Outdoor help-seeking method and device, electronic equipment and computer readable storage medium

Abstract

The embodiment of the application discloses an outdoor help-seeking method and device, electronic equipment and computer readable storage medium, and relates to the technical field of outdoor sports safety; the method comprises the following steps: acquiring multiple source parameters in real time, wherein the multiple source parameters comprise weather parameters, environment parameters, sports parameters and physiological parameters; modeling the multiple source parameters at multiple time points to obtain risk scores at the multiple time points; time integrating the risk scores corresponding to the multiple time points included in a target period to obtain a comprehensive risk score of the target period; when the comprehensive risk score of the target period is higher than a period score threshold, issuing a reminder to make a target user perform a safety confirmation operation; and when the safety confirmation operation is not detected within a preset time length, sending help-seeking information. In this way, it can be judged whether to seek help accurately.

Description

Technical Field

[0001] This application relates to the field of outdoor sports safety technology, specifically to an outdoor rescue method, device, electronic device, and computer-readable storage medium. Background Technology

[0002] Wearable devices such as smart bracelets and smartwatches have been widely used in outdoor sports such as hiking, trail running, and long-distance running. However, the automatic identification and emergency assistance mechanisms of mainstream wearable devices for accidental injuries and sudden dangers in outdoor environments are still not accurate enough. They are easily affected by short-term data fluctuations, which can lead to misjudgments of the user's danger status and incorrect determination of the distress trigger conditions. This can easily lead to unnecessary rescue actions and cause ineffective consumption and waste of public rescue resources. Summary of the Invention

[0003] This application provides an outdoor distress signaling method, device, electronic device, and computer-readable storage medium that can accurately determine whether to request distress.

[0004] In a first aspect, embodiments of this application provide an outdoor distress signaling method, applied to a wearable device, comprising: Real-time acquisition of multi-source parameters, including weather parameters, environmental parameters, motion parameters, and physiological parameters; The multi-source parameters at multiple time points are modeled to obtain risk scores at multiple time points; The risk scores corresponding to each of the multiple moments included in the target time period are integrated over time to obtain the comprehensive risk score for the target time period. When the overall risk score for the target time period is higher than the time period score threshold, a reminder is issued to prompt the target user to perform a security confirmation operation; If the security confirmation operation is not detected within a preset time period, a distress message is sent.

[0005] Secondly, embodiments of this application provide an outdoor distress device, applied to wearable devices, comprising: The acquisition module is used to acquire multi-source parameters in real time, including weather parameters, environmental parameters, motion parameters, and physiological parameters. The modeling module is used to model the multi-source parameters at multiple time points to obtain risk scores at multiple time points; The integration module is used to integrate the risk scores corresponding to multiple moments within the target time period over time to obtain the comprehensive risk score for the target time period. The reminder module is used to issue a reminder to the target user when the overall risk score for the target time period is higher than the time period score threshold, so that the target user can perform a security confirmation operation. The distress module is used to send a distress message if the security confirmation operation is not detected within a preset time period.

[0006] Thirdly, embodiments of this application also provide an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps in the above-described outdoor rescue method.

[0007] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in the above-described outdoor rescue method.

[0008] Fifthly, embodiments of this application also provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described in embodiments of this application.

[0009] The embodiments of this application have the following beneficial effects: It can acquire multi-source parameters in real time, including weather, environmental, motion, and physiological parameters. Weather and environmental parameters accurately recreate complex outdoor conditions, effectively identifying external hazards such as severe weather, low temperatures, and low light. Motion parameters reflect user exercise intensity and posture changes, accurately identifying dangerous behaviors such as falls and sudden cessation of movement. Physiological parameters monitor abnormal states such as physical exhaustion and sudden discomfort. Multi-dimensional data collaboration effectively improves the reliability of risk assessment. Modeling multi-source parameters at multiple times quantifies risk factors such as weather fluctuations, environmental changes, abnormal movements, and physiological fluctuations into risk scores, enabling refined and dynamic risk identification. It can also continuously analyze data within a target time period. The risk score is integrated over time to calculate a comprehensive risk value, which can effectively avoid single misjudgments caused by instantaneous data fluctuations and occasional anomalies, weaken the interference of short-term invalid data, and ensure the continuity and objectivity of risk assessment. The comprehensive risk score and time period score threshold are compared to determine potential safety risks and trigger human-computer interaction verification simultaneously. Only when no user safety confirmation operation is detected within a preset time period will a distress message be automatically sent. This can effectively distinguish between accidental falls without injury, normal movement fluctuations, and real danger scenarios, and significantly reduce the probability of false distress triggers. In this way, a distress message will only be automatically sent when the user is in danger and unable to perform a safety confirmation operation, thereby accurately sending out automatic rescue and avoiding the waste of rescue resources. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of the steps of an outdoor rescue method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the structure of a wearable device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an outdoor rescue device provided in one embodiment of this application; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0013] Wearable devices typically have multiple built-in sensors and wireless communication capabilities, enabling them to record users' movement trajectories, speeds, altitudes, and various physical indicators in real time, providing data support and a degree of safety for outdoor enthusiasts. Some wearable devices also possess basic environmental awareness capabilities, capable of receiving weather warnings or monitoring ambient temperature. However, in the complex and ever-changing wilderness environment, relying solely on existing functions is still insufficient for timely detection and response to sudden dangers.

[0014] The safety monitoring functions of wearable devices are mostly limited to single anomaly detection. For example, fall detection functions analyze the user's fall state using gyroscopes and accelerometers. When a severe fall is detected and the user remains still for about 60 seconds, it will automatically dial emergency services and send a distress message containing the user's location. Some sports watches have accident detection functions that can send text messages or emails to preset emergency contacts via a connected mobile phone when a severe impact is detected during activities such as cycling. However, these functions mainly focus on kinematic anomalies such as falls and collisions, and do not fully consider the impact of environmental factors. If a user does not fall significantly but encounters danger due to sudden environmental changes such as a blizzard, traditional fall detection functions may not trigger an alarm in time.

[0015] One of the core limitations of wearable devices is the lack of integration between environmental warnings and user status. While some high-end sports watches offer severe weather warnings and alerts for changes in air pressure and temperature, these warnings are often independent, requiring users to assess the situation and take appropriate action. If a user loses mobility or consciousness due to environmental factors, the device lacks a mechanism to combine environmental hazards with the user's status, potentially missing crucial rescue opportunities. Furthermore, emergency response capabilities in unresponsive situations are insufficient. When users are in danger and unable to call for help independently, most devices, aside from fall and accident detection functions, lack a dedicated "long-term inactivity / no-movement alarm" function. Mobile terminals in the elderly care field have proposed "long-term inactivity alarms," ​​meaning that when a phone is inactive for an extended period, it should automatically send a notification to emergency contacts. However, this mechanism is not yet applied in outdoor sports scenarios. If an outdoor enthusiast goes out alone and remains unreachable or stationary for an extended period, related products may struggle to automatically notify others, potentially delaying rescue efforts.

[0016] Wearable devices have made several improvements to fall detection functionality, such as extending it to head-mounted devices and combining sensor data with audio monitoring to more systematically identify fall events. The core principle is to detect a suspected fall and then determine whether the user is moving in stages. If there is no movement within a certain period of time, an emergency communication is triggered. This solution can solve the problem of automatic SOS when a user falls and loses consciousness. However, it only addresses abnormal personal physiological states and cannot cope with external environmental threats. It cannot be effective when a user encounters danger due to non-fall-related environmental or terrain hazards such as slipping, getting lost, etc.

[0017] The "accident detection" function of related wearable devices is mainly aimed at collision accidents during running and cycling. It relies on the threshold judgment of motion sensors (such as sudden changes in acceleration) to infer the occurrence of an accident, and then sends a preset distress message through the mobile phone. This function is practical in scenarios such as cycling falls, but it is not sensitive to slow-onset environmental crises such as hypothermia. Moreover, if the mobile phone signal is poor or the user does not carry a mobile phone, the alarm message may fail to be sent. At the same time, this function lacks linkage with weather warning data and cannot adjust its sensitivity or decision logic according to environmental risks.

[0018] Standalone environmental monitoring instruments (such as portable lightning detectors, barometers, etc.) and mobile applications can provide weather alerts, but these devices and applications are usually not integrated with personal status and do not have automatic distress call functions; while high-end sports watches can improve safety with features such as barometric altitude warnings and route deviation reminders, these reminders cannot automatically escalate into distress calls when the user is in danger and unable to respond.

