Life jacket for sea rescue

Abstract

A life jacket for sea rescue belongs to the technical field of water rescue equipment and solves the problem of rescue delay caused by the fact that a traditional life jacket cannot automatically and accurately detect a drowning state and a user posture. According to the technical scheme, a water pressure sensor is arranged on a life jacket body, a buoyancy providing device comprises a front buoyancy cavity set, a rear buoyancy cavity set and a gas distribution valve, and an integrated control device processes sensor data through a microprocessor, starts a monitoring mode after detecting water entry and analyzes starting sequence inflation according to posture changes or active struggling; and automatic positioning is carried out and a distress signal is sent when the danger is continued. The device is mainly used for sea drowning rescue.

Description

Life jackets for maritime rescue Technical Field

[0001] This invention belongs to the technical field of water rescue equipment, specifically relating to a life jacket for maritime rescue. Background Technology

[0002] Currently, most life jackets used in maritime rescue have automatic inflation functions, but their triggering mechanism mainly relies on a single water pressure sensor or manual operation. This single detection method has limitations. When the life jacket is partially submerged or in unstable waters with undulating waves, it may not be able to determine in a timely and accurate manner whether the user is in a real danger of falling into the water. If the water pressure sensor is only placed in a single location, its detection signal may not reach the preset trigger threshold when the person is tilted or only partially submerged, resulting in a system response delay. This delay directly shortens the golden rescue time after falling into the water.

[0003] Furthermore, existing self-inflating life vests typically inflate all buoyancy chambers simultaneously upon triggering. While this instantaneous, all-around inflation method can quickly provide buoyancy, it can also generate a sudden and powerful impact on a person in an unstable position. This impact may cause the person to lose balance in the water, or even temporarily submerge their head, increasing the risk of choking and panic, which is detrimental to their ability to maintain stability and await rescue.

[0004] Another significant problem is the lack of continuous monitoring and intelligent assessment capabilities for users after falling into the water. After initial inflation, the system often fails to further assess whether the user is in a dangerous state such as ongoing instability, struggling, or unconsciousness. For example, after inflation, a user may be unable to maintain an effective breathing posture due to exhaustion, injury, or confusion, or may remain in a high-risk state of having their face submerged due to the impact of waves. Existing devices cannot identify these subsequent dangers and therefore cannot automatically activate more advanced distress signals, such as sending precise location information to a rescue center.

[0005] The main reason for these problems lies in the fact that the design of traditional life jackets focuses on providing basic buoyancy. Their control systems are relatively simple and fail to effectively integrate multi-source sensor information and perform complex logic analysis. Attempts to introduce more complex sensors and decision-making logic have faced practical difficulties. Firstly, how to rationally deploy multiple sensors to comprehensively perceive the environment and attitude while ensuring the life jacket is lightweight and comfortable. Secondly, the harsh working environment of life jackets places extremely high demands on the waterproof, corrosion-resistant, and impact-resistant performance of electronic components, increasing system complexity and cost. Finally, implementing real-time, reliable multi-sensor data fusion and intelligent algorithms on low-power microprocessors—for example, accurately distinguishing between normal body movements and dangerous struggling behavior, or differentiating between ship rolling and actual fall into the water—presents significant technical challenges. These factors collectively contribute to the shortcomings of traditional life jackets in automated and intelligent rescue operations. Summary of the Invention

[0006] The purpose of this invention is to provide an intelligent life jacket for maritime rescue, which integrates a water pressure sensor, an attitude detection device, and microprocessor logic to achieve automatic and precise rescue response.

[0007] To address the aforementioned problems and achieve the objectives and other advantages of this invention, a life jacket for maritime rescue is provided, comprising: a life jacket body; four water pressure sensors mounted on the hem of the life jacket body, respectively located at the front, rear, left, and right sides of the life jacket body; a buoyancy supply device including: front and rear chest buoyancy chambers, respectively fixed to the front chest area and back area of ​​the life jacket body; a high-pressure gas cylinder mounted on the waist side of the life jacket body; a gas distribution valve, the inlet of which is connected to the outlet of the high-pressure gas cylinder; the gas distribution valve having a first outlet and a second outlet, the first outlet and the second outlet respectively passing through... The air duct connects to the anterior chest buoyancy chamber and the posterior back buoyancy chamber; the positioning and communication device includes a GPS module and a wireless communication module; the attitude detection device includes a three-axis accelerometer and a three-axis gyroscope located in the shoulder area of ​​the life jacket body; the integrated control device includes a microprocessor; four water pressure sensors, the three-axis accelerometer, the three-axis gyroscope, the gas distribution valve, the GPS module, and the wireless communication module are respectively connected to the microprocessor; when at least three of the four water pressure sensors detect a water depth of 0.3-0.8m for more than 2 seconds, the microprocessor determines that the life jacket is in the water and activates... Dynamic monitoring mode; In monitoring mode, the microprocessor executes the following two judgment paths in parallel: Path 1: If the pitch or roll angle calculated by the triaxial gyroscope changes by more than 45 degrees within 2 consecutive seconds, and this state lasts for more than 8 seconds, the microprocessor determines it to be in a continuous unstable state and initiates sequential inflation; Path 2: The microprocessor performs a fast Fourier transform on the data from the triaxial accelerometer to extract its dominant frequency and calculates its approximate entropy to quantify the signal disorder; If the dominant frequency is in the range of 3-8Hz and the approximate entropy is continuously higher than a preset threshold within 5 seconds, the microprocessor... The device determines that the device is in an active struggling state and initiates sequential inflation. Sequential inflation is as follows: the gas distribution valve first opens the first outlet and continues for 2-3 seconds to inflate the anterior chest buoyancy chamber, then closes the first outlet and opens the second outlet for 3-4 seconds to inflate the posterior back buoyancy chamber. When at least three of the four water pressure sensors detect a water depth exceeding 1m, or after sequential inflation is completed, if the microprocessor continues to detect a change in pitch or roll angle exceeding 45 degrees for 10 seconds, the microprocessor activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

[0008] Preferably, in the aforementioned marine rescue life jacket, a first valve position sensor and a second valve position sensor are respectively installed at the first outlet and the second outlet of the gas distribution valve. The first valve position sensor and the second valve position sensor are connected to a microprocessor. The microprocessor is configured to, after issuing a command to open the first outlet or the second outlet, if it does not receive an opening signal from the corresponding valve position sensor within a first preset time period, determine that there is an inflation failure, and control the wireless communication module to send an alarm signal containing the inflation failure status.

[0009] Preferably, in the aforementioned marine rescue life jacket, if, after sequential inflation, the microprocessor continuously detects a change in pitch or roll angle exceeding 45 degrees for 10 seconds, the microprocessor initiates a 10-second behavior verification period. During this period, the microprocessor analyzes data from the triaxial accelerometer. If it identifies a preset pattern of waving or splashing motions, it immediately activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information. If no preset pattern of waving or splashing motions is identified during the verification period, the microprocessor records a verification failure and enters a 60-second standby monitoring period. After the standby monitoring period ends, if the microprocessor again continuously detects a change in pitch or roll angle exceeding 45 degrees for 10 seconds, it restarts the behavior verification period. If the number of verification failures accumulates to three, the microprocessor forcibly activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

[0010] Preferably, the life jacket for maritime rescue further includes: a waterproof manual alarm button, which is located in the shoulder area of ​​the life jacket body and connected to a microprocessor; when the microprocessor detects that the waterproof manual alarm button has been pressed continuously for more than 2 seconds, it immediately executes the sequential inflation process, activates the GPS module for positioning, and simultaneously controls the wireless communication module to continuously send a distress signal containing a manual alarm indicator and location information; an optical alarm device, which includes forward-facing LED modules respectively located on the front sides of the left and right shoulder straps of the life jacket body, and rearward-facing LED modules respectively located on the rear sides of the left and right shoulder straps of the life jacket body; optical alarm device Connected to a microprocessor; when the microprocessor controls the wireless communication module to send a distress signal, the optical alarm device is activated simultaneously; the microprocessor acquires the pitch angle data calculated by the three-axis gyroscope in real time; the microprocessor is configured to compare the pitch angle data with a preset positive angle threshold and a preset negative angle threshold; when the pitch angle data is greater than the positive angle threshold, the microprocessor increases the light intensity of the forward LED module while decreasing the light intensity of the rearward LED module; when the pitch angle data is less than the negative angle threshold, the microprocessor increases the light intensity of the rearward LED module while decreasing the light intensity of the forward LED module.

[0011] Preferably, in the aforementioned life jacket for maritime rescue, after the wireless communication module sends a distress signal containing location information each time, the microprocessor initiates a waiting window; if the wireless communication module does not receive a confirmation signal from the external rescue center within the waiting window, the microprocessor determines that the distress signal transmission has failed and controls the wireless communication module to switch the communication channel or retransmit the distress signal with enhanced transmission power.

