically triggers an adaptive emergency mechanism if an abnormality or fault is detected in the sensor.

2. The sensor data fusion based underway ADCP survey vessel heading alert system of claim 1, wherein, The sensor data acquisition module specifically includes: Connect to the RMC100 electronic compass via RS422 interface, set the sampling frequency to 10-50Hz, read the target ship's magnetic north heading angle data in real time, and analyze the target ship's magnetic north heading angle data according to the IEC-61162-1 standard, normalizing the target ship's magnetic north heading angle data to a value within the range of 0°~360°.

3. The sensor data fusion based underway ADCP survey vessel heading alert system of claim 2, wherein, The sensor data acquisition module also includes: For the acquired magnetic north heading angle data of the target ship, the ellipsoid fitting algorithm is used to fit the data points in the three-dimensional space of the acquired magnetic north heading angle data of the target ship using the least squares method to obtain the best ellipsoid model for compensation, eliminating the hard iron and soft iron error of the RMC100 electronic compass, and converting the compensated magnetic north heading angle data of the target ship into an absolute heading angle. The hard iron error is a fixed error that manifests as a shift in different directions; the soft iron error varies with direction and is compensated by fitting a set of known target ship magnetic north heading angle data points.

4. The sensor data fusion based underway ADCP survey vessel heading alert system of claim 3, wherein, The sensor data acquisition module also includes: Connecting to the BMI088 gyroscope via the SPI interface, the sampling frequency is set to 200-1000Hz to read the three-axis angular velocity data of the BMI088 gyroscope in real time. Time integration is performed on the read three-axis angular velocity data to calculate the change in the BMI088 gyroscope's three-axis angular velocity data relative to the heading angle, which is used as the heading increment of the BMI088 gyroscope's three-axis angular velocity data. High-frequency noise is suppressed by low-pass filtering, and zero bias error is removed to obtain the compensated heading increment of the BMI088 gyroscope's three-axis angular velocity data.

5. The sensor data fusion based underway ADCP survey vessel heading alert system of claim 4, wherein, The data processing module specifically includes: Define the current magnetic north heading angle of the target ship and the state information of the three-axis angular velocity of the BMI088 gyroscope as it changes over time; Using the three-axis angular velocity of the BMI088 gyroscope as input, a heading angle state equation is established, and the heading prediction of the target ship at the next moment is used as output. The specific expression for the heading angle state equation is as follows: ; where x is the target ship course angle state vector, k is the time step, F is the dynamic change of the target ship at time to time , is the course prediction of the target ship at time to time , B is the BMI088 gyroscope three-axis angular velocity input matrix, is the three-axis angular velocity of the BMI088 gyroscope at time ; Using the target ship's absolute heading angle as input, an observation equation is established, and the predicted value of the target ship's absolute heading angle is used as output; The specific expression for the observation equation is as follows: ; wherein, is the target ship's absolute heading angle at time is the absolute heading angle observation prediction, is the observation matrix mapping the target ship's absolute heading angle state vector to the observation space, is the observation noise.

6. The sensor data fusion based underway ADCP survey vessel heading alert system of claim 5, wherein, The data processing module also includes: By integrating the state equation and observation equation, an attitude estimation state space model is established. The Kalman filter algorithm is used to process the model, which is divided into two stages: prediction and update. In the prediction stage, based on the state equation and combined with the state vector and output of the BMI088 gyroscope at the previous moment, the target ship's heading angle state vector at the current moment is predicted. In the update stage, the target ship's absolute heading angle data is introduced and compared with the actual observed target ship absolute heading angle in combination with the observation matrix to remove observation noise. The predicted value of the target ship's heading angle state vector at the current moment is weighted and fused with the observed value of the target ship's absolute heading angle to obtain the optimal estimate of the target ship's heading angle. The real-time heading angle after the electronic compass-gyroscope data fusion is obtained.

7. The navigational ADCP measurement vessel direction warning system based on sensor data fusion according to claim 6, characterized in that, The lighting control module specifically includes: Based on the real-time heading angle after data fusion from the electronic compass and gyroscope, a specific range of heading angle is set. If the real-time heading angle is in the range of 0° to 45°, it corresponds to the state of facing away from the operator and triggers the green light. If the real-time heading angle is in the range of 45° to 135°, it corresponds to the state of turning and triggers the yellow light. If the real-time heading angle is in the range of 135° to 180°, it corresponds to the state of returning to measurement and triggers the red light. A state hold counter is set up for anti-shake logic judgment. In each processing cycle, the interval to which the current heading angle belongs is detected. If it is the same as the interval to which the previous cycle belongs, the state hold counter is incremented. If they are different, the counter is reset to zero.

8. The navigational ADCP measurement vessel direction warning system based on sensor data fusion according to claim 7, characterized in that, The lighting control module also includes: For the state-holding counter incrementing, if the real-time heading angle remains continuously and stably within a certain range for more than 1 second, a corresponding color control signal is dynamically generated. Based on the interval judgment result, a corresponding PWM control signal is generated to drive the three-color LED indicator. If the duty cycle of the green channel PWM is 100%, it indicates that the green light is flashing. If the duty cycles of the red and green channels PWM are both 50%, it indicates that the yellow light is flashing. If the duty cycle of the red channel PWM is 100%, it indicates that the red light is flashing. By combining ambient light sensors, the ambient light intensity data of the target ship is acquired in real time, and a mapping relationship between ambient light intensity and PWM duty cycle is established. The brightness of LED lights is dynamically adjusted, increasing the brightness in dimly lit environments and decreasing the brightness in brightly lit environments.

9. The navigational ADCP measurement vessel direction warning system based on sensor data fusion according to claim 8, characterized in that, The fault self-diagnosis module specifically includes: Set a timer to trigger a self-test process every 30 seconds for the RMC100 electronic compass. The self-test reads the current geomagnetic field strength data and determines whether the geomagnetic field strength is within a reasonable range. If the data exceeds the reasonable range, the abnormality is recorded and an early warning mechanism is triggered, which controls the indicator light to turn on red. Perform a self-test on the BMI088 gyroscope, read the output value of the BMI088 gyroscope, calculate the zero bias, determine whether the zero bias exceeds the set threshold, and if it exceeds the threshold, record the abnormality and trigger the early warning mechanism. If any abnormality is detected during the self-test of the RMC100 electronic compass and BMI088 gyroscope, the abnormality timestamp and abnormality type will be recorded immediately, and the data quality and status of the electronic compass and gyroscope sensors will be monitored in real time.

10. The navigational ADCP measurement vessel direction warning system based on sensor data fusion according to claim 9, characterized in that, The fault self-diagnosis module also includes: When the equipment is started, the user is required to point the equipment in a known direction, record the sensor reading at this time, set it as the reference value, set the state transition equation and observation equation, obtain the relationship between the sensor reading and the actual heading, and each time a new target ship absolute heading angle observation value is obtained, calculate the residual between the target ship absolute heading angle observation value and the actual value, perform filtering estimation based on the residual, update the sensor parameters, and reduce its deviation. If any sensor malfunction is detected, the system immediately enters emergency mode and automatically switches to single-sensor operation mode. If a malfunction is detected in the lighting module, the system automatically activates the backup audible alarm device, which emits a continuous alarm sound via a buzzer. All fault events, including sensor and lighting module malfunctions, are uploaded to the host computer in real time via a 2.4GHz radio, enabling remote recording and monitoring of fault events.

Images