[0019] In summary, the relevant technologies can only address some outdoor safety issues in a fragmented way, possessing single functions such as fall detection, weather warnings, route navigation, and location tracking. However, they lack an integrated solution and cannot fuse multi-source information to determine potential dangers encountered by outdoor enthusiasts and automatically trigger distress calls. Especially in scenarios such as unaccompanied long-distance hiking and trail running, when multiple adverse factors such as severe weather, physical exhaustion, and remote locations combine, wearable devices struggle to achieve intelligent identification and timely response to emergencies.

[0020] To solve or partially solve the above-mentioned technical problems, in one embodiment, such as Figure 1 As shown, an outdoor distress signaling method is provided. Although the logical sequence of steps is illustrated in the schematic diagram, in some cases, the steps shown or described may be performed in a different order than those shown in the accompanying drawings. Specifically, this outdoor distress signaling method can be applied to wearable devices, which may include, but are not limited to, one or more of smart bracelets, smartwatches, sports armbands, smart glasses, sports headphones, smart helmets, chest detectors, and smart backpacks.

[0021] Figure 2 This is a schematic diagram of the structure of a wearable device provided in an embodiment of this application, as shown below. Figure 2 As shown, the wearable device may include: an environmental sensor, an inertial sensor, a physiological sensor, a positioning module, a processor and memory, a communication module, and a user interaction module.

[0022] Environmental sensors can include, but are not limited to: ambient light sensors (ALS), temperature sensors, humidity sensors, and / or barometers. Ambient light sensors detect ambient light levels to determine daytime, nighttime, and changes in lighting conditions. Ambient light levels directly affect outdoor visibility and judgment; low-light environments increase the risk of bumps, getting lost, and sudden dangers. Temperature sensors detect ambient temperature; abnormally high or low temperatures can cause hypothermia, heatstroke, and other physiological problems, increasing safety hazards during outdoor activities. Humidity sensors detect ambient humidity; excessively high or low humidity can affect comfort and thermoregulation, easily inducing discomfort and increasing the probability of outdoor accidents. Barometers detect atmospheric pressure to calculate altitude; drastic fluctuations in air pressure may indicate sudden weather changes, and sudden changes in altitude can trigger altitude sickness, exhaustion, and significantly increase the risk of accidents during outdoor activities.

[0023] An inertial sensor, consisting of an IMU unit composed of a three-axis accelerometer and a gyroscope, can collect users' motion data in real time, monitor changes in users' cadence and limb posture, as well as impact events such as falls and collisions, providing motion data support for accurate detection of outdoor accidents.

[0024] Physiological sensors may include, but are not limited to, heart rate sensors, blood oxygen sensors, and / or pressure sensors. Heart rate sensors collect real-time heart rate data, which directly reflects exercise load and fatigue levels, quickly identifying dangerous physiological states such as abnormally high or low heart rates. Blood oxygen sensors collect blood oxygen saturation data, enabling timely detection of hypoxia, altitude sickness, and other issues, providing early warnings of abnormal bodily functions. Pressure sensors detect external forces such as pressure and impacts, aiding in the assessment of falls, bumps, and other accidents. The data collected by these physiological sensors work together to comprehensively characterize a user's health status, providing physiological evidence for assessing risks of outdoor fatigue, discomfort, and accidental injuries, thus improving the comprehensiveness and accuracy of outdoor danger detection.

[0025] The positioning module can perform geographic location positioning based on the Global Positioning System (GPS) and / or the Global Navigation Satellite System (GNSS) to obtain information such as the user's real-time location and movement speed.

[0026] The communication module can obtain local weather forecasts and weather warnings from relevant platforms (such as weather forecasting platforms). Weather parameters reflect external environmental conditions such as temperature, humidity, air pressure, and meteorological changes. Severe weather and drastic weather fluctuations can easily lead to safety hazards such as heatstroke, hypothermia, hypoxia, and extreme weather disasters, serving as an important basis for assessing outdoor environmental risks and predicting the probability of potential danger. The communication module can also send distress signals to external entities (such as rescue organizations or pre-set contacts) when necessary. Furthermore, the communication module can connect to the user's mobile phone via Bluetooth technology or to other user terminals (such as other users' wearable devices).

[0027] User interaction modules may include, but are not limited to: screens, indicator lights, vibration motors, buttons, buzzers, and / or speakers. Screens and indicator lights can be used to display notifications, vibration motors can be used for tactile alerts, buttons and touchscreens can be used for user input confirmation, and buzzers and speakers can be used for audible alarms.

[0028] The processor can coordinate and process multi-source data, completing parameter modeling, risk calculation, logical judgment, and instruction scheduling, and realizing the computational control for risk assessment, early warning, and distress triggering. The memory can be used to store acquired environmental, motion, physiological, and other data, as well as phased risk data, ensuring continuous recording of time-series data and providing data storage support for integral calculations and risk tracing.

[0029] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the priority of the embodiments.

[0030] according to Figure 1 The outdoor distress signaling method shown includes at least steps S110 to S150, which are described in detail below: In step S110, multi-source parameters are acquired in real time.

[0031] Multi-source parameters are critical parameters related to user safety in outdoor scenarios. They provide data support for subsequent risk assessments, modeling calculations, and other tasks, avoiding misjudgments or omissions due to single data sources. Multi-source parameters can include, but are not limited to, weather parameters, environmental parameters, motion parameters, and / or physiological parameters. Wearable devices can collect corresponding multi-source parameters in real time using various sensors, and the communication module of the wearable device can obtain the corresponding multi-source parameters from relevant data sources.

[0032] Weather parameters can be obtained by acquiring the current location through the positioning module, and then by obtaining local weather parameters (such as temperature, whether it is raining, whether it is in a red rainstorm warning area, a yellow thunderstorm warning area, etc.) through the Internet. As an auxiliary measure, or when the device has no network, information such as air pressure, temperature and / or humidity can be collected by environmental sensors, and then weather parameters can be inferred based on air pressure, temperature and / or humidity, such as whether there is a sudden drop in air pressure or temperature in a short period of time.

[0033] Environmental parameters can be collected in real time by environmental sensors, such as ambient light intensity, ambient humidity, and / or air pressure. Ambient light intensity directly affects the user's visibility. Low-light environments (such as dusk, night, and dense forests) can easily cause users to bump into things, get lost, and increase the risk of accidental falls.

[0034] Motion parameters reflect a user's motion state and changes in movement. These parameters can include stillness duration, cadence, stride length, movement speed, exercise intensity, changes in limb posture, and impact events (such as falls and collisions). Motion parameters can be collected through inertial sensors built into wearable devices. Cadence, stride length, and movement speed reflect a user's exercise load; a sudden drop in cadence or abrupt changes in speed may indicate exhaustion or an accidental fall. Data on changes in limb posture can accurately identify abnormal movements, such as excessive body tilt angles or sudden changes in movement amplitude, providing direct evidence for judging falls and collisions. Impact event data can quickly capture sudden collisions and falls, rapidly identifying potential physical injuries to the user.

[0035] Physiological parameters directly reflect a user's physical condition and may include, but are not limited to, heart rate, blood oxygen saturation, and / or body pressure intensity. These parameters can be collected by integrated physiological sensors in the device. Abnormally high heart rate (far exceeding the normal exercise heart rate range), sudden drop in heart rate, or excessive heart rate fluctuations may indicate dangerous conditions such as physical exhaustion or cardiac discomfort. Blood oxygen saturation data can help detect issues like hypoxia and altitude sickness, especially in high-altitude outdoor settings where low blood oxygen levels can directly threaten a user's life. Pressure sensors can detect external pressure or impacts on the body, helping to assess injuries caused by falls, bumps, and other accidents, further improving the accuracy of risk detection.

[0036] Each acquired multi-source parameter can have a corresponding timestamp. The acquisition frequency of each multi-source parameter can be dynamically adjusted according to scenario requirements and data type; for example, weather parameters can be acquired every half hour, and motion parameters can be acquired every second. The acquired multi-source parameters are preprocessed to filter out instantaneous interference data during sensor acquisition, ensuring the accuracy and validity of each set of parameters.

[0037] In step S120, the multi-source parameters at multiple time points are modeled to obtain risk scores at multiple time points.

[0038] Since multi-source parameters cover multiple dimensions and different parameters have different weights in their impact on outdoor safety, it is possible to model each multi-source parameter to quantify the risk level corresponding to each multi-source parameter.

[0039] The modeling process can begin by standardizing the multi-source parameters, converting parameters of different units and magnitudes (such as air temperature, heart rate, and cadence) into standardized data of the same magnitude, thus avoiding modeling bias caused by differences in parameter magnitudes.