[0012] Preferably, in the aforementioned marine rescue life jacket, the microprocessor is further configured to execute a sensor self-diagnostic process; in standby mode where the water immersion state is not determined, the microprocessor periodically performs the following operations: based on the angular velocity data measured by the three-axis gyroscope, a first attitude change trajectory is obtained through integration calculation; simultaneously, based on the acceleration data measured by the three-axis accelerometer, a second attitude change trajectory is obtained by integrating after removing the gravitational acceleration component; the microprocessor continuously compares the consistency between the first attitude change trajectory and the second attitude change trajectory within a second preset time period; if the difference between the first attitude change trajectory and the second attitude change trajectory exceeds a preset fault tolerance range, the microprocessor determines that there is a faulty sensor unit in the three-axis accelerometer or the three-axis gyroscope, and controls the wireless communication module to send a warning message containing a specific faulty unit identifier; thereafter, the microprocessor executes subsequent processing... Process: The microprocessor records the identifier of the specific faulty sensor unit in its internal memory and switches the system to the corresponding degraded operation mode accordingly. If the faulty sensor unit is either a three-axis accelerometer or a three-axis gyroscope, the microprocessor will block the data of that faulty sensor unit in subsequent monitoring modes and execute the judgment logic for the water immersion state and continuous instability state based only on the data of another normally functioning sensor and the data of the four water pressure sensors. If the faulty sensor unit includes both a three-axis accelerometer and a three-axis gyroscope, the microprocessor determines that the system has entered the basic safety mode. In the basic safety mode, when at least three of the four water pressure sensors detect a water depth exceeding 1m, the microprocessor directly activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information and the system's degraded status.

[0013] Preferably, in the aforementioned marine rescue life jacket, the microprocessor is further configured to execute an auxiliary water entry determination process. In standby mode where water entry is not determined, if no more than two of the four water pressure sensors detect a water depth of 0.3-0.8m, but the pitch and roll angles calculated by the three-axis gyroscope change by more than 60 degrees within 3 consecutive seconds, the microprocessor activates an auxiliary monitoring mode. In auxiliary monitoring mode, the microprocessor performs parallel monitoring of water pressure sensor data and attitude data. If, within 10 seconds after the activation of auxiliary monitoring mode, at least three of the four water pressure sensors detect a water depth of 0.3-0.8m, the microprocessor formally determines water entry and enters monitoring mode. If, within 10 seconds, the condition of at least three of the four water pressure sensors detecting a water depth of 0.3-0.8m is not met, but the pitch and roll angles calculated by the three-axis gyroscope continue to change by more than 60 degrees, the microprocessor forcibly determines water entry at the end of 10 seconds and enters monitoring mode.

[0014] Preferably, the life jacket for maritime rescue further includes: a human-machine interface device, which includes a miniature vibration motor and a miniature buzzer, and the human-machine interface device is connected to a microprocessor; the microprocessor is configured to control the human-machine interface device to output different prompt modes according to different system operating states; when the microprocessor determines that the system is in the water and starts the monitoring mode, it controls the miniature vibration motor to generate continuous short vibrations; when the microprocessor determines that the system is in a state of continuous instability or active struggle and starts sequential inflation, it controls the miniature buzzer to emit intermittent sounds; when the microprocessor controls the wireless communication module to continuously send a distress signal containing location information, it controls the miniature vibration motor and the miniature buzzer to work synchronously.

[0015] Preferably, in the aforementioned marine rescue life jacket, after the microprocessor controls the wireless communication module to continuously send a distress signal containing location information for a third preset duration, it automatically switches to a low-power tracking mode. In low-power tracking mode, the microprocessor periodically executes a work cycle, each work cycle including a short activation window and a long sleep window. During the activation window, the microprocessor activates the GPS module for positioning and controls the wireless communication module to send a distress signal containing the latest location information once. During the sleep window, the microprocessor shuts down the GPS module and puts the wireless communication module into a low-power standby state. The microprocessor has multiple pre-stored correspondences between power thresholds and sleep window durations, and the microprocessor selects and sets the corresponding sleep window duration based on the threshold range where the remaining power of the battery module is located.

[0016] Preferably, in the aforementioned marine rescue life jacket, the microprocessor is further configured to execute a hazard clearance determination process; while the wireless communication module continuously sends distress signals, the microprocessor continuously monitors the data from the four water pressure sensors and the data from the three-axis gyroscope; when all four water pressure sensors continuously detect a water depth of less than 0.1m, and the pitch and roll angles calculated by the three-axis gyroscope change by less than 10 degrees within 30 seconds, the microprocessor determines that the user has escaped the water hazard; the microprocessor then controls the wireless communication module to stop sending distress signals and causes the entire system to exit the rescue state and return to the standby mode where it has not been determined to be in the water.

[0017] The present invention has at least the following beneficial effects: The present invention aims to provide an intelligent life jacket for maritime rescue. By innovatively integrating multi-source sensors and intelligent control algorithms, it solves the technical bottlenecks of traditional life jackets in terms of accuracy of water-fall detection, inflation safety, continuous status monitoring and rescue reliability, and builds a comprehensive rescue system that integrates automatic detection, intelligent judgment, active alarm and long-term standby.

[0018] This invention establishes a multi-dimensional water-fall determination mechanism by combining a four-point distributed water pressure sensor layout with an attitude detection device. This not only significantly improves the accuracy of identifying standard water-fall postures, but also effectively addresses complex scenarios such as side entry into water and partial immersion, ensuring that the system can reliably start under any dangerous water-fall conditions. This fundamentally avoids the problems of missed judgments and misjudgments in traditional single-sensor solutions.

[0019] This invention innovatively designs a phased sequential inflation control strategy. By precisely controlling the opening and closing sequence of the gas distribution valve, it achieves sequential inflation of the buoyancy chambers in the front chest and back. This aims to effectively alleviate the instantaneous impact force generated by the traditional simultaneous inflation method, help the person who has fallen into the water to smoothly obtain buoyancy and gradually restore balance, and significantly reduce the risk of secondary injury caused by inflation impact, such as choking, panic, and loss of posture.

[0020] This invention establishes a struggle behavior recognition model based on Fast Fourier Transform and approximate entropy calculation, and a dangerous state monitoring logic based on continuous posture instability, enabling the life jacket to have a deep perception capability of the user's state after falling into the water. It can make intelligent decisions and automatically trigger the location and distress signal transmission in different dangerous states such as continuous instability, active struggle or confusion, ensuring that distress information is sent within the golden rescue time.

[0021] This invention establishes a comprehensive system health management and fault response system, including inflation valve position monitoring, sensor self-diagnosis and cross-validation, and graded degradation operation modes. It aims to monitor the working status of key execution units and sensing units in real time, promptly detect and report faults, and automatically switch to backup safety strategies when components fail, thereby greatly improving the system's durability and overall reliability in harsh environments.

[0022] This invention introduces a low-power tracking mode and an intelligent power management strategy. By dynamically adjusting the working cycle and the sleep window based on the remaining power, it solves the problem of equipment battery life during long-term waiting for rescue, ensuring that the positioning and communication functions remain effective throughout the rescue process. At the same time, through the danger clearance judgment process, the alarm is automatically stopped after the user is safe, avoiding unnecessary consumption of power and communication resources.

[0023] This invention emphasizes user experience and psychological support, integrating a manual alarm button to provide an active distress signal and designing a multimodal human-computer interaction device. Through differentiated vibrations, sounds, and intelligently adjusted optical signals, it clearly and in real time conveys the system's operating status to the user, effectively reducing their panic and uncertainty in critical moments, enhancing their confidence in waiting for rescue, and improving the overall efficiency of the rescue process from a humanistic perspective.

[0024] This invention ultimately realizes a complete rescue closed loop, from accurate hazard perception, intelligent decision-making and response, reliable action execution to long-term and continuous support, significantly improving the automation, intelligence level and success rate of maritime rescue.

[0025] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Detailed Implementation

[0026] The present invention will now be described in further detail so that those skilled in the art can implement it based on the description.

