A method and system for visual tracking of close contact of vessels

Abstract

The application provides a ship close contact visual tracking method and a tracking system. The method provided by the application is applied to the tracking system, the tracking system comprises a plurality of beacons arranged at specific positions of a ship, a server arranged on the ship, and a tracker held by each person on the ship, the method realizes accurate positioning and dynamic trajectory tracking of a high-risk target tracker under the premise of protecting user privacy through cooperation of the beacons and the tracker, in combination with anonymous identifier technology and real-time signal strength analysis; in addition, through efficient aggregation and space-time data analysis of the server on a contact event, a visual interface superimposed on a ship planar map is generated, a target activity path, close contacts and contact duration are intuitively displayed, the tracking efficiency of an epidemic transmission chain in a ship closed environment and the prevention and control response capability are significantly improved, and the security of user privacy data is improved.

Description

Technical Field

[0001] This application relates to the field of wireless positioning technology, and in particular to a visual tracking method and system for close contact between ships. Background Technology

[0002] Ships, being enclosed spaces with dense populations and frequent contact, especially large vessels like cruise ships, pose a significant risk of infectious disease transmission due to the prolonged shared living and working conditions of crew and passengers. During long voyages at sea, ships are typically far from shore-based medical resources, making them highly vulnerable to rapid spread of outbreaks and serious threats to the health and safety of crew and passengers. Therefore, real-time monitoring and tracing of contact among passengers on board to prevent virus transmission and implement timely countermeasures has become a pressing challenge.

[0003] Traditional contact tracing methods rely heavily on manual recording and periodic inspections, lacking real-time accuracy and being susceptible to human oversight. They fail to effectively achieve efficient, accurate, and privacy-protected close contact tracing. Therefore, developing an intelligent tracking method that can track contact in real time, protect privacy, and facilitate timely responses to contact risks by ship management personnel is crucial for improving ship safety management and optimizing health risk response. Summary of the Invention

[0004] In view of this, this application provides a visual tracking method and system for close contact on ships, which can accurately track close contact events while protecting personal privacy.

[0005] Specifically, this application is implemented through the following technical solution:

[0006] The first aspect of this application provides a visual tracking method for close contact on ships. The method is applied to a tracking system, which includes multiple beacons located at specific locations on the ship, a server mounted on the ship, and a tracker held by each passenger on board. The method includes:

[0007] The target tracker in a high-risk state among all trackers periodically broadcasts a first Bluetooth signal to notify the other trackers besides the target tracker of its own status information via the first Bluetooth signal; the first Bluetooth signal carries the anonymous identifier of the target tracker; the anonymous identifier is an identifier obtained by anonymizing the device identifier of the target tracker.

[0008] When the other tracker receives the first Bluetooth signal, and determines that it is in close contact with the target tracker based on the first Bluetooth signal, it records and saves the contact event. The contact event includes the target tracker's anonymous identifier, the other tracker's own device identifier, a timestamp representing the start point of the contact, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the other tracker's location information during the contact duration. The other tracker's location information is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the ship.

[0009] After successfully establishing a connection with the server, the other trackers will report all saved contact events to the server.

[0010] The server determines the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determines the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker.

[0011] The server overlays the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, which is then displayed to the passengers.

[0012] A second aspect of this application provides a tracking system, the tracking system comprising multiple beacons located at specific locations on a ship, a server installed on the ship, and a tracker held by each passenger on the ship; wherein:

[0013] The target tracker in a high-risk state among all trackers is used to periodically broadcast a first Bluetooth signal to notify the other trackers besides the target tracker of its own status information through the first Bluetooth signal; the first Bluetooth signal carries the anonymous identifier of the target tracker; the anonymous identifier is an identifier after anonymizing the device identifier of the target tracker.

[0014] The other tracker, upon receiving the first Bluetooth signal and determining, based on the first Bluetooth signal, that the other tracker and the target tracker are in close proximity, record and save a contact event; wherein the contact event includes the anonymous identifier of the target tracker, the device identifier of the other tracker, a timestamp representing the start point of the contact, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the location information of the other tracker during the contact duration; the location information of the other tracker is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the vessel;

[0015] The other trackers are also used to report all saved contact events to the server after a successful connection is established with the server;

[0016] The server is configured to determine the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determine the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker.

[0017] The server is also used to overlay the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, and to display the visual tracking interface to passengers.