[0040] It can be based on preset rules to model multi-source parameters. For example, different value ranges can be set for each type of parameter, and each value range can be mapped to a risk score. Thus, the risk score corresponding to the parameter can be determined according to the value range corresponding to each type of parameter.

[0041] This can be modeled based on multiple parameters. For example, the model sets a safe threshold range for each type of parameter. When a parameter exceeds the safe range, a corresponding risk score is assigned based on the degree of exceedance. For instance, a heart rate exceeding the normal exercise range by more than 50 beats per minute is assigned a high-risk score; a temperature exceeding the safe range by more than 10°C is assigned a medium-risk score; and light intensity below the safe threshold is assigned a low-risk score. Simultaneously, the model can consider the correlation between parameters. For example, the combination of low temperature and high humidity will result in an additive risk score because the risk of hypothermia in this environment is much higher than in a single low temperature or high humidity environment. The combination of abnormally high heart rate and a sudden drop in cadence will be judged as high-risk, as this may indicate physical exhaustion or cardiac discomfort in the user.

[0042] The risk score for a given moment is obtained by weighted summation of the risk scores corresponding to various multi-source parameters at the same time. Based on big data analysis of outdoor safety scenarios and combined with expert experience, reasonable risk weights can be assigned to each type of parameter. For example, physiological parameters (such as heart rate and blood oxygen) are directly related to user safety and therefore have the highest weight; extreme temperature and sudden changes in air pressure in weather parameters, and falls and impacts in exercise parameters, have the next highest weight; auxiliary parameters such as light intensity in the environment have relatively lower weights, ensuring that the modeling results accurately reflect the impact of different risk factors.

[0043] Optionally, the multi-source parameters at each moment can be converted into a standardized risk score of 0-1, with higher scores indicating higher risk. Alternatively, the multi-source parameters at each moment can be converted into a standardized risk score of 0-100. For example, 0-30 points represent low risk (user status is normal, no security risks), 31-60 points represent medium risk (minor security risks exist, requiring attention), and 61-100 points represent high risk (significant risk of encounter exists, requiring timely warning). The risk score at each moment is updated in real time, accurately capturing instantaneous risk changes, achieving refined and dynamic risk identification, and avoiding misjudgments caused by abnormal single parameters or instantaneous fluctuations.

[0044] In step S130, the risk scores corresponding to each of the multiple moments included in the target time period are integrated over time to obtain the comprehensive risk score of the target time period.

[0045] In outdoor scenarios, users' activity levels and environmental conditions can fluctuate momentarily. For example, a user's sudden urge to run can cause momentary abnormalities in physiological parameters, leading to a higher risk score at a single moment. Similarly, a temporary decrease in ambient light due to cloud cover can cause a momentary increase in risk score. However, these momentary fluctuations are not actual safety hazards, and relying solely on risk scores at a single moment can easily lead to false alarms and false distress calls. Therefore, setting a reasonable target time period and integrating risk scores over multiple moments can filter out interference from occasional abnormal data and mitigate the impact of short-term, invalid fluctuations.

[0046] A target time period refers to a pre-defined fixed time length (similar to a time window) used to integrate continuous risk data. The step size is the time interval between two adjacent target time periods, which may overlap. For example, if the target time period is set to 10 minutes with a step size of 1 minute, there can be two target time periods within 11 minutes: minutes 1 to 10 and minutes 2 to 11. The setting of the target time period should be combined with the actual needs of the outdoor sports scenario, taking into account both sensitivity and accuracy. For example, it can be set to 5-10 minutes; it can also be dynamically adjusted according to the intensity of the exercise. For example, when the exercise intensity is high (such as trail running), the target time period can be shortened to 5 minutes to ensure timely capture of sudden risks; when the exercise intensity is low (such as hiking), the target time period can be extended to 10 minutes to reduce the impact of instantaneous fluctuations. The time integral can be calculated by plotting time on the horizontal axis and the risk score at each moment on the vertical axis, calculating the area enclosed by the risk score curve within the target time period and the time axis. This area is the comprehensive risk score for the target time period.

[0047] The calculation of the comprehensive risk score can effectively integrate the continuous risk status within a target time period. For example, if there are 1-2 moments with instantaneous high-risk scores (such as accidental tripping) within the target time period, and the remaining moments are low-risk scores, the comprehensive risk score will fall within the low-risk range after integration, avoiding misjudgment. If multiple moments within the target time period consistently show medium-to-high risk scores (such as persistently abnormal heart rate or persistently adverse environmental conditions), the comprehensive risk score will increase significantly, thus accurately reflecting the user's ongoing safety hazards. In addition, the comprehensive risk score can be adjusted based on the risk change trend within the target time period. If the risk score shows an upward trend (such as a gradually increasing heart rate or a continuous decrease in air pressure), the comprehensive risk score will be appropriately increased to warn of potential escalating risks; if the risk score shows a downward trend (such as the user returning to normal exercise or improved environmental conditions), the comprehensive risk score will be appropriately decreased to avoid over-warning.

[0048] The comprehensive risk score obtained by integrating time can comprehensively and objectively reflect the overall safety status of users within the target time period. It avoids the interference of instantaneous fluctuations and can capture continuous risks and hidden dangers, thus improving the accuracy of outdoor risk detection.

[0049] In step S140, when the overall risk score for the target time period is higher than the time period score threshold, a reminder is issued to prompt the target user to perform a security confirmation operation.

[0050] The comprehensive risk score for the target time period is compared with a preset time period score threshold. When the comprehensive risk score is higher than the time period score threshold, an alert can be triggered by vibration, ringtone, or text message. This alert guides the user to confirm safety, thereby building a dual early warning mechanism of machine judgment and manual verification, effectively reducing the probability of false alarms and false distress calls.

[0051] The setting of time-period score thresholds needs to be combined with big data analysis of outdoor safety scenarios and user needs, and adopt a hierarchical threshold design, divided into low threshold, medium threshold and high threshold, each corresponding to different warning levels, to ensure the targeting and flexibility of warnings. The low threshold corresponds to a comprehensive risk score of 60 points, the medium threshold corresponds to 75 points, and the high threshold corresponds to 90 points. When the comprehensive risk score is higher than the low threshold (60 points) and lower than the medium threshold (75 points), a mild reminder (such as a vibration reminder) is issued to prompt the user to pay attention to their own status and changes in the environment; when the comprehensive risk score is higher than the medium threshold (75 points) and lower than the high threshold (90 points), a moderate reminder (such as a vibration + voice reminder) is issued to force the user to confirm safety; when the comprehensive risk score is higher than the high threshold (90 points), a severe reminder (such as continuous vibration + high-frequency voice reminder) is issued, while shortening the subsequent preset confirmation time to ensure rapid response to potential emergencies.

[0052] The alert system is designed to take into account the unique characteristics of outdoor scenarios, employing a multimodal alert mode to prevent users from ignoring a single alert method. The main alert methods include device vibration, voice announcements, and screen notifications. Voice announcements use clear and concise prompts, such as "There is a safety hazard in the current environment; please check your safety" and "Abnormal heart rate; please stop exercising and check your status," ensuring users quickly receive alerts during exercise. Screen notifications display the current overall risk score, abnormal parameters, and safety confirmation instructions, allowing users to easily view detailed information.

[0053] In one embodiment, the security confirmation operation may include simple operations such as pressing a device button, confirming via screen touch, or responding with a voice, to avoid users being unable to confirm in a timely manner due to complex operations.

[0054] In another embodiment, the safety confirmation operation can also be a micro-challenge action used to confirm whether the user is awake and capable of operation. This micro-challenge action does not require complex learning from the user, conforms to the operating habits during outdoor sports, and has clear movement characteristics, facilitating accurate detection by the device. For example, the micro-challenge action could be pressing and holding the side button for 3 seconds, rotating the wrist approximately 90°, or double-tapping the screen and holding it stably. Whether the target user (the wearer of the wearable device) has completed the safety confirmation operation can be detected based on the inertial measurement unit of the wearable device. For example, the inertial measurement unit can capture limb movements to determine whether the user has the ability to exercise self-control.

[0055] The micro-challenge actions effectively distinguish between two scenarios: "the user is conscious and capable" and "the user is unconscious and unable to operate." This adapts to various states that users may be in during outdoor activities, ensuring the reliability of safety confirmation results. The inertial measurement unit can quickly and reliably determine whether the user has completed the safety confirmation operation without relying on complex external equipment, ensuring that the entire confirmation process is efficient and convenient. At the same time, it further improves the accuracy of safety confirmation and provides a reliable basis for subsequent distress triggering.