[0027] This invention provides a life jacket for maritime rescue, comprising: a life jacket body; four water pressure sensors installed at the hem of the life jacket body, located at the front, rear, left, and right sides of the life jacket body respectively; and a buoyancy providing device comprising: a front chest buoyancy chamber assembly and a rear chest buoyancy chamber assembly, respectively fixed to the front chest area and back area of ​​the life jacket body; the front and rear chest buoyancy chamber assemblies are made of high-strength composite fabric material, and after inflation, the total volume of the front chest chamber assembly is not less than 35L, and the total volume of the rear chest chamber assembly is not less than 25L, which can provide a net buoyancy of more than 150N for a standard adult, effectively ensuring that the head of the person falling into the water remains above the surface.A high-pressure gas cylinder is installed on the waist side of the life jacket body; a gas distribution valve has an inlet connected to the outlet of the high-pressure gas cylinder; the gas distribution valve has a first outlet and a second outlet, which are respectively connected to the front chest buoyancy chamber and the back buoyancy chamber via air guide tubes; a positioning and communication device, including a GPS module and a wireless communication module; an attitude detection device, including a three-axis accelerometer and a three-axis gyroscope located in the shoulder area of ​​the life jacket body; an integrated control device, including a microprocessor; four water pressure sensors, a three-axis accelerometer, a three-axis gyroscope, a gas distribution valve, a GPS module, and a wireless communication module (preferably integrating Beidou short message communication and 406MHz). The dual-mode module with EPIRB function (ensuring communication coverage and reliability in a wide sea area) is connected to the microprocessor. When at least three of the four water pressure sensors detect a water depth of 0.3-0.8m (0.3m as the lower limit to avoid false triggering by splashing water, and 0.8m as the upper limit, exceeding which indicates approaching a dangerous depth) for more than 2 seconds, the microprocessor determines that the water has entered the water and initiates monitoring mode. In monitoring mode, the microprocessor executes the following two judgment paths in parallel: Path 1: If the pitch or roll angle calculated by the three-axis gyroscope changes by more than 45 degrees within 2 seconds, and this state lasts for more than 8 seconds, the microprocessor determines that the water is in a state of continuous instability and initiates sequential inflation. Path 2: The microprocessor performs a fast Fourier transform on the data from the three-axis accelerometer to extract its dominant frequency and calculates its approximate entropy to quantify the signal disorder. The preset threshold for the approximate entropy is set to 0.5 (through extensive experimental verification, an approximate entropy of 0.5 can effectively distinguish struggling water). (In contrast to normal operation); if the main frequency is in the range of 3-8Hz and the approximate entropy is continuously higher than the preset threshold for 5 seconds, the microprocessor determines that it is in an active struggling state and starts sequential inflation; the sequential inflation is as follows: the gas distribution valve first opens the first outlet and continues for 2-3 seconds to inflate the anterior chest buoyancy chamber, then closes the first outlet and opens the second outlet for 3-4 seconds to inflate the posterior back buoyancy chamber; for example, the gas distribution valve first opens the first outlet and continues for a first time period (e.g., 2.5 seconds) to inflate the anterior chest buoyancy chamber, then closes the first outlet and opens the second outlet for a second time period (e.g., 3.5 seconds) to inflate the posterior back buoyancy chamber; when at least three of the four water pressure sensors detect a water depth of more than 1m, or after the sequential inflation is completed, the microprocessor still continuously detects a change in pitch or roll angle of more than 45 degrees for 10 seconds, then the microprocessor activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

[0028] Life jackets used for maritime rescue typically employ a single water pressure sensor mounted on the chest area of ​​the jacket body to detect when someone has fallen into the water. When the sensor detects a water depth reaching a preset value, such as 0.5m, it immediately triggers the inflation device to simultaneously inflate the buoyancy chambers in the front chest and back. This design is prone to misjudgment or delayed response when partially submerged or disturbed by waves. Furthermore, the instantaneous impact of inflation can cause the person in the water to lose balance, increasing the risk of choking. In addition, current technology lacks attitude monitoring capabilities and cannot identify persistent instability or struggling after falling into the water, thus failing to automatically send distress signals. Rescue operations rely on manual operation, resulting in low efficiency.

[0029] This solution utilizes four water pressure sensors positioned at the front, back, left, and right sides of the life jacket hem for multi-point detection. When at least three sensors detect a water depth of 0.3-0.8m for at least 2 seconds, the microprocessor determines the user is in water and activates monitoring mode. This improves the accuracy and reliability of water entry detection, avoiding misjudgments caused by localized immersion. In monitoring mode, the microprocessor executes two decision paths in parallel. Path one calculates the change in pitch or roll angle using a three-axis gyroscope. If the change exceeds 45 degrees within 2 seconds and continues for 8 seconds, it is determined to be a state of continuous instability, thus identifying a dangerous situation where the user cannot maintain balance independently. Path two performs a Fast Fourier Transform on the three-axis accelerometer data to extract the dominant frequency and calculates the approximate entropy. If the dominant frequency is within the 3-8Hz range and the approximate entropy remains above a preset threshold for 5 seconds, it is determined to be a state of active struggle, thus detecting the user's violent struggling behavior. Water depths between 0.8m and 1.0m are considered by the system as a continuous danger state that has triggered water entry and is in monitoring mode. Within this depth range, the triggering of a distress signal no longer depends on further increases in water depth, but rather on the user's posture stability or whether sequential inflation has been completed. The system continuously runs two judgment paths in monitoring mode. If persistent instability or active struggling is detected, sequential inflation is executed; if posture instability persists for a specified time after sequential inflation, location and distress signals are triggered. Therefore, this water depth range is a crucial stage for the system to actively assess the user's state and decide on subsequent rescue actions through multi-sensor fusion logic. In the sequential inflation method, the gas distribution valve first opens the first outlet for 2-3 seconds to inflate the chest buoyancy chamber, then closes the first outlet and opens the second outlet for 3-4 seconds to inflate the back buoyancy chamber. This staged inflation reduces the impact of sudden inflation and helps the user gradually regain stability. When the water depth exceeds 1 meter, or if the microprocessor still detects a pitch or roll angle change exceeding 45 degrees for 10 seconds after sequential inflation, the GPS module is activated for location, and the wireless communication module is controlled to continuously send distress signals containing location information. This ensures automatic distress signaling when danger persists, improving rescue response speed.

[0030] Example 1: The life jacket structure includes a life jacket body. Four water pressure sensors are installed at the hem of the life jacket body, located at the front, rear, left, and right sides of the life jacket body, respectively. The buoyancy supply device includes a front chest buoyancy chamber assembly and a rear chest buoyancy chamber assembly, fixed to the front chest and back areas of the life jacket body, respectively. A high-pressure gas cylinder is installed on the waist side of the life jacket body. The inlet of the gas distribution valve is connected to the outlet of the high-pressure gas cylinder. The gas distribution valve has a first outlet and a second outlet, which are connected to the front chest buoyancy chamber assembly and the rear chest buoyancy chamber assembly, respectively, via air guide tubes. The positioning and communication device includes a GPS module and a wireless communication module. The attitude detection device includes a three-axis accelerometer and a three-axis gyroscope located in the shoulder area of ​​the life jacket body. The integrated control device includes a microprocessor, and the four water pressure sensors, the three-axis accelerometer, the three-axis gyroscope, the gas distribution valve, the GPS module, and the wireless communication module are respectively connected to the microprocessor. In a simulated water-fall scenario, a user wearing a life jacket fell into the water. Four water pressure sensors detected a water depth of 0.5m within 1.5 seconds, and the microprocessor determined the user was in the water and activated monitoring mode after 2 seconds. Due to wave impact, the user's posture became unstable, with pitch changes reaching 50 degrees for 10 seconds. The microprocessor determined this instability and initiated sequential inflation. The gas distribution valve first opened the first outlet for 2.5 seconds to inflate the chest, then closed the first outlet and opened the second outlet for 3.5 seconds to inflate the back. After sequential inflation, the user's buoyancy increased, but the posture remained unstable, with pitch changes exceeding 45 degrees for 10 seconds. The microprocessor activated GPS positioning and sent a distress signal. The rescue center received the signal within 30 seconds and dispatched a rescue team. The results showed that the life jacket accurately identified the water-fall state and dangerous posture, reduced impact through sequential inflation, automatically sent distress signals, and improved rescue efficiency.

[0031] Comparative Example 1 life jacket structure includes the same main body as Example 1. Comparative Example 1 also includes a high-pressure gas cylinder, GPS module, and wireless communication module. However, Comparative Example 1 uses only one water pressure sensor, installed on the chest of the main body. Although its buoyancy supply device includes front and back buoyancy chambers, the gas distribution valve has only one outlet, connected to both the front and back buoyancy chambers via an air duct to simultaneously inflate all chambers. Comparative Example 1 does not include any attitude detection device, such as a three-axis accelerometer or gyroscope. Its control device is a simple control unit, connected only to a single water pressure sensor and gas distribution valve, lacking any monitoring mode judgment logic. In the test, when a user fell into the water, the single water pressure sensor detected a water depth of 0.5m after 2 seconds and triggered inflation. All buoyancy chambers inflated simultaneously, causing the user to lose balance due to the impact, and their head submerged in the water, leading to choking. Since the life jacket lacks attitude monitoring functionality, it cannot detect continuous instability or struggle, and therefore does not send a distress signal. Users need to rely on manually calling the police or waiting for rescue, which increases the risk of rescue delays.

[0032] Results: Comparative Example 1 relies on a single sensor, which is prone to detection delays or failures in partial immersion situations. In addition, the sudden impact of inflation increases the risk of choking, and the lack of subsequent status monitoring makes it impossible to automatically call for help, resulting in low rescue efficiency.

[0033] In another embodiment, the life jacket for marine rescue is provided with a first valve position sensor and a second valve position sensor at the first and second outlets of the gas distribution valve, respectively. The first and second valve position sensors are connected to a microprocessor. The microprocessor is configured to, after issuing a command to open the first or second outlet, if it does not receive an opening signal from the corresponding valve position sensor within a first preset time period (e.g., 2 seconds), determine that there is an inflation failure, and control the wireless communication module to send an alarm signal containing the inflation failure status.

[0034] In actual use of life jackets for maritime rescue, although the sequential inflation logic can effectively provide buoyancy, the gas distribution valve itself may fail to open properly due to blockage, mechanical jamming, or circuit failure, resulting in inflation failure. Because the system lacks verification of the execution results, neither the user nor the rescue center is aware of this critical malfunction, leaving the person in the water in a fatal state without buoyancy protection.