[0018] The ship close contact visualization tracking method and system provided in this application achieves accurate tracking and visualization of close contact situations involving high-risk individuals while fully protecting the privacy of passengers. By anonymizing the identifier of the target tracker device, it ensures that the individual's real identity is not exposed during the entire data collection and interaction process, effectively preventing privacy leaks. Simultaneously, leveraging Bluetooth signal strength and beacon-assisted positioning technology, the system can accurately identify close contact events between personnel and construct the dynamic activity trajectory of the target tracker based on spatiotemporal information. Through analysis of contact events, the server not only reconstructs the target personnel's movement path within the ship but also accurately identifies their close contacts and contact duration at each trajectory point, ultimately displaying this information in an intuitive and interactive manner on the ship's planar map. This solution achieves highly efficient and accurate contact tracing and risk identification without relying on personal identification data, providing intelligent and low-intrusion technical support for epidemic prevention and control and safety management. Attached Figure Description

[0019] Figure 1 A flowchart of an embodiment of the visual tracking method for close contact between ships provided in this application;

[0020] Figure 2 A schematic diagram of a tracking system shown in an exemplary embodiment of this application;

[0021] Figure 3 A schematic diagram of Embodiment 2 of the visual tracking method for close contact between ships provided in this application;

[0022] Figure 4 A schematic diagram illustrating the principle of a visual tracking interface provided for an exemplary embodiment of this application. Detailed Implementation

[0023] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0024] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0025] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0026] Existing ship personnel tracking and contact recording systems face several major technical challenges:

[0027] Insufficient real-time capability: Traditional methods are unable to capture information about personnel contact in real time and cannot provide timely feedback on contact data, leading to delays in epidemic control measures.

[0028] Privacy protection issues: Directly exposing device identifiers or personally identifiable information during tracking can easily lead to privacy leaks, requiring effective anonymization.

[0029] Accuracy issues: Existing contact recording methods cannot accurately pinpoint the time, location, and duration of contact events, and lack comprehensive analysis across spatial and temporal dimensions.

[0030] Inadequacies in data analysis and presentation: The processing of contact events and data is often fragmented, lacking effective structured reports and visualizations, making it difficult for ship managers to make quick judgments.

[0031] To address the aforementioned technical issues, this application proposes a visual tracking method for close contact between ships. It employs Bluetooth signals and server-based data processing technology to construct a comprehensive, accurate, real-time, and privacy-protected contact tracking system.

[0032] (1) Bluetooth signal broadcasting and contact recording: When there are passengers on board the ship in a high-risk situation, the target tracker will periodically broadcast a first Bluetooth signal to inform other trackers of its anonymous identifier and status information. If a tracker that receives this signal comes into close contact with the target tracker, it will record the contact event.

[0033] (2) Anonymization and privacy protection: The device identifier of the target tracker is generated by anonymization to ensure that the identity information of passengers is not leaked and that sensitive personal data is not exposed during contact tracking.

[0034] (3) Contact event reporting and data processing: After all contact events are saved locally, the data is reported through a connection with the server. The server calculates the activity trajectory of the target tracker based on the received contact events and determines the other trackers closely connected at each trajectory point and their contact duration.

[0035] (4) Visualization and Data Analysis: The server overlays information such as the target tracker's activity trajectory, close contact trackers, and contact duration onto the ship's plan map to generate a visual tracking interface, which is displayed in real time to passengers and ship managers. Through the interface, passengers can clearly understand their contact with other people, and ship managers can perform data statistics, risk analysis, and take timely prevention and control measures.

[0036] The technical solution of this application solves the above-mentioned technical problems and has the following significant effects:

[0037] Real-time and accurate: Through periodic broadcasting of Bluetooth signals and real-time recording of contact events, the system can accurately capture the contact information of every passenger on board the ship, avoiding the delays and omissions of manual recording, and realizing real-time tracking and efficient feedback of contact situations.

[0038] Privacy Protection: Anonymous identifiers are used instead of device identifiers to ensure effective protection of passengers' personal privacy. The system only records necessary contact information, avoiding the risk of identity information leakage.

[0039] Data processing and analysis capabilities: By combining contact event data with activity trajectories, the server can perform multi-dimensional data analysis, including contact duration and close contact statistics, to help ship managers accurately understand the health risk status inside the ship and provide a scientific basis for decision-making.

[0040] Visualized and structured reports: By generating visual tracking interfaces and structured reports, ship managers can intuitively and quickly obtain contact data and risk information, which greatly improves emergency response efficiency and management accuracy, and enhances the safety of ship operations.

[0041] In summary, this technical solution innovatively solves the problems of privacy protection, real-time performance, and accuracy in ship contact tracking by combining Bluetooth signal tracking, anonymization technology, real-time data processing, and visualization, providing a more intelligent and efficient solution for ship management.