[0056] Wearable devices can record user confirmation actions. If the user completes the safety confirmation action, it means that the current risk is a misjudgment or can be handled by the user. The wearable device can reset the risk assessment process and continue to collect multi-source parameters for monitoring in real time. If the user does not complete the safety confirmation action, it enters the distress trigger process to ensure that rescue can be initiated in a timely manner when the user is actually in danger.

[0057] In step S150, if the safety confirmation operation is not detected within a preset time period, a distress message is sent.

[0058] If a user fails to complete the safety confirmation operation within a preset time period, it is determined that the user may be in danger and has lost the ability to operate independently. This will automatically trigger the sending of a distress message, enabling timely initiation of emergency rescue while ensuring both the rigor of the early warning and the timeliness of the rescue.

[0059] The preset duration can be set based on the overall risk score and the needs of the outdoor scenario to avoid false distress calls due to too short a duration or rescue delays due to too long a duration. The preset duration can be inversely proportional to the overall risk score; the higher the overall risk score, the shorter the preset duration.

[0060] Within a preset time period, the wearable device will continuously monitor the user's safety confirmation actions. If no confirmation actions are detected, it will determine that the user is in danger and unable to seek help on their own, and will immediately send a distress message.

[0061] In one embodiment, the content of the distress message enables rescuers to quickly obtain critical information, including but not limited to: the target user's real-time coordinates (obtained via a positioning module), the risk type, and a timestamp. This timestamp can be the current timestamp, the timestamp corresponding to the target time period, or the timestamp of an event such as a collision.

[0062] The distress message may also include, but is not limited to: the current overall risk score, abnormal parameters (such as abnormal heart rate, low temperature, etc.), device ID, and preset emergency contact information.

[0063] SOS messages can be sent simultaneously through multiple channels to ensure timely delivery of rescue information. For example, they can be sent to the user's preset emergency contacts (such as family and friends) via SMS, telephone, or push notifications from a dedicated app to inform the emergency contacts of the user's distress situation and real-time location. They can also be sent to outdoor rescue platforms to connect with professional rescue organizations, providing rescuers with accurate location and distress information, thus shortening rescue response time.

[0064] After sending a distress signal, the wearable device can continuously collect parameters from multiple sources, update the user's status and location information in real time, and push the data simultaneously to emergency contacts and the rescue platform. This ensures that rescuers can monitor the user's movements in real time, improving rescue efficiency. Furthermore, the wearable device can activate an emergency mode to reduce power consumption, ensuring continuous operation until rescuers arrive, further guaranteeing the smooth progress of the rescue.

[0065] By using a preset buffer period and sending distress signals through multiple channels, false distress calls caused by user negligence or misoperation are avoided. This ensures that when a user is truly in danger and unable to call for help independently, rescue can be initiated in a timely manner, providing ultimate protection for the safety of outdoor users.

[0066] The technical solution adopted in this application embodiment can acquire multi-source parameters such as weather parameters, environmental parameters, motion parameters, and physiological parameters in real time. Weather and environmental parameters can accurately recreate complex outdoor scene conditions, effectively identifying external risk factors such as severe weather, low temperatures, and low light. Motion parameters can reflect the user's exercise intensity and posture changes, accurately identifying dangerous behavioral characteristics such as falls and sudden cessation of movement. Physiological parameters can monitor abnormal states such as physical exhaustion and sudden discomfort. Multi-dimensional data collaboration can effectively improve the reliability of risk assessment. Modeling multi-source parameters at multiple times quantifies risk factors such as weather fluctuations, environmental changes, abnormal movements, and physiological fluctuations into risk scores, enabling refined and dynamic risk identification. By integrating continuous risk scores over time within a target period to calculate a comprehensive risk value, the system effectively avoids single misjudgments caused by instantaneous data fluctuations and occasional anomalies, mitigates interference from short-term invalid data, and ensures the continuity and objectivity of risk assessment. The system compares the comprehensive risk score with time-period score thresholds to identify potential safety risks and simultaneously triggers human-computer interaction verification. Only when no user safety confirmation action is detected within a preset timeframe will a distress signal be automatically sent. This effectively distinguishes between accidental falls without injury, normal movement fluctuations, and genuine distress scenarios, significantly reducing the probability of false distress triggers. Thus, a distress signal is only automatically sent when a user is in danger and unable to perform a safety confirmation action, ensuring accurate automatic distress calls and avoiding waste of rescue resources.

[0067] In one embodiment, a user wearing a smartwatch is hiking alone in a mountainous area at 14:20 on a given day. The smartwatch activates the outdoor distress signal method described in this solution, which can acquire and process various multi-source parameters in real time. After 14:25, the outdoor weather in the mountainous area deteriorates significantly. The smartwatch's environmental sensors monitor in real time that the temperature drops from 18°C ​​to 14°C within 10 minutes, and the ambient light intensity decreases by approximately 60% within 5 minutes. Simultaneously, the smartwatch's positioning module, combined with map data, calculates that the target user's current location is more than 3 kilometers away from the nearest road, indicating a remote outdoor environment.

[0068] The smartwatch fuses acquired multi-source parameters every 30 seconds according to a preset cycle. These parameters include weather parameters (temperature, temperature change rate), environmental parameters (ambient light intensity, ambient light change rate, user remoteness), and motion parameters (user's motion state). Using a preset multi-dimensional risk modeling algorithm, it calculates a risk score H for each moment. In this embodiment, the risk score H = 0.72. The smartwatch then continuously integrates the risk scores H from multiple consecutive moments according to a preset target time period and time integration algorithm to obtain a comprehensive risk score Dose. This comprehensive risk score Dose characterizes the cumulative degree of comprehensive risk faced by the target user within the target time period.

[0069] Around 2:34 PM, the smartwatch's motion sensor detected that the target user had remained stationary for more than 6 minutes in the middle of the hiking route, and their posture was abnormally prone. At this point, the overall risk score (Dose) had exceeded the preset time-limited score threshold. Based on this, the smartwatch automatically generated a 6-minute "safety confirmation window" and triggered the target user to perform a safety confirmation operation through a combination of strong vibration and a buzzer, to confirm whether the target user was conscious and capable of performing actions. This safety confirmation operation consisted of preset micro-challenge actions, specifically: pressing and holding the side button of the smartwatch for 3 seconds, rotating the wrist approximately 90 degrees, double-tapping the screen, and maintaining a stable posture. The smartwatch used its built-in inertial measurement unit to detect the target user's body movements and actions in real time to determine whether the micro-challenge actions were completed. The detection showed that the target user did not complete any valid micro-challenge actions within the 6-minute safety confirmation window, and the target user also did not perform any valid actions within the second safety confirmation window. Based on this, the smartwatch can determine that the target user may have lost the ability to act independently or be conscious, and immediately generate and sign a "Proof of Presence in Incomplete Consciousness" (PoLT). This proof of presence may include the target user's risk accumulation process, abnormal posture data, and records of two failed security confirmations.

[0070] The smartwatch automatically sends a distress message to the target user's preset emergency contacts through its built-in wireless communication module. The distress message may include the target user's real-time coordinates, event timestamp (around 14:34), a summary of the deteriorating outdoor environment (sudden drop in temperature, decrease in ambient light, user is in a remote mountainous area), and the core statement that "the target user is unable to complete the conscious verification in a high-risk outdoor environment".

[0071] The system then enters continuous monitoring mode, updating the user's location to the preset emergency contact every 2 minutes. If the user resumes walking at 14:39 and successfully completes a strong wakefulness challenge, the system will immediately send a cancellation request message to withdraw the distress call. The entire process requires no user intervention. The system forms a complete closed loop from the continuous accumulation of risk, failure to confirm wakefulness, to automatically initiating a distress call, effectively avoiding the life-threatening risks caused by users being unable to call for help themselves in a coma or severely injured situation.

[0072] Based on the above technical solution, as an embodiment, considering that a single device's judgment of "whether in distress" is prone to false alarms (rest) or missed alarms (device without network / user disabled), teammates or companions can be introduced as on-site witnesses to assist in the distress call. Specifically, this outdoor distress call method may further include: joining a team, which includes the target user and fellow users; when the risk score at the target time is higher than the time score threshold, broadcasting a witness request to the team and obtaining the voting results; the witness request is used to request team members to vote on whether the target user is in distress; when the voting results indicate that the target user is in distress, identifying the alarm user from the team, so that the alarm user's user terminal automatically sends the target user's distress information; the network quality corresponding to the alarm user is higher than the network quality threshold.