[0035] This solution adds a first valve position sensor and a second valve position sensor at the first and second outlets of the gas distribution valve, respectively, and connects them to a microprocessor. The purpose is to directly monitor the actual physical state of the valve actuator and confirm whether it has responded to the microprocessor's opening command. When the microprocessor issues a command to open the first or second outlet, a monitoring window of a first preset duration is initiated. Within this window, if no opening signal is received from the corresponding valve position sensor, it indicates that the valve has failed to open successfully, and the microprocessor determines that an inflation failure has occurred. This mechanism extends the system's monitoring scope from command issuance to action completion, achieving closed-loop control of the critical inflation process. After determining an inflation failure, the microprocessor immediately controls the wireless communication module to send an alarm signal containing the inflation failure status. This allows the rescue center, upon receiving the signal, not only to obtain the user's location but also to clearly understand that the user is facing the extreme danger of buoyancy device failure, thus enabling priority and emergency dispatch of rescue forces, while also guiding the user to take other self-rescue measures.

[0036] Results: By adding valve position sensor monitoring and fault diagnosis logic, this solution addresses the core safety hazard of undetected inflation system failures. It transforms a potentially fatal malfunction into an identifiable and reportable event, significantly enhancing the safety of life vests in real-world failure scenarios and providing crucial information for rescue responses.

[0037] In another embodiment, in the aforementioned marine rescue life jacket, if, after sequential inflation, the microprocessor continuously detects a change in pitch or roll angle exceeding 45 degrees for 10 seconds, the microprocessor initiates a 10-second behavior verification period. During this period, the microprocessor analyzes data from the triaxial accelerometer. If it identifies a pre-defined pattern of waving or splashing motions, it immediately activates the GPS module for positioning and controls the wireless communication module to continuously transmit a distress signal containing location information. During the behavior verification period, the microprocessor analyzes the triaxial accelerometer data, identifying the pattern of waving or splashing motions through time and frequency domain analysis. Specifically, the microprocessor performs a Fast Fourier Transform (FFT) on the triaxial accelerometer data to extract frequency features and calculates the power spectral density of the signal in the 1-3Hz range. If periodic peaks (e.g., 1-2 times per second) are detected within this frequency range for 3 consecutive seconds, and the amplitude exceeds a pre-defined threshold (e.g., 0.5g), it is determined to conform to the pattern of waving or splashing motions. If such characteristics are identified, the GPS module is immediately activated for positioning, and the wireless communication module is controlled to continuously send a distress signal containing location information. It's important to note that while the behavioral verification period is designed as a final step to confirm intent before triggering a distress call, the system always prioritizes the user's safety above all else. Therefore, during the behavioral verification period, the microprocessor executes another risk assessment algorithm based on vital sign proxy indicators in parallel. This algorithm quantifies the user's activity intensity by analyzing the overall energy and complexity of the triaxial accelerometer data. If, during the behavioral verification period, no regular movements are identified, but the user's activity intensity is consistently below a preset coma activity threshold (e.g., signal amplitude consistently below 0.1g), the system will prioritize determining that the user may have lost the ability to independently seek help due to coma or extreme exhaustion. In this case, the system will immediately interrupt the standard verification process and directly forcefully activate the positioning and distress call functions, ensuring the fastest possible response at the user's most dangerous moment and avoiding rescue delays caused by waiting for the verification period to end.

[0038] To address potentially complex situations, the system has established clear handling priorities: During the behavioral verification period, if the microprocessor simultaneously detects both "persistent attitude instability" (pitch or roll angle changes exceeding 45 degrees) and "activity intensity below the coma threshold," the system will prioritize the "coma threshold" assessment. This is because extremely low activity intensity combined with persistent attitude instability is a strong indication that the user has lost consciousness and is in a dangerous state at the mercy of the water flow, far more dangerous than conscious attitude instability. In this case, the system will immediately interrupt all verification processes and forcibly activate the location and distress call functions. This priority setting ensures the system makes the fastest and most conservative response when the user faces the highest risk to their life.

[0039] If no pre-defined regular waving or splashing motion is detected during the behavior verification period, the microprocessor records a verification failure and enters a 60-second standby monitoring period. After the standby monitoring period ends, if the microprocessor detects a change in pitch or roll angle exceeding 45 degrees for 10 seconds, the behavior verification period is restarted. If the number of verification failures reaches three, the microprocessor forcibly activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

[0040] When using a life jacket for maritime rescue, the user may remain in a state of continuous instability even after sequential inflation. This state can be caused by various reasons, such as the user attempting to swim or adjust their posture, or losing consciousness and being unable to control their body due to unconsciousness. If the system immediately sends a distress signal based solely on posture data continuously exceeding a threshold, it may lead to prematurely issuing a distress call when the user is still capable of self-rescue, resulting in the misuse of rescue resources, or delaying crucial rescue due to a lack of further confirmation when the user has truly lost their ability to move.

[0041] After sequential inflation, if the microprocessor continuously detects a pitch or roll angle change exceeding 45 degrees for 10 seconds, it does not immediately send a distress signal. Instead, it initiates a 10-second behavioral verification period. The purpose is to distinguish between two distinct danger levels: "able to move but with unstable posture" and "confused or incapacitated." During the behavioral verification period, the microprocessor analyzes data from the triaxial accelerometer, specifically identifying pre-defined, regular hand-waving or splashing motions. If these conscious distress behaviors are identified, the positioning and communication modules are immediately activated to send a distress signal, enabling a rapid response.

[0042] If no regular actions are identified during the behavior verification period, the system does not immediately abandon the attempt but records a verification failure and enters a 60-second standby monitoring period. This provides a window for observing possible changes in the user's state, avoiding an immediate judgment of the worst-case scenario due to a brief interruption of consciousness or physical exhaustion. After the standby monitoring period ends, if the postural instability condition is met again, the behavior verification period is restarted for a second confirmation. This cyclical logic aims to capture potential fluctuations in the user's state. Only when the number of verification failures accumulates to three, indicating that the user has remained in an unstable state after multiple check windows and has not exhibited any conscious, regular actions, does the system forcibly activate location tracking and send a distress signal. This is essentially a high-confidence assessment that the user may have fallen into a coma or be unable to act.

[0043] Results: By introducing a behavior verification period, a standby monitoring period, and a cumulative failure mechanism, this solution effectively distinguishes between users' active activities and passive dangerous states. It significantly reduces the possibility of false distress signals due to users' temporary and autonomous activities, while ensuring the final identification and rescue of users who are truly incapacitated through multiple verifications, achieving a balance between precise allocation of rescue resources and life protection in critical moments.

[0044] In another embodiment, the marine rescue life jacket further includes: a waterproof manual alarm button located in the shoulder area of ​​the life jacket body and connected to a microprocessor; when the microprocessor detects that the waterproof manual alarm button has been pressed continuously for more than 2 seconds, it immediately executes a sequential inflation process, activates the GPS module for positioning, and simultaneously controls the wireless communication module to continuously send a distress signal containing a manual alarm indicator and location information; an optical alarm device, including forward-facing LED modules respectively located on the front sides of the left and right shoulder straps of the life jacket body, and rearward-facing LED modules respectively located on the rear sides of the left and right shoulder straps of the life jacket body; the optical alarm device is connected to the microprocessor. The device is configured to: when the microprocessor controls the wireless communication module to send a distress signal, simultaneously activate the optical alarm device; the microprocessor acquires the pitch angle data calculated by the three-axis gyroscope in real time; the microprocessor is configured to: compare the pitch angle data with a preset positive angle threshold (e.g., +30 degrees) and a preset negative angle threshold (e.g., -30 degrees); when the pitch angle data is greater than the positive angle threshold, the microprocessor increases the light intensity of the forward LED module while decreasing the light intensity of the rearward LED module; when the pitch angle data is less than the negative angle threshold, the microprocessor increases the light intensity of the rearward LED module while decreasing the light intensity of the forward LED module.

[0045] In the application of life jackets for maritime rescue, the basic approach relies on automatic detection and triggering. However, when the automatic system fails to activate for any reason, or when the user anticipates impending danger, there is a lack of a rapid and reliable method for manual intervention to immediately initiate all rescue procedures. Furthermore, after a radio distress signal is sent, how to effectively improve the optical visibility of the person in the water, especially at night or in poor visibility conditions, and further provide clear directional guidance to approaching rescue forces, is also a problem that urgently needs to be solved.

[0046] This solution addresses the issue of manual triggering by adding a waterproof manual alarm button. This button is located on the shoulder area of ​​the life jacket, making it easy for the user to find and operate. When the microprocessor detects that the button has been pressed continuously for more than 2 seconds, it immediately executes the entire rescue process, including sequential inflation and sending a distress signal with a manual alarm indicator. This design allows the user to proactively call for help in any situation, ensuring a timely rescue response. Simultaneously, the solution incorporates an optical alarm device, consisting of forward-facing LED modules located on the front of the left and right shoulder straps and rearward-facing LED modules on the back of the left and right shoulder straps. When the system sends a distress signal, the optical alarm device activates simultaneously, significantly increasing the user's visibility on the sea surface by emitting a strong light.

[0047] To optimize the effectiveness of the optical alarm and transmit directional information, this solution further introduces an intelligent light intensity adjustment mechanism. The microprocessor acquires the pitch angle data calculated by the three-axis gyroscope in real time and compares it with a positive angle threshold and a negative angle threshold. When the pitch angle data is greater than the positive angle threshold, indicating that the user may be in an upward-facing, supine position, the microprocessor increases the light intensity of the forward-facing LED module while decreasing the light intensity of the rearward-facing LED module, ensuring the light source primarily illuminates the water in front of the user, facilitating identification by rescuers approaching from the front. Conversely, when the pitch angle data is less than the negative angle threshold, indicating that the user may be in a downward-facing, supine position, the microprocessor increases the light intensity of the rearward-facing LED module while decreasing the light intensity of the forward-facing LED module, directing the main light source backward. This dynamic adjustment function allows the optical signal to more accurately indicate the user's body orientation, providing rescuers with crucial directional reference.

[0048] Results: This solution, by introducing a manual alarm mechanism, empowers users with the ability to proactively survive in emergencies, compensating for the potential blind spots of purely automatic systems. The addition of optical alarm devices and their intelligent direction adjustment function greatly enhances the visibility and discoverability of those who have fallen into the water in complex sea conditions. This not only aids in long-distance search and rescue but also provides clear directional guidance for close-range rescue, thus constructing a comprehensive rescue guarantee from proactive triggering to efficient discovery.

[0049] In another embodiment, in the life jacket for maritime rescue, after the wireless communication module sends a distress signal containing location information each time, the microprocessor initiates a waiting window; if the wireless communication module does not receive a confirmation signal from the external rescue center within the waiting window, the microprocessor determines that the distress signal transmission has failed and controls the wireless communication module to switch the communication channel or retransmit the distress signal with enhanced transmission power.

[0050] In the application of life jackets for maritime rescue, automatically sending a distress signal containing location information is a core function. However, in complex marine environments or remote waters, wireless communication signals may be interfered with, blocked, or extremely weak due to the distance from the rescue center. A single signal transmission carries a high risk of failure. If the system cannot confirm successful delivery of the distress signal after transmission and lacks any remedial measures, the person in the water may be left helpless and isolated because crucial distress information has not been transmitted.

[0051] This solution addresses the issue of unreliable signal transmission by establishing a communication confirmation and retransmission mechanism. After the wireless communication module transmits a distress signal containing location information, the microprocessor does not immediately proceed to the next step but instead initiates a specific waiting window. The purpose of this waiting window is to allow sufficient time to receive a confirmation signal from the external rescue center, which serves as the basis for determining the success of the communication. This is a proactive communication quality assessment.

[0052] If the wireless communication module fails to receive a confirmation signal from the external rescue center within the waiting window, the microprocessor will determine that the distress signal transmission has failed. This determination triggers the system's autonomous recovery procedure. The microprocessor then controls the wireless communication module to execute two possible error correction strategies: one is to switch the communication channel to avoid potential interference or congestion; the other is to use enhanced transmission power to improve signal strength and coverage, and then retransmit the distress signal. This design ensures that after a transmission failure, the system can automatically and immediately try alternative solutions, significantly increasing the probability of the distress message being successfully delivered.

[0053] Results: This solution effectively overcomes the communication uncertainty of a single transmission mode by introducing a communication confirmation mechanism and an adaptive retransmission strategy after failure. It elevates the transmission of distress signals from a potentially failed single action into a reliable communication process with feedback and error correction capabilities, greatly enhancing the life jacket's ability to ensure successful information transmission in harsh communication environments and providing critical information link reliability for rescue operations.

[0054] In another embodiment, the microprocessor in the aforementioned marine rescue life jacket is further configured to execute a sensor self-diagnostic process. In standby mode, where the water immersion state is not determined, the microprocessor periodically performs the following operations: Based on angular velocity data measured by a three-axis gyroscope, a first attitude change trajectory is obtained through integration calculation; simultaneously, based on acceleration data measured by a three-axis accelerometer, a second attitude change trajectory is obtained by integrating after removing the gravitational acceleration component. Specifically, based on the acceleration data measured by the three-axis accelerometer, the gravitational acceleration component is estimated and separated from the acceleration data in real time using a Kalman filter-based attitude calculation algorithm to obtain pure motion acceleration data; then, the motion acceleration data is integrated to obtain the second attitude change trajectory; the microprocessor continuously compares the consistency between the first and second attitude change trajectories within a second preset time period (e.g., 30 seconds); if the difference between the first and second attitude change trajectories exceeds a preset tolerance range (e.g., the average difference in attitude angles exceeds 10%), the microprocessor... The system detects a faulty sensor unit in either the three-axis accelerometer or the three-axis gyroscope and controls the wireless communication module to send a warning message containing a specific faulty unit identifier. Subsequently, the microprocessor executes the following processing steps: it records the specific faulty sensor unit identifier in its internal memory and switches the system to the corresponding degraded operating mode accordingly. If the faulty sensor unit is either the three-axis accelerometer or the three-axis gyroscope, the microprocessor, in subsequent monitoring modes, masks the data from that faulty sensor unit and only uses data from another normally functioning sensor and the four water pressure sensors to determine the water immersion state and the continuous instability state. If the faulty sensor unit includes both the three-axis accelerometer and the three-axis gyroscope, the microprocessor determines that the system has entered a basic safety mode. In basic safety mode, when at least three of the four water pressure sensors detect a water depth exceeding 1m, the microprocessor directly activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information and the system's degraded status.

[0055] During long-term use of life jackets for maritime rescue, core attitude detection devices such as triaxial accelerometers and triaxial gyroscopes may experience performance drift or latent malfunctions due to long-term vibration, humid environments, or component aging. If the system cannot identify the abnormal state of these critical sensors before or during an incident and continues to rely on their data for critical judgments, it may lead to misjudgments of the water entry status or dangerous attitude, resulting in serious problems such as misinflation, failure to inflate, or false distress signals, directly affecting the reliability of the rescue.

[0056] This solution addresses this potential issue by introducing a sensor self-diagnostic process. In standby mode, where the system is not determined to be in a water-immersed state, the microprocessor periodically performs a self-test. Its core principle is to cross-validate the data obtained from measuring the motion of an object using data from a three-axis gyroscope and a three-axis accelerometer. Specifically, the microprocessor calculates the first attitude change trajectory based on the angular velocity data measured by the three-axis gyroscope through integration. Simultaneously, it calculates the second attitude change trajectory based on the acceleration data measured by the three-axis accelerometer, after removing the gravitational acceleration component and performing integration. Within a second preset time period, the microprocessor continuously compares the consistency between these two attitude change trajectories, which originate from different physical principles.

[0057] If the difference between the two trajectories exceeds the preset fault tolerance range, it indicates that at least one sensor unit outputs abnormally. Based on this, the microprocessor determines that there is a faulty sensor unit in the three-axis accelerometer or three-axis gyroscope, and immediately controls the wireless communication module to send a warning message containing a specific faulty unit identifier, thereby providing early warning to maintenance personnel or users.

[0058] Subsequently, the system does not completely shut down but enters an intelligent degraded operation mode. The microprocessor records the identifier of the faulty sensor unit in its internal memory and switches the system strategy accordingly. If the fault involves only one of the three-axis accelerometer and three-axis gyroscope, the microprocessor will mask the data of the faulty unit in subsequent monitoring modes and continue to execute the judgment logic for the water immersion state and the continuous instability state based only on the data of the other normally functioning sensor and the data of the four water pressure sensors, preserving core functions to the greatest extent possible. If both attitude sensors are judged to be faulty, the system enters a basic safety mode. In this mode, the judgment logic is greatly simplified. When at least three of the four water pressure sensors detect a water depth exceeding 1m, the positioning and communication functions are directly activated to send a distress signal containing the system's degraded status.

[0059] Results: This solution achieves proactive monitoring of the health status of critical sensors through periodic sensor self-diagnosis and cross-validation, enabling timely detection of latent faults. By establishing a tiered degradation operation strategy, it ensures that even in the event of partial or complete sensor failure, the life vest can still provide the highest level of safety commensurate with the current hardware capabilities, greatly enhancing the overall robustness and survivability of the system under real-world complex operating conditions.

[0060] In another embodiment, the microprocessor in the aforementioned marine rescue life jacket is also configured to execute an auxiliary water entry determination process. This process follows the principle of "high threshold triggering and continuous observation confirmation," specifically as follows: Step 1: Triggering the auxiliary monitoring mode. In standby mode where no water entry is determined, if no more than two of the four water pressure sensors detect a water depth of 0.3-0.8m, but simultaneously, the pitch and roll angles calculated by the three-axis gyroscope change by more than 60 degrees within 3 consecutive seconds, the system determines that the body may have experienced a severe loss of control similar to falling into the water. The microprocessor then initiates a 10-second auxiliary monitoring mode for depth assessment. The reliability of this auxiliary water entry determination logic is based on the fusion judgment of the signals of "severe attitude change" and "water contact." The precondition of "no more than two water pressure sensors detecting water depth" is set precisely to capture scenarios of partial body entry into the water or non-standard posture water entry. Meanwhile, during the 10-second observation period in the auxiliary monitoring mode, the system requires that the water pressure or attitude anomaly signal be continuous. This effectively filters out brief, severe attitude swaying on the deck caused by ship pitching. This is because, in non-overwater situations, it is difficult for the human body to simulate a severe and continuous rolling motion lasting more than 10 seconds with roll and pitch angles exceeding 60 degrees. Furthermore, in the tests of this embodiment, we simulated deck environments under various ship navigation conditions, verifying that this logic has extremely high reliability in such scenarios, with a false trigger rate of less than 1‰.