[0073] Teams can be formed, including the target user and multiple users traveling together. Devices within the team can interact via short-range wireless links, enabling beacon communication, message broadcasting, and collaborative security information exchange. During team formation, security authentication and communication link establishment are completed to ensure secure interaction and reliable data transmission between team devices, preventing interference from external devices, spoofing, or other malicious actions. Once the team is formed, team members wearing compatible devices (such as smartwatches or smartphones) automatically activate team mode. Devices broadcast beacon messages every 2 seconds, which can include key information such as team identifier, device identifier, real-time signal strength (RSSI), device battery level, and network connection status. Each device receives beacon signals broadcast by teammates in real time, maintaining its local neighbor table and updating teammates' locations, signal strength trends, communication capabilities, and battery status. This provides foundational data for subsequent short-range gating, voting rights determination, and alarm user screening. For example, member B's device can receive 15 beacons broadcast by member A within 30 seconds, recording the RSSI value received each time to form an RSSI sequence, used to determine if its real-time distance to member A meets the short-range requirement.

[0074] During outdoor activities, each device continuously acquires multi-source parameters according to a preset cycle and calculates the risk score at each moment in real time. This score is then compared with a preset moment score threshold to achieve a preliminary determination of suspected distress. When the risk score of the target user at any given moment exceeds the preset moment score threshold, it can be determined that the target user is in suspected distress. A witness request can then be broadcast to the team with established communication connections, simultaneously initiating a collective verification process to avoid the limitations of independent judgment by a single device. Optionally, this witness request is a standardized message used to request team members to vote on whether the target user is actually in distress. The message contains a unique event identifier, the severity level of the event, and preliminary risk data of the target user (such as impact intensity, abnormal posture information, etc.).

[0075] Collect valid votes within the team to determine if the target user is in danger. If the votes indicate that the target user is in danger, select the user from all team members to ultimately report the distress call. The selection process can prioritize devices with network quality exceeding a preset threshold based on the communication link status of each member's terminal. The user's terminal will then automatically send the distress call information to the target user.

[0076] By adopting the technical solution of this application embodiment, a team consisting of the target user and fellow users can be formed without relying on the communication and operational capabilities of the target user's own device. This can effectively solve the problems of single device network anomalies and user disability preventing them from actively seeking help. The consensus verification of teammates on site improves the accuracy of distress judgment, which can significantly reduce the probability of false alarms and shorten the rescue response time in real distress situations.

[0077] Based on the above technical solution, as an example, to ensure the authenticity and validity of the voting results, a proximity gating rule can be set. Only team members who meet the proximity condition have the right to vote; that is, only members who are truly close to the target user and can directly observe the target user's status can participate in the voting. This avoids remote members blindly voting due to a lack of understanding of the situation, further reducing the probability of misjudgment. Specifically, beacon messages can be periodically broadcast to the team to update the team's network topology information and determine the distance to neighboring users of the target user. Based on the network topology information and the distance to neighboring users, valid voting users are determined from the peer users. Broadcasting a witness request to the team and obtaining the voting results can include: broadcasting the witness request to the valid voting users and obtaining the voting results corresponding to the valid voting users. Alarm users can be identified from the valid voting users.

[0078] Once the team is formed, the user terminals (including wearable devices) of the team members periodically broadcast beacon messages to the team. Each device receives and parses the beacon messages in real time, dynamically updates the network topology information of the team, and determines the distance of the target user's neighboring users based on RSSI values.

[0079] The core judgment parameter for proximity gate control is RSSI (Bluetooth signal strength). The preset proximity threshold is RSSI ≥ -70dBm, and this signal strength must be maintained for t_prox = 15 seconds. If an optional UWB (Ultra-Wideband) ranging module is configured, it can also assist in judging the distance ≤ 30 meters and the number of ranging measurements ≥ 3 times. Dual verification ensures that the member is within the proximity range of the target user.

[0080] Based on the updated network topology information, the communication connection status of each member in the team is determined. Combined with the distance data of nearby users, valid voting users are identified from the peer users. Only users who meet the proximity threshold (RSSI≥-70 dBm for 15 seconds) are selected as valid voting users to ensure that the voters are close personnel who can observe the situation in real time, thus ensuring the authenticity and validity of the voting results.

[0081] Specifically, after receiving a witness request, each team member's device automatically retrieves its locally recorded RSSI sequence to verify whether its signal strength and that of the target user meet the proximity gating requirements. For example, if target user A falls, member B's device detects an average RSSI of -63 dBm with A for 18 seconds, meeting the proximity condition and gaining voting rights. Member C's device detects an average RSSI of -68 dBm with A for 16 seconds, also meeting the condition and gaining voting rights. However, member D's device detects an RSSI of approximately -78 dBm with A, failing to meet the proximity threshold and thus lacking voting rights; it can only receive event notifications and cannot participate in the voting. Only members who pass the proximity gating will have their user terminals display a voting UI, offering voting options such as "Danger" and "Safe," with a default 15-second timeout rule. If no manual vote is cast within 15 seconds, the device automatically ignores the voting result for that user terminal, ensuring efficient convergence of the voting process.

[0082] Members with voting rights (such as B and C) combine their observations of the target user's status on-site (e.g., whether they have fallen, responded, or whether the surrounding environment is dangerous) to complete their voting within the voting window. After voting, each member device automatically generates a proximity evidence digest. This digest includes an RSSI histogram within 15 seconds, a timestamp, a device identifier, and a user signature. This digest is used to verify that the member was indeed within close proximity and that the voting operation occurred within the preset window, ensuring the traceability and credibility of the vote. The proximity evidence digest is small enough for rapid transmission. Each member device binds the voting result with the proximity evidence digest and digital signature, generates a witness response, and sends it to all devices in the team. The digital signature is generated using the device's private key to prevent the voting result from being tampered with.

[0083] After all voting responses are received, the system summarizes and statistically analyzes the voting results to determine whether the voting results indicate that the target user is in danger. Voting thresholds can be preset, with a danger threshold t_danger=2, meaning that if two or more votes are "dangerous," the system determines that the target user is in danger; a safety threshold t_safe=2, meaning that if two or more votes are "safe," the system determines that the target user is not in danger and triggers a no-outbound-call downgrade process. For example, if members B and C both vote "dangerous," reaching the danger threshold t_danger=2, the system determines that target user A is indeed in danger and immediately initiates the subsequent SOS process; if only one vote is "dangerous" and the rest are "unreachable," then no immediate SOS is triggered, and the system only continuously monitors the target user's status to avoid false alarms.

[0084] After determining that the voting results indicate that the target user is in distress, it is necessary to identify the alarm user from the team. The alarm user's user terminal will automatically send the target user's distress message. The core screening criterion can be that the network quality of the alarm user is higher than a preset network quality threshold, to ensure that the distress message can be sent to the emergency contact and rescue platform in a timely and stable manner, and to solve problems such as network outages, poor signal, and insufficient power that may exist in the target user's own device, so as to avoid missed reports due to the target user's device being unable to communicate.

[0085] The system can weighted and sum the communication capability score, terminal battery score, and proximity ranking score of each user to determine the score of each team member, and identify the team member with the highest score as the alarm user. For example, if a user terminal has a cellular network connection, its communication capability score is 1; if it only has a Bluetooth connection, its communication capability score is 0.5; and if it has no network connection, its communication capability score is 0. The user terminal's battery score can be obtained by mapping the user terminal's battery level to a value from 0 to 1. The system can rank each team member based on their distance from the target user, and determine their proximity ranking score based on this ranking, where a higher ranking results in a higher proximity ranking score.

[0086] Once the alarm user is identified, their user terminal (such as a smartwatch or linked mobile phone) automatically sends a distress message without user intervention. The distress message may include, but is not limited to: the target user's real-time coordinates (obtained via a positioning module), the risk type, and a timestamp. It may also include a witness certificate as credible evidence. This certificate is automatically generated by the system by aggregating valid votes and includes an event identifier, a judgment result, a hash of the witness signer set, and digital signatures of all valid votes. Aggregated signature technology can be used to further compress the message size, proving the authenticity of the target user's distress situation to emergency contacts and rescue platforms, thus enhancing the credibility of the distress message.