[0061] The specific anti-interference verification experiment was conducted on a simulated deck, using a mechanical platform to reproduce the ship's roll (±15 degrees, period 5-8 seconds) and pitch (±10 degrees, period 4-6 seconds) movements under sea state 4 (wave height 1.5-2.5 meters, corresponding to significant wave height). In this severe and continuous swaying environment, testers simulated typical deck work actions (such as walking, falling, and carrying items), conducting over 1000 test cycles. The experimental results show that this judgment logic can effectively distinguish between "active and passive human movement on the deck" and "uncontrolled rolling at the moment of falling into the water and subsequent events," with a false trigger rate of less than 1‰. This demonstrates that the combination of a 10-second observation period and a 60-degree amplitude threshold can reliably filter out interference in non-fall-in-the-water scenarios.

[0062] Step 2: Confirmation and Judgment During the Assisted Monitoring Period. During this 10-second assisted monitoring mode, the microprocessor executes parallel monitoring of water pressure sensor data and attitude data. It will officially determine the user as being in the water when either of the following conditions is met: Condition A (Water Pressure Confirmation): If, within 10 seconds of the assisted monitoring mode being activated, at least three of the four water pressure sensors detect a water depth of 0.3-0.8m, indicating that the body is fully submerged, the microprocessor will officially determine the user as being in the water and enter monitoring mode. Condition B (Continuous Abnormal Attitude Confirmation): If, within 10 seconds, the condition that at least three of the four water pressure sensors detect a water depth of 0.3-0.8m is not met, but the pitch and roll angles calculated by the three-axis gyroscope consistently exceed 60 degrees until the end of the 10-second observation period, this indicates that the user is continuously in a state of violent tumbling and is likely in the water but not fully submerged in their life jacket. In this case, the microprocessor will forcibly determine the user as being in the water and enter monitoring mode at the end of the 10 seconds.

[0063] The parameters in this judgment logic (60-degree amplitude, 3-second duration, and 10-second observation) are optimized based on extensive marine fall-over dynamics experiments and ship deck activity simulations. A rapid attitude change of up to 60 degrees is a typical characteristic of loss of control during a fall-over, effectively filtering out most deck turbulence and daily activities. The subsequent 10-second continuous observation reliably distinguishes between brief accidents such as "deck slippage" and genuine "continuous loss of control during a fall-over," ensuring high system reliability in complex marine environments, with a measured false trigger rate of less than 0.1%.

[0064] In practical applications of life jackets for maritime rescue, users may not enter the water in a standard posture with their entire body submerged simultaneously. For example, if a user falls into the water sideways or is partially obscured, only a few of the four water pressure sensors may be able to contact the water surface in time and detect a depth of 0.3-0.8m. In this case, the system will not be able to enter monitoring mode because the condition that at least three sensors must detect water depth is not met, resulting in a significant delay in responding to the fall into the water and missing a critical early rescue opportunity.

[0065] This solution addresses the risk of missed detections in this specific scenario by introducing an auxiliary water entry detection process. In standby mode, where water entry is not detected, the system does not rely entirely on water pressure sensor data. When no more than two of the four water pressure sensors detect a water depth of 0.3-0.8m, but simultaneously the pitch and roll angles calculated by the three-axis gyroscope change by more than 60 degrees within 3 consecutive seconds, the microprocessor activates the auxiliary monitoring mode. The purpose of this design is to capture anomalies where the body posture changes drastically despite weak water pressure signals, which is typically a characteristic of the moment of falling into water.

[0066] In auxiliary monitoring mode, the microprocessor performs parallel monitoring of water pressure sensor data and attitude data, conducting a comprehensive observation for 10 seconds. Its judgment logic has two levels. First, if at least three of the four water pressure sensors detect a water depth of 0.3-0.8m within 10 seconds of the auxiliary monitoring mode being activated, the microprocessor officially determines the user is in water and enters monitoring mode. This handles situations where slight delays in the entry action or water fluctuations cause temporary suboptimal sensor readings. Second, if the trigger conditions of at least three water pressure sensors are not met within 10 seconds, but the pitch and roll angles calculated by the three-axis gyroscope consistently exceed 60 degrees, indicating that the user is continuously in an abnormal posture such as violent tumbling, the microprocessor forcibly determines the user is in water and enters monitoring mode at the end of the 10 seconds. This ensures that the system will not completely fail due to the local limitations of the water pressure sensors when strong abnormal postures persist.

[0067] Results: This solution establishes a complementary water entry detection mechanism by fusing water pressure signals and signals of significant attitude changes. It effectively solves the system triggering delay problem caused by partial immersion or special landing postures, significantly improving the recognition sensitivity and reliability of life jackets in complex and non-standard water landing scenarios, and providing users with more timely basic protection.

[0068] In another embodiment, the life jacket for maritime rescue further includes a human-machine interface device, comprising a miniature vibration motor and a miniature buzzer, connected to a microprocessor. The microprocessor is configured to control the human-machine interface device to output different prompt modes according to different system operating states. When the microprocessor determines that the system is in the water and activates the monitoring mode, it controls the miniature vibration motor to generate continuous short vibrations. When the microprocessor determines that the system is in a state of continuous instability or active struggle and activates sequential inflation, it controls the miniature buzzer to emit intermittent sounds. When the microprocessor controls the wireless communication module to continuously send distress signals containing location information, it controls the miniature vibration motor and the miniature buzzer to work synchronously.

[0069] During the use of life jackets for maritime rescue, the system can automatically perform a series of complex operations, including water-fall detection, inflation, and distress signaling. However, when users are in danger, they may find it difficult to visually understand the system's current status due to panic, poor lighting, or being submerged in water. They cannot confirm whether the life jacket has correctly detected water-fall, whether it is inflating, or whether a distress signal has been sent. This uncertainty about the system's status can exacerbate the user's panic and psychological burden, preventing them from taking reasonable actions to cooperate with the rescue.

[0070] This solution addresses the problem of unclear state perception by integrating a human-computer interaction device. This device includes a miniature vibration motor and a miniature buzzer, which are connected to a microprocessor. The microprocessor is configured to control these devices to output differentiated prompting modes based on different system operating states, converting the system's internal logical state into tactile and auditory signals that the user can directly perceive.

[0071] When the microprocessor determines that the system is submerged and activates monitoring mode, it controls a miniature vibrating motor to produce continuous, short vibrations. This tactile cues quietly inform the user that the system has confirmed the submersion and entered an alert state without disturbing the surrounding environment, giving them peace of mind. When the microprocessor determines that the system is in a state of continuous instability or active struggling and initiates sequential inflation, it controls a miniature buzzer to emit intermittent beeps. This sound signal clearly indicates to the user that the inflation rescue procedure has been triggered, preparing them and alleviating the shock caused by sudden inflation. When the microprocessor controls the wireless communication module to continuously send a distress signal containing location information, it controls the miniature vibrating motor and the miniature buzzer to work synchronously. This strong combination of tactile and auditory cues aims to clearly inform the user that a distress message has been sent and rescue has been initiated, thus providing them with maximum psychological support and confidence in waiting for rescue.

[0072] Results: This solution establishes a status communication bridge between the system and the user through multimodal human-computer interaction design. It effectively eliminates the user's uncertainty and anxiety about the equipment's operating status in critical moments, enabling them to remain calm through timely status feedback and enhancing their confidence in being rescued, thereby improving the overall effectiveness of the rescue process from a psychological perspective.

[0073] In another embodiment, the life jacket used for maritime rescue automatically switches to low-power tracking mode after the microprocessor controls the wireless communication module to continuously send a distress signal containing location information for a third preset duration (e.g., 10 minutes). In low-power tracking mode, the microprocessor periodically executes a work cycle, each cycle including a short activation window (e.g., 10 seconds) and a long sleep window (e.g., 5 minutes, 10 minutes, or 20 minutes, the specific duration depending on the battery level). To ensure accurate battery level detection, the system employs an intelligent battery metering algorithm based on the voltage-load curve. This algorithm is calibrated for specific battery modules before leaving the factory and can effectively compensate for measurement errors caused by temperature changes and load fluctuations. The correspondence between the pre-stored battery level threshold and the sleep window in the microprocessor is set after rigorous power consumption simulation and endurance calculations. For example, the setting of a 20-minute hibernation window when the battery level drops below 30% is a trade-off made to maximize system survival time while ensuring that the rescue center can still receive location updates at an acceptable frequency (e.g., at least 3 times per hour) within the normal search and rescue radius. Meanwhile, during the hibernation window, the system is not completely "asleep"; the core water pressure sensor module and accelerometer continue to operate at extremely low power, continuously monitoring for sudden changes in water depth and attitude. Once an anomaly is detected, the system can be immediately awakened, interrupting the hibernation cycle, thus ensuring that the user remains under safe monitoring even in low-power mode.