[0087] By adopting the technical solution of this application embodiment, beacon messages are periodically broadcast to the team, which can dynamically monitor the communication connection status, location distribution, and equipment operation of each member's devices. Simultaneously, it accurately calculates the distance between the target user and nearby users, providing precise and real-time data support for the subsequent selection of valid voting users. This avoids invalid votes or misjudgments caused by lagging topology information or distance judgment errors. Determining valid voting users based on network topology information and the distance to nearby users can strictly screen out peers within close range of the target user who can observe the situation in real time, excluding members who are far away and unable to understand the actual situation. This ensures that voters have the qualifications to truly witness, improving the authenticity and reliability of voting results from the source, effectively avoiding false alarms and missed alarms caused by independent judgment by a single device, and guaranteeing the accuracy and timeliness of outdoor collaborative rescue.

[0088] Based on the above technical solution, as an embodiment, the outdoor distress method may further include: acquiring preset rules and preset routes; the preset rules include parameter thresholds; the preset routes include safe and unsafe areas; the safe areas include a starting area, a rest area, and a destination area; determining a first risk probability corresponding to each of the multi-source parameters based on the multi-source parameters and their respective parameter thresholds; determining a second risk probability based on the area where the target user is located when the static duration reaches a preset duration; the second risk probability corresponding to the target user in the unsafe area is greater than the second risk probability corresponding to the target user in the safe area; summing the first risk probability and the second risk probability to obtain a risk probability; issuing a reminder to the target user to perform a safety confirmation operation when the risk probability exceeds a probability threshold; and sending a distress message if no safety confirmation operation is detected within a preset duration.

[0089] Preset rules can include parameter thresholds for various multi-source parameters and risk level classification standards. Preset routes can clearly mark the target user's preset travel trajectory and the areas traversed, while also dividing the area into safe and unsafe zones. Safe zones can include the starting area, rest area, and destination area, while unsafe zones can be dangerous areas such as complex terrain in the wild or areas without signal coverage. After obtaining the target user's multi-source parameters, a risk assessment is performed on each multi-source parameter based on the parameter thresholds of various multi-source parameters in the preset rules to determine the primary risk probability corresponding to each parameter.

[0090] If a target user remains stationary in a certain area for a preset duration, the area type (safe zone / unsafe zone) can be determined, and the probability of a second risk can be calculated based on the area type. If the target user remains stationary in a safe zone for an extended period, they may be resting, and the probability of a second risk is relatively low. Conversely, if the target user remains stationary in an unsafe zone for an extended period, the probability of a second risk is higher. The probability of a second risk can be determined based on the duration of stillness and the area type in which they remain stationary; the longer the stillness, the higher the probability of a second risk. The probability of a second risk is higher when stationary in an unsafe area than when stationary in a safe area.

[0091] The first risk probability and the second risk probability can be weighted and summed to obtain the risk probability. The weights of the first risk probability and the second risk probability can be set according to actual needs. When the risk probability exceeds a preset probability threshold, an alert can be issued to prompt the target user to perform a security confirmation operation; if no security confirmation operation is detected within a preset time period, a distress message is sent. The specific methods for issuing the alert to prompt the target user to perform a security confirmation operation and sending a distress message if no security confirmation operation is detected within the preset time period can be referred to the above.

[0092] In one embodiment, the preset rule can stipulate that when a preset combination of environmental hazard conditions and user abnormal conditions is met, it is determined that there is a dangerous situation (the risk probability exceeds the probability threshold).

[0093] Optionally, the wearable device can obtain local weather warning information corresponding to the current location of the target user through the communication module. When the weather warning level reaches a preset threshold, it can be determined that the environmental danger conditions are met. The warning level threshold can be a red warning or an orange warning, such as a red warning for heavy rain or an orange warning for thunderstorms. Such warnings indicate that there are significant safety hazards in the outdoor environment.

[0094] Optionally, the wearable device can collect environmental parameters in real time through environmental sensors (ambient light sensor, temperature sensor). When it detects both a sudden decrease in light intensity and a sudden drop in temperature, it can be determined that the environmental emergency conditions are met. The conditions for a sudden decrease in light intensity can be: during non-sunset periods, the ambient light intensity decreases by ≥50% within 5 minutes; the conditions for a sudden drop in temperature can be: the ambient temperature decreases by ≥4℃ within 10 minutes. Such abnormal signals usually indicate the approach of severe weather such as storms or frontal processes.

[0095] Optionally, the wearable device combines the safe and unsafe areas of the preset route and uses the positioning module to locate the target user's position in real time. When it detects that the target user is continuously stationary in the unsafe area and the stationary time reaches the preset time (such as 5 minutes), it is determined that the user's abnormal condition is met. The unsafe area is an area with complex terrain, weak signal, and high rescue difficulty in the preset route. The safe area includes the starting area, the rest area and the ending area.

[0096] Optionally, the wearable device detects the target user's motion state through an inertial sensor. When it detects a fall or abnormal impact event, it determines that the user's abnormal condition is met. The criteria for determining a fall / abnormal impact are: a peak acceleration ≥3.0g and an angular velocity ≥300° / s are detected. Such signals indicate that the user may have suffered physical trauma.

[0097] Optionally, wearable devices can monitor the vital signs of the target user through physiological sensors (such as heart rate sensors). When a continuous low heart rate (such as heart rate < 40 beats / minute) or the disappearance of the heart rate signal is detected, it is determined that the user's abnormal condition is met, further improving the comprehensiveness of the risk assessment.

[0098] When both environmental hazards and user anomalies are met simultaneously, the situation is classified as highly dangerous. For example, if a wearable device receives an orange lightning warning (meeting environmental hazards) and simultaneously detects that the target user has been stationary in an unsafe area for more than 5 minutes without any interactive response (meeting user anomalies), the system will directly classify it as a highly dangerous situation and prioritize triggering the subsequent emergency response process.

[0099] In the context of environmental hazards, "remaining still in an unsafe area for an extended period of time" will be considered a key abnormal signal and can be used as a separate strong judgment condition. That is, when the target user is in an unsafe area and the duration of stillness reaches a preset threshold (such as 6 minutes), and environmental hazards are detected (such as a sudden dimming of light), even if no fall or abnormal vital signs are detected, the system will determine it as a suspected distress situation.

[0100] Through the above combined logic, when it is determined that the target user is in a possible danger situation, the subsequent tiered triggering strategy is immediately implemented, and the first-level prompt (triggering the user's safety confirmation operation) and the second-level action (sending a distress message) are executed in sequence to ensure that rescue can be initiated in a timely manner when the user is actually in danger, while avoiding unnecessary rescue caused by false alarms.

[0101] By adopting the technical solution of this application embodiment, and combining preset rules with preset routes that divide safe and unsafe areas, a first risk probability can be quantified based on multi-source parameters and corresponding thresholds, and a second risk probability can be determined based on the user's static duration and the type of area they are in. The risk weight of unsafe areas is automatically increased. Through the fusion and summarization of the two-dimensional risk probabilities, the comprehensive risk is accurately quantified, and the danger is judged by combining the combination of environmental and abnormal user conditions. The safety confirmation and automatic rescue process are triggered in a graded manner. This effectively reduces the probability of misjudgment by a single indicator, distinguishes between regular rest and real danger in the wild, and improves the accuracy of outdoor risk identification and the rationality of rescue response.

[0102] Based on the above technical solution, as an example, machine learning can be used to intelligently determine whether a user is in danger. Time-series features can be generated based on the multi-source parameters acquired in real time; these features include: velocity change curves, acceleration impact waveforms, environmental parameter change trends, weather warning levels, and / or poster height; the time-series features are input into a risk prediction model to obtain a risk probability; when the risk probability exceeds a probability threshold, an alert is issued to prompt the target user to perform a safety confirmation operation; if no safety confirmation operation is detected within a preset time period, a distress message is sent.

[0103] It can deeply mine the continuous variation patterns of data based on various multi-source parameters, and then generate standardized time series features. Time series features may include, but are not limited to: velocity change curves, acceleration impact waveforms collected by inertial sensors, continuous variation trends of environmental parameters such as temperature, humidity and light intensity, real-time synchronized weather warning levels, and dynamic fluctuation data of altitude. This time series feature can reconstruct the entire process of user movement, limb movement status, evolution of the surrounding environment and changes in terrain elevation. Unlike the traditional method of determining discrete parameters at a single moment, it can effectively capture hidden danger signals such as short-term sudden changes, slow deterioration, and trend anomalies, and comprehensively reflect the overall safety status in outdoor scenarios.

[0104] After standardizing and processing time-series features, they are input into a pre-trained risk prediction model. Leveraging the model's powerful feature extraction and correlation analysis capabilities, the model performs fusion calculations, correlation assessments, and risk projections on multiple types of time-series features, ultimately outputting quantifiable and comparable real-time risk probabilities. This risk prediction model can autonomously learn the evolution patterns of hazards in different outdoor scenarios, accurately identifying various potential dangers such as environmental degradation, abnormal behavior, physical exhaustion, and terrain hazards. It overcomes the limitations of traditional fixed threshold rules, improving the intelligence and generalization ability of risk identification in complex outdoor environments.