[0074] During the active window, the microprocessor activates the GPS module for positioning and controls the wireless communication module to send a distress signal containing the latest location information. During the sleep window, the microprocessor shuts down the GPS module and puts the wireless communication module into a low-power standby state. The microprocessor has pre-stored multiple power thresholds and their corresponding sleep window durations. For example, when the remaining battery power is above 70%, the sleep window duration is 5 minutes; when the remaining battery power is between 30% and 70%, the sleep window duration is 10 minutes; and when the remaining battery power is below 30%, the sleep window duration is 20 minutes.

[0075] The microprocessor selects and sets the corresponding sleep window duration based on the threshold range of the remaining battery power of the battery module.

[0076] During maritime rescue operations using life jackets, once a distress signal is initiated, the system needs to remain operational to update its location and await rescue. However, both the GPS module and the wireless communication module are high-power devices. If they operate at full power continuously after the distress signal is sent, the battery module will be quickly depleted. This could result in the life jacket losing its positioning and communication functions due to battery depletion before rescuers arrive, making subsequent search efforts extremely difficult, or even rendering all previous efforts futile.

[0077] This solution addresses the battery life issue during extended rescue operations by introducing a low-power tracking mode. After the microprocessor-controlled wireless communication module continuously transmits a distress signal containing location information for a third preset duration, the system considers the emergency alarm information to have been sufficiently sent and automatically switches to low-power tracking mode. In this mode, the microprocessor no longer allows the system to operate at full power continuously, but instead periodically executes a carefully designed work cycle. Each work cycle includes a brief activation window and a long sleep window.

[0078] During the brief activation window, the microprocessor reactivates the GPS module to locate itself, acquire the latest location information, and controls the wireless communication module to send a distress signal containing this updated location information. This ensures the regular updating of location information. During the subsequent long sleep window, the microprocessor shuts down the high-power GPS module and puts the wireless communication module into a low-power standby state, thereby significantly reducing the overall system power consumption.

[0079] To further enhance intelligent power management, the microprocessor pre-stores multiple power thresholds and their corresponding sleep window durations. The microprocessor determines in real-time which threshold range the remaining battery power falls into, and automatically selects and sets the corresponding sleep window duration accordingly. For example, when the battery is fully charged, the sleep window may be shorter and its position updated more frequently; when the battery is low, the sleep window will automatically extend to conserve power in an extreme way, maximizing the system's uptime.

[0080] Results: This solution successfully solved the battery life problem of life vests during long waits for rescue by employing a work cycle and a power-based intelligent scheduling strategy. While maintaining the effectiveness of core positioning and communication functions, it significantly extended battery life, ensuring continuous tracking of those in the water and greatly improving the efficiency of subsequent search operations and the likelihood of a successful rescue.

[0081] In another embodiment, the microprocessor in the aforementioned maritime rescue life jacket is also configured to execute a hazard clearance determination process. While the wireless communication module continuously transmits distress signals, the microprocessor continuously monitors data from four water pressure sensors and a three-axis gyroscope. When all four water pressure sensors continuously detect a water depth of less than 0.1m, and the pitch and roll angles calculated by the three-axis gyroscope change by less than 10 degrees within 30 seconds, the microprocessor determines that the user has escaped the water hazard. This determination condition is set quite strictly, primarily to prevent misinterpretation of the distress signal due to waves briefly lifting the life jacket or the user briefly maintaining stability in the water, ensuring the continuity of the rescue link in real-world hazardous situations. We understand that in scenarios such as on a rescue ship deck, the life jacket may become partially wet. In laboratory simulations, we verified that when the user's body is out of the water and placed on a stable platform, even with a small amount of residual splash water, a 30-second observation period is sufficient for most water pressure sensor readings to return to normal and meet the determination condition. Furthermore, the system design also considers external intervention as an auxiliary means of resolving emergencies. For example, after rescuers arrive, they can press and hold the waterproof manual alarm button for more than 5 seconds to send a forced deactivation command to the microprocessor, manually causing the system to exit rescue mode.

[0082] Furthermore, to further enhance the system's applicability and reliability, the hazard clearance determination process also integrates a manual intervention mechanism. When the microprocessor detects that the waterproof manual alarm button has been triggered in a specific mode (e.g., pressed and held for more than 5 seconds), it also determines that the user has escaped danger. This design aims to address special scenarios where automatic judgment may fail (such as when a life jacket is partially snagged and soaked), allowing users or rescuers to proactively inform the system that the danger has been cleared. It serves as an important supplement and backup to the automatic judgment logic.

[0083] The microprocessor then controls the wireless communication module to stop sending distress signals and puts the entire system out of rescue mode, returning it to standby mode where it has not been determined to be in water.

[0084] After a life jacket completes a rescue response at sea, the system continuously sends distress signals. However, if the system fails to automatically recognize this safe state once the user is successfully rescued and reaches shore or leaves the water on their own, it will continue sending distress signals and issuing optical alarms indefinitely. This not only continuously consumes valuable battery power, shortening the equipment's lifespan, but more seriously, it continuously sends incorrect alarm messages to the rescue center, occupies emergency communication channels, interferes with rescue command, and may mislead rescue forces to locations where the danger has subsided, resulting in a serious waste of public rescue resources.

[0085] This solution addresses the issue of automatic exit from rescue status by introducing a hazard resolution determination process. While the wireless communication module continues to send distress signals, the microprocessor does not cease operation but continuously monitors data from four water pressure sensors and a three-axis gyroscope in parallel. It analyzes this crucial data to determine whether the user's actual environment and situation have undergone a fundamental change.

[0086] The determination is based on very strict conditions, requiring the simultaneous fulfillment of two requirements. First, all four water pressure sensors must continuously detect a water depth of less than 0.1 meters, indicating that the user's body is essentially and completely out of the water. Second, the pitch and roll angles calculated by the three-axis gyroscope must change by less than ten degrees over a continuous thirty seconds, indicating that the user's posture is highly stable, likely standing, lying down, or on a stable platform, rather than struggling in the water. Only when both conditions are met simultaneously does the microprocessor comprehensively determine that the user is out of danger in the water.

[0087] Once the danger is determined to be over, the microprocessor executes the exit procedure. It controls the wireless communication module to immediately stop sending distress signals, disables human-machine interaction devices such as optical alarms, and smoothly exits the entire system from the high-power rescue state to a low-power standby mode where the system is not determined to be submerged. In this mode, the system is ready to respond to the next possible danger.

[0088] Results: This solution endows life jackets with the ability to intelligently identify safety status and automatically reset. It effectively prevents unnecessary energy consumption and the occupation of rescue communication resources after an emergency has ended, avoids false alarms interfering with the rescue system, improves the overall operational efficiency and professionalism of the rescue system, and also helps extend the standby time and service life of the life jackets themselves.

[0089] The core judgment thresholds involved in this invention (such as water depth 0.3-0.8m, attitude angle 45 degrees and 60 degrees, approximate entropy threshold of 0.5, and positive and negative angle thresholds of ±30 degrees) were all optimized and determined based on extensive laboratory water simulation experiments, ergonomic data, and marine environmental data analysis. For example, the approximate entropy threshold of 0.5 was determined by performing machine learning classification on hundreds of sets of acceleration data from falling into the water and swimming normally, resulting in the highest recognition accuracy. All threshold settings are implemented to minimize the false alarm rate while ensuring safety.

[0090] All electronic units (including control boards, sensors, and batteries) are sealed in an IP68-rated waterproof enclosure. The main control microprocessor uses a low-power, high-performance ARM Cortex-M4 core chip. Its built-in hardware floating-point unit (FPU) and digital signal processing (DSP) instruction set ensure that complex algorithms such as Fast Fourier Transform (FFT) and approximate entropy can be calculated in much less than 1 second, meeting the system's real-time requirements.

[0091] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details.

Claims

1. A life jacket for maritime rescue, characterized in that, include: The main body of the life jacket; Four water pressure sensors are installed at the bottom of the life jacket body, and are located at the front, back, left and right of the life jacket body respectively; A buoyancy supply device includes: front and rear chest buoyancy chambers, fixed to the front chest and back areas of the life jacket body, respectively; a high-pressure gas cylinder installed on the waist side of the life jacket body; a gas distribution valve with its inlet connected to the outlet of the high-pressure gas cylinder; the gas distribution valve has a first outlet and a second outlet, which are respectively connected to the front chest buoyancy chambers and the rear back buoyancy chambers via air guide tubes; a positioning and communication device, including a GPS module and a wireless communication module; an attitude detection device, including a three-axis accelerometer and a three-axis gyroscope located in the shoulder area of ​​the life jacket body; an integrated control device, including a microprocessor; four water pressure sensors, the three-axis accelerometer, the three-axis gyroscope, the gas distribution valve, the GPS module, and the wireless communication module are respectively connected to the microprocessor; when at least three of the four water pressure sensors detect a water depth of 0.3-0.8m for more than 2 seconds, the microprocessor determines that the body is in water and starts the monitoring mode; in the monitoring mode, the microprocessor executes the following two judgment paths in parallel: Path 1: If the three-axis gyroscope calculates... If the pitch or roll angle changes by more than 45 degrees within 2 seconds and continues in this state for more than 8 seconds, the microprocessor determines it to be in a state of continuous instability and initiates sequential inflation. Path 2: The microprocessor performs a fast Fourier transform on the data from the triaxial accelerometer to extract its dominant frequency and calculates its approximate entropy to quantify the disorder of the signal. If the dominant frequency is in the range of 3-8Hz and the approximate entropy is continuously higher than the preset threshold for 5 seconds, the microprocessor determines it to be in a state of active struggle and initiates sequential inflation. Sequential inflation is as follows: the gas distribution valve first opens the first outlet and continues for 2-3 seconds to inflate the anterior chest buoyancy chamber, then closes the first outlet and opens the second outlet and continues for 3-4 seconds to inflate the posterior back buoyancy chamber. When at least three of the four water pressure sensors detect a water depth of more than 1m, or after sequential inflation is completed, the microprocessor still detects a pitch or roll angle change of more than 45 degrees for 10 seconds, then the microprocessor activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

2. The life jacket for maritime rescue as described in claim 1, characterized in that, A first valve position sensor and a second valve position sensor are respectively installed at the first outlet and the second outlet of the gas distribution valve. The first valve position sensor and the second valve position sensor are connected to a microprocessor. The microprocessor is configured to, when issuing a command to open the first outlet or the second outlet, if it does not receive an opening signal from the corresponding valve position sensor within a first preset time period, determine that there is an inflation fault, and control the wireless communication module to send an alarm signal containing the inflation fault status.

3. The life jacket for maritime rescue as described in claim 1, characterized in that, If, after sequential inflation, the microprocessor continuously detects a pitch or roll angle change exceeding 45 degrees for 10 seconds, it initiates a 10-second behavior verification period. During this period, the microprocessor analyzes data from the triaxial accelerometer. If it identifies a pre-defined pattern of waving or splashing motions, it immediately activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information. If no pre-defined pattern of waving or splashing motions is identified during the verification period, the microprocessor records a verification failure and enters a 60-second standby monitoring period. After the standby monitoring period ends, if the microprocessor again continuously detects a pitch or roll angle change exceeding 45 degrees for 10 seconds, it restarts the behavior verification period. If the number of verification failures accumulates to three, the microprocessor forcibly activates the GPS module for positioning and controls the wireless communication module to continuously send a distress signal containing location information.

4. The life jacket for maritime rescue as described in claim 1, characterized in that, Also includes: A waterproof manual alarm button is located on the shoulder area of ​​the life jacket and connected to a microprocessor. When the microprocessor detects that the waterproof manual alarm button has been pressed continuously for more than 2 seconds, it immediately executes the sequential inflation process, activates the GPS module for positioning, and simultaneously controls the wireless communication module to continuously send a distress signal containing a manual alarm indicator and location information. An optical alarm device includes forward-facing LED modules located on the front sides of the left and right shoulder straps of the life jacket, and rearward-facing LED modules located on the rear sides of the left and right shoulder straps. The optical alarm device is connected to the microprocessor. When the microprocessor detects that the waterproof manual alarm button has been pressed continuously for more than 2 seconds, it immediately executes the sequential inflation process, activates the GPS module for positioning, and simultaneously controls the wireless communication module to continuously send a distress signal containing a manual alarm indicator and location information. When the processor controls the wireless communication module to send a distress signal, it simultaneously activates the optical alarm device. The microprocessor acquires the pitch angle data calculated by the three-axis gyroscope in real time. The microprocessor is configured to compare the pitch angle data with a preset positive angle threshold and a preset negative angle threshold. When the pitch angle data is greater than the positive angle threshold, the microprocessor increases the light intensity of the forward LED module while decreasing the light intensity of the rear LED module. When the pitch angle data is less than the negative angle threshold, the microprocessor increases the light intensity of the rear LED module while decreasing the light intensity of the forward LED module.

5. The life jacket for maritime rescue as described in claim 1, characterized in that, After the wireless communication module sends a distress signal containing location information each time, the microprocessor starts a waiting window. If the wireless communication module does not receive a confirmation signal from the external rescue center within the waiting window, the microprocessor determines that the distress signal transmission has failed and controls the wireless communication module to switch the communication channel or retransmit the distress signal with enhanced transmission power.

6. The life jacket for maritime rescue as described in claim 1, characterized in that, The microprocessor is also configured to execute a sensor self-diagnostic process; in standby mode where the device is not determined to be in water, the microprocessor periodically performs the following operations: based on the angular velocity data measured by the triaxial gyroscope, it calculates a first attitude change trajectory through integration; simultaneously, based on the acceleration data measured by the triaxial accelerometer, it calculates a second attitude change trajectory by integrating after removing the gravitational acceleration component; the microprocessor continuously compares the consistency between the first and second attitude change trajectories within a second preset time period; if the difference between the first and second attitude change trajectories exceeds a preset fault tolerance range, the microprocessor determines that there is a faulty sensor unit in the triaxial accelerometer or triaxial gyroscope, and controls the device to... The line communication module sends a warning message containing a specific fault unit identifier. Subsequently, the microprocessor executes the following processing steps: the microprocessor records the specific fault sensor unit identifier in its internal memory and switches the system to the corresponding degraded operating mode accordingly. If the fault sensor unit is either a three-axis accelerometer or a three-axis gyroscope, the microprocessor, in subsequent monitoring modes, masks the data from that fault sensor unit and executes the judgment logic for the water immersion state and continuous instability state based solely on the data from another normally functioning sensor and the four water pressure sensors. If the fault sensor unit includes both a three-axis accelerometer and a three-axis gyroscope, the microprocessor determines that the system has entered a basic safety mode. In basic safety mode, when at least three of the four water pressure sensors detect a water depth exceeding 1m, the microprocessor directly activates the GPS module for positioning and controls the wireless communication module to continuously send distress signals containing location information and system degradation status.

7. The life jacket for maritime rescue as described in claim 1, characterized in that, The microprocessor is also configured to execute an auxiliary water entry determination process; in standby mode where water entry is not determined, if no more than two of the four water pressure sensors detect a water depth of 0.3-0.8m, but the pitch and roll angles calculated by the three-axis gyroscope change by more than 60 degrees within 3 consecutive seconds, the microprocessor will activate the auxiliary monitoring mode; in the auxiliary monitoring mode, the microprocessor will execute water pressure sensor data monitoring and attitude data monitoring in parallel. If, within 10 seconds of the activation of the auxiliary monitoring mode, at least three of the four water pressure sensors detect a water depth of 0.3-0.8m, the microprocessor officially determines that the water is in the water state and enters the monitoring mode. If, within 10 seconds, the condition that at least three of the four water pressure sensors detect a water depth of 0.3-0.8m is not met, but the pitch and roll angles calculated by the three-axis gyroscope continue to change by more than 60 degrees, the microprocessor forcibly determines that the water is in the water state and enters the monitoring mode at the end of the 10 seconds.

8. The life jacket for maritime rescue as described in claim 1, characterized in that, Also includes: The human-machine interface device includes a miniature vibration motor and a miniature buzzer, and is connected to a microprocessor. The microprocessor is configured to control the human-machine interface device to output different prompt modes according to different operating states of the system. When the microprocessor determines that the system is in a water-entry state and activates the monitoring mode, it controls the miniature vibration motor to generate continuous short vibrations. When the microprocessor determines that the system is in a state of continuous instability or active struggle and activates sequential inflation, it controls the miniature buzzer to emit intermittent sounds. When the microprocessor controls the wireless communication module to continuously send distress signals containing location information, it controls the miniature vibration motor and the miniature buzzer to work synchronously.

9. The life jacket for maritime rescue as described in claim 1, characterized in that, After the microprocessor controls the wireless communication module to continuously send a distress signal containing location information for a third preset duration, it automatically switches to low-power tracking mode. In low-power tracking mode, the microprocessor periodically executes a work cycle, each work cycle including a short activation window and a long sleep window. During the activation window, the microprocessor activates the GPS module for positioning and controls the wireless communication module to send a distress signal containing the latest location information once. During the sleep window, the microprocessor shuts down the GPS module and puts the wireless communication module into a low-power standby state. The microprocessor has multiple pre-stored relationships between power thresholds and sleep window durations. The microprocessor selects and sets the corresponding sleep window duration based on the threshold range in which the battery module's remaining power falls.

10. The life jacket for maritime rescue as described in claim 1, characterized in that, The microprocessor is also configured to execute a hazard clearance determination process. While the wireless communication module is continuously sending distress signals, the microprocessor continuously monitors the data from the four water pressure sensors and the three-axis gyroscope. When all four water pressure sensors continuously detect a water depth of less than 0.1m, and the pitch and roll angles calculated by the three-axis gyroscope change by less than 10 degrees within 30 seconds, the microprocessor determines that the user has escaped the water hazard. The microprocessor then controls the wireless communication module to stop sending distress signals and causes the entire system to exit the rescue state and return to the standby mode, where the user has not been identified as being in the water.