[0105] When the real-time risk probability output by the model exceeds the system's preset probability threshold, an active early warning mechanism can be triggered. This mechanism issues safety reminders through multimodal methods such as vibration, beeping, and screen pop-ups, guiding the target user to complete a safety confirmation operation in a timely manner. This verifies whether the user maintains a clear mind and the ability to act independently, forming a two-way human-machine verification mechanism to reduce false triggering issues.

[0106] If no valid confirmation operation or interaction is detected from the user within the preset fixed confirmation time range, it is determined that the user has most likely lost the ability to act or be conscious and is unable to deal with danger on their own. The system will automatically initiate the emergency response process, proactively send out standardized distress information, and promptly contact emergency contacts to carry out rescue and support.

[0107] The intelligent algorithm path, employing the technical solution of this application embodiment, can adapt to complex situations and reduce false alarms and missed alarms. For example, it may determine that a user is not resting normally but experiencing a sudden accident by identifying a combination of abnormal acceleration before the movement stops and a sudden change in the environment. This risk prediction model can execute a simple model locally on the device, or upload data to the cloud via a mobile phone for analysis and then issue a decision.

[0108] In one embodiment, a trail runner is running a long distance in mountainous terrain. A smartwatch, in "Trail Running Mode," continuously monitors the area. Around the midpoint of the 50km mark (approximately 20km), the watch receives an orange lightning warning (indicating a severe thunderstorm) via its mobile network. Simultaneously, the sky darkens rapidly, and the ambient light sensor shows a 60% drop in light intensity and a 4°C drop in temperature within 5 minutes. The runner slows down and eventually stops (possibly to avoid the weather or due to exhaustion). The watch detects that the runner has been stationary for over 7 minutes outside of aid stations, and based on the combined environmental and physical condition, determines the situation to be dangerous: approaching severe weather + prolonged stationary position midway through the route. The system immediately issues a Level 1 alert: the device vibrates and sounds an alarm, and the screen displays "You may be in danger. Are you safe? We will automatically send a distress signal." Because the runner may have lost mobility, they fail to respond within 60 seconds. The watch then initiated a Level 2 emergency call: It sent an SOS text message via cellular network to its pre-set emergency contacts (such as family members or the race's rescue team): "Suspected distress: Trail runner [Name], location <longitude, latitude>, may be incapacitated due to thunderstorms. Please contact or rescue him immediately." The watch then attempted to dial the local emergency number, stopping when no one answered or the call could not be connected (this scenario assumes sending the message to a contact is sufficient). Upon receiving the message, the emergency contact immediately tried to contact the athlete by phone without success, and then sent the location to a nearby rescue team. Fortunately, rescuers quickly set off and found the athlete, unconscious due to hypothermia, near the location and successfully provided medical assistance.

[0109] In one embodiment, several hikers agreed to trek through a remote valley. One hiker, Xiao Zhang, was equipped with the smartwatch of this invention. During the trek, the group stretched out, with Xiao Zhang lagging behind. The watch operated offline, relying primarily on its built-in sensors. In the afternoon, fog suddenly rolled in and heavy rain began (due to the lack of network signal, the watch detected severe weather signals through rapid pressure drops and dimming light). Xiao Zhang then slipped and fell down a slope. Although he did not sustain serious injuries, he was unable to climb back up immediately. His location was concealed, and his teammates were unaware that he had fallen behind. The watch's accelerometer detected the violent fall and indicated that Xiao Zhang remained largely still afterward. Given that they were in the middle of the route, the weather was bad, and he had fallen, the system immediately triggered a level one alert. Xiao Zhang was conscious but startled and did not immediately operate the watch. About one minute later, the watch automatically sent an SOS. Due to its offline status, the watch could not immediately notify others, but in SOS mode, it periodically attempted to establish any available connection. As Xiao Zhang slowly moved to a slightly higher position, his watch detected a weak cell phone signal and immediately sent a distress message via SMS. One of the other hikers in the group received the message (Xiao Zhang had previously set his teammate as an emergency contact): "Xiao Zhang appears to be in distress, last location coordinates XXX, time HH:MM, please check immediately." The teammates immediately returned, located Xiao Zhang according to the coordinates, rescued him, and safely evacuated him. This embodiment demonstrates the characteristic of outdoor distress signals that continue to attempt to send out distress signals even in the absence of a network, and the advantage of combining fall detection with environmental perception to improve detection accuracy. If there were only a standard fall alarm without environmental and stationary assessments, Xiao Zhang might have been misjudged as not seriously injured due to slight movement after the fall, thus missing the opportunity for rescue. This embodiment, however, significantly improves the reliability of hazard identification through a multi-factor approach. Ultimately, it successfully facilitated mutual rescue among teammates, reducing accident losses.

[0110] The technical solution adopted in this application can issue early warnings in the early stages of environmental deterioration and abnormal user behavior, reminding users to correct their behavior or avoid risks; once the user is unable to respond, the distress information can be automatically transmitted, minimizing the information gap within the "golden rescue time".

[0111] This is especially important for people who engage in long-distance exercise alone, essentially providing them with a tireless safety monitor. Multi-sensor information fusion and a phased confirmation mechanism reduce the likelihood of false alarms. For example, a user resting midway through the exercise might be identified as stationary, but if the weather is normal and the user actively cancels the notification, a false alarm will not be triggered. This is more intelligent than devices that rely solely on a single sensor (such as alarming simply by detecting stationary movement).

[0112] Compatible with various outdoor activities and different device network conditions, it can send distress messages through various means (cellular network / satellite / companion relay) whether in the countryside with cell phone signal or in the deep mountains without network coverage, thus improving the system's robustness. Its algorithm parameters can also be flexibly adjusted according to hiking, running, mountaineering and other modes to meet the needs of different exercise intensities and rhythms.

[0113] Wearable devices have an automatic alarm function in case of danger, which can improve the sense of security of users and their families. Users can reduce their psychological burden when exploring, and their families can also be informed of the situation in a timely manner through the emergency contact mechanism.

[0114] To facilitate better implementation of the outdoor rescue method of this application, this application also provides an outdoor rescue device based on the above-described outdoor rescue method. The meanings of the terms used are the same as in the above-described outdoor rescue method, and specific implementation details can be found in the description of the method embodiments.

[0115] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of the outdoor rescue device provided in the embodiments of this application, wherein the outdoor rescue device includes: The acquisition module 301 is used to acquire multi-source parameters in real time, including weather parameters, environmental parameters, motion parameters, and physiological parameters. Modeling module 302 is used to model the multi-source parameters at multiple times to obtain risk scores at multiple times; The integration module 303 is used to perform time integration on the risk scores corresponding to each of the multiple moments included in the target time period to obtain the comprehensive risk score of the target time period. The reminder module 304 is used to issue a reminder to the target user to perform a security confirmation operation when the comprehensive risk score of the target time period is higher than the time period score threshold. The distress module 305 is used to send a distress message if the safety confirmation operation is not detected within a preset time period.

[0116] In one embodiment, the device further includes: The "Join" module is used to join a team, which includes the target user and fellow users. The request broadcast module is used to broadcast a witness request to the team and obtain the voting results when the risk score at the target time is higher than the time score threshold; the witness request is used to request the team members to vote on whether the target user is in danger; The user identification module is used to identify an alarm user from the team when the voting result indicates that the target user is in distress, so that the user terminal of the alarm user automatically sends the distress message of the target user; the network quality corresponding to the alarm user is higher than the network quality threshold.

[0117] In one embodiment, the device further includes: The message broadcast module is used to periodically broadcast beacon messages to the team in order to update the team's network topology information and determine the distance to the target user's neighboring users; The effective determination module is used to determine the valid voting users from the peer users based on the network topology information and the distance of the neighboring users; The request broadcast module is specifically used to: broadcast the witness request to the valid voting user and obtain the voting result corresponding to the valid voting user.

[0118] In one embodiment, the device further includes: The route acquisition module is used to acquire preset rules and preset routes; the preset rules include parameter thresholds; the preset routes include safe areas and unsafe areas; the safe areas include a starting area, a rest area, and a destination area. The first probability determination module is used to determine the first risk probability corresponding to each of the multi-source parameters based on the multi-source parameters and the parameter thresholds corresponding to each of the multi-source parameters. The second probability determination module is used to determine a second risk probability based on the area where the target user is located when the static duration reaches a preset duration; the second risk probability of the target user in the unsafe area is greater than the second risk probability of the target user in the safe area. The total probability determination module is used to summarize the first risk probability and the second risk probability to obtain the risk probability; The first confirmation module is used to issue a reminder when the risk probability exceeds a probability threshold, so that the target user can perform a security confirmation operation.

[0119] In one embodiment, the device further includes: The sequence determination module is used to generate time series features based on the multi-source parameters acquired in real time; the time series features include: velocity change curves, acceleration impact waveforms, environmental parameter change trends, weather warning levels, and / or poster heights; The probability prediction module is used to input the time series features into the risk prediction model to obtain the risk probability; The second confirmation module is used to issue a reminder to the target user when the risk probability exceeds the probability threshold, so that the target user can perform a security confirmation operation.

[0120] In one embodiment, the safety verification operation is a micro-challenge action used to verify whether the target user is awake and capable of operation; The device further includes: An operation detection module is used to detect whether the target user has completed the security confirmation operation based on the inertial measurement unit of the wearable device.

[0121] In one embodiment, the distress module 305 is specifically used for: The multi-source parameters are analyzed to determine the risk type; Obtain the coordinate information of the target user; Get the timestamp; The distress message is generated and sent based on the risk type, the coordinate information, and the timestamp.

[0122] The technical solution adopted in this application embodiment can acquire multi-source parameters such as weather parameters, environmental parameters, motion parameters, and physiological parameters in real time. Weather and environmental parameters can accurately recreate complex outdoor scene conditions, effectively identifying external risk factors such as severe weather, low temperatures, and low light. Motion parameters can reflect the user's exercise intensity and posture changes, accurately identifying dangerous behavioral characteristics such as falls and sudden cessation of movement. Physiological parameters can monitor abnormal states such as physical exhaustion and sudden discomfort. Multi-dimensional data collaboration can effectively improve the reliability of risk assessment. Modeling multi-source parameters at multiple times quantifies risk factors such as weather fluctuations, environmental changes, abnormal movements, and physiological fluctuations into risk scores, enabling refined and dynamic risk identification. By integrating continuous risk scores over time within a target period to calculate a comprehensive risk value, the system effectively avoids single misjudgments caused by instantaneous data fluctuations and occasional anomalies, mitigates interference from short-term invalid data, and ensures the continuity and objectivity of risk assessment. The system compares the comprehensive risk score with time-period score thresholds to identify potential safety risks and simultaneously triggers human-computer interaction verification. Only when no user safety confirmation action is detected within a preset timeframe will a distress signal be automatically sent. This effectively distinguishes between accidental falls without injury, normal movement fluctuations, and genuine distress scenarios, significantly reducing the probability of false distress triggers. Thus, a distress signal is only automatically sent when a user is in danger and unable to perform a safety confirmation action, ensuring accurate automatic distress calls and avoiding waste of rescue resources.

[0123] Specific limitations regarding outdoor rescue devices can be found in the above description of outdoor rescue methods, and will not be repeated here. Each module in the aforementioned outdoor rescue device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0124] In addition, this application also provides an electronic device, such as Figure 4As shown, it illustrates the structural diagram of the electronic device involved in this application, specifically: The electronic device may include components such as a processor 401 with one or more processing cores and a memory 402 with one or more computer-readable storage media. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 401 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 402, and by calling data stored in the memory 402, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 401.

[0125] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0126] In one embodiment, the electronic device further includes a power supply 403 that supplies power to the various components. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 403 may also include one or more DC or AC power supplies, recharging systems, power equipment debugging circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0127] In one embodiment, the electronic device may further include an input unit 404, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0128] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 401 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 402 according to the following instructions, and the processor 401 runs the applications stored in the memory 402, thereby implementing the steps in any of the exception handling methods provided in the embodiments of this application.

[0129] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0130] In one embodiment, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the methods described in any embodiment of this application.

[0131] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described in any embodiment of this application.

[0132] In some embodiments, a computer program product is also provided, including a computer program or instructions that, when executed by a processor, implement the methods described in any embodiment of this application.

[0133] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0134] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0135] Therefore, this application provides a computer-readable storage medium storing a computer program that can be loaded by a processor to perform the steps in any of the outdoor rescue methods provided in this application.

[0136] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0137] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0138] Since the instructions stored in the computer-readable storage medium can execute the steps of any of the outdoor rescue methods provided in this application, the beneficial effects that any of the outdoor rescue methods provided in this application can achieve can be realized, as detailed in the preceding embodiments, and will not be repeated here.

[0139] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0140] The above provides a detailed description of an outdoor rescue method, device, electronic device, and computer-readable storage medium provided in this application. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. An outdoor distress signaling method, characterized in that, Applications in wearable devices, including: Real-time acquisition of multi-source parameters, including weather parameters, environmental parameters, motion parameters, and physiological parameters; The multi-source parameters at multiple time points are modeled to obtain risk scores at multiple time points; The risk scores corresponding to each of the multiple moments included in the target time period are integrated over time to obtain the comprehensive risk score for the target time period. When the overall risk score for the target time period is higher than the time period score threshold, a reminder is issued to prompt the target user to perform a security confirmation operation; If the security confirmation operation is not detected within a preset time period, a distress message is sent.

2. The method according to claim 1, characterized in that, The method further includes: Join a team, which includes the target user and fellow users; When the risk score at the target time is higher than the time score threshold, a witness request is broadcast to the team, and the voting results are obtained; the witness request is used to request the team members to vote on whether the target user is in danger; When the voting result indicates that the target user is in distress, an alarm user is identified from the team so that the user terminal of the alarm user automatically sends the distress message of the target user; the network quality corresponding to the alarm user is higher than the network quality threshold.

3. The method according to claim 2, characterized in that, The method further includes: The team periodically broadcasts beacon messages to update the team's network topology information and determine the distance to neighboring users of the target user; Based on the network topology information and the distance to neighboring users, valid voting users are determined from the peer users; The process of broadcasting a witness request to the team and obtaining the voting results includes: The witness request is broadcast to the valid voting users, and the voting results corresponding to the valid voting users are obtained.

4. The method according to claim 1, characterized in that, The method further includes: Obtain preset rules and preset routes; the preset rules include parameter thresholds; the preset routes include safe areas and unsafe areas; the safe areas include a starting area, a rest area, and a destination area. Based on the multi-source parameters and the parameter thresholds corresponding to each of the multi-source parameters, determine the first risk probability corresponding to each of the multi-source parameters; A second risk probability is determined based on the area where the target user is located when the static duration reaches a preset duration; the second risk probability of the target user in the unsafe area is greater than the second risk probability of the target user in the safe area. Summarize the first risk probability and the second risk probability to obtain the risk probability; When the probability of risk exceeds a probability threshold, a reminder is issued to prompt the target user to perform a security confirmation operation.

5. The method according to claim 1, characterized in that, The method further includes: Based on the multi-source parameters acquired in real time, time series features are generated; the time series features include: velocity change curves, acceleration impact waveforms, environmental parameter change trends, weather warning levels, and / or poster heights; The time series features are input into the risk prediction model to obtain the risk probability; When the probability of risk exceeds a probability threshold, a reminder is issued to prompt the target user to perform a security confirmation operation.

6. The method according to claim 1, characterized in that, The safety confirmation operation is a micro-challenge action used to confirm whether the target user is awake and capable of operating. The method further includes: The inertial measurement unit of the wearable device detects whether the target user has completed the security confirmation operation.

7. The method according to claim 1, characterized in that, Sending a distress message includes: The multi-source parameters are analyzed to determine the risk type; Obtain the coordinate information of the target user; Get the timestamp; The distress message is generated and sent based on the risk type, the coordinate information, and the timestamp.

8. An outdoor distress signaling device, characterized in that, Applications in wearable devices, including: The acquisition module is used to acquire multi-source parameters in real time, including weather parameters, environmental parameters, motion parameters, and physiological parameters. The modeling module is used to model the multi-source parameters at multiple time points to obtain risk scores at multiple time points; The integration module is used to integrate the risk scores corresponding to multiple moments within the target time period over time to obtain the comprehensive risk score for the target time period. The reminder module is used to issue a reminder to the target user when the overall risk score for the target time period is higher than the time period score threshold, so that the target user can perform a security confirmation operation. The distress module is used to send a distress message if the security confirmation operation is not detected within a preset time period.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the outdoor rescue method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the outdoor distress method as described in any one of claims 1 to 7.

Images