[0042] The following specific embodiments are given to illustrate the technical solution of this application in detail.

[0043] Figure 1 This is a flowchart of an embodiment of the visual tracking method for close contact between ships provided in this application. Figure 2 This is a schematic diagram of a tracking system illustrating an exemplary embodiment of this application. Please refer to... Figure 1 and Figure 2 The method provided in this embodiment is applied to a tracking system, which includes multiple beacons set at specific locations on a ship, a server set on the ship, and a tracker held by each passenger on the ship.

[0044] It should be noted that a beacon is a wireless transmitter based on Bluetooth Low Energy technology that can broadcast Bluetooth signals to the surrounding area. In this application, multiple beacon devices are placed at specific locations on the ship, and trackers held by passengers receive Bluetooth signals emitted by the multiple beacon devices. The location of the tracker is indirectly calculated using the signals received by the tracker, thereby locating the passengers.

[0045] It should be noted that multiple beacons are deployed in several main areas within the ship's cabin, and the specific deployment locations are determined according to actual needs, and are not limited in this application. For example, in one embodiment, multiple beacons are deployed in the corridors within the ship's cabin, each beacon approximately 1.7m above the ground, with one beacon placed every 10m along the corridor. In addition, beacons can also be placed in public areas and at room entrances within the cabin. In this case, the beacons are installed on the cabin walls, 1.5 to 2.0m above the ground. This ensures that the signals emitted by the beacons can cover the entire area, while preventing the broadcast signals emitted by the beacons from being blocked by people or furniture.

[0046] Specifically, each beacon can broadcast a Bluetooth signal at a predetermined period (for ease of distinction, the Bluetooth signal emitted by the beacon is referred to as the second Bluetooth signal in this application). The Bluetooth signal has a frequency of 1 Hz, a transmission power of 4 dBm, operates on the 2.4 GHz channel, and is broadcast alternately on BLE channels 37, 38, and 39. This ensures stable signal coverage of the Bluetooth signals emitted by multiple beacons on the ship.

[0047] It should be noted that after the beacons are installed on the ship, the installation positions of multiple beacons need to be recorded. Specifically, for each beacon, the cabin number, floor, and relative coordinates of the beacon are stored on the ship's server and synchronously updated to each tracker. The relative coordinates of each beacon are determined as follows: a spatial coordinate system is established with the bow as the origin, and the position of each beacon in this spatial coordinate system is determined as its relative coordinates (the relative coordinate position of each beacon in the spatial coordinate system established with the bow as the origin should have an error of less than 0.5m compared to its actual position in space).

[0048] Furthermore, the tracking system also includes a tracker held by each passenger on board. Each passenger's tracker receives Bluetooth signals broadcast from multiple beacons to the surrounding area to determine their location. This tracker can be a mobile phone or other terminal (e.g., a wearable device).

[0049] Specifically, when each passenger boards the ship, a tracking application is installed on their mobile phone or other device. Upon first launch, the application displays a privacy statement detailing the data collected and its purpose, and requires the passenger to agree to the statement. After the passenger agrees, the application generates and assigns a 128-bit device identifier to the tracker, uniquely identifying the tracker and the passenger associated with it for subsequent identification. Furthermore, the application retrieves the latest deployment information for multiple beacons from a server via the ship's local area network. If the device is connected to the internet, the deployment information is downloaded and cached locally; if offline, it is pre-installed in the application's default configuration folder. As described above, the deployment information for multiple beacons includes each beacon's identifier, its location coordinates (within ±0.5m of the actual location), and detailed data on its installation height.

[0050] In summary, by installing a tracking application on a passenger's mobile phone or other device, that device can be used as a tracker to monitor the passenger. It should be noted that how to use a tracker to monitor passengers will be explained later and will not be elaborated upon here.

[0051] Furthermore, the tracking system also includes a server installed on the ship, which is connected to the ship's local area network. The server can receive data from the trackers and also allow operators to send information to the trackers through the server.

[0052] Please continue to refer to Figure 1 The following describes the visual tracking method for close contact between ships provided in this application. Specifically, the visual tracking method for close contact between ships provided in this application includes:

[0053] S101. The target tracker in a high-risk state among all trackers periodically broadcasts a first Bluetooth signal to notify the other trackers besides the target tracker of its own status information through the first Bluetooth signal.

[0054] The first Bluetooth signal carries an anonymous identifier of the target tracker; the anonymous identifier is an identifier obtained by anonymizing the device identifier of the target tracker.

[0055] It should be noted that when a passenger on board is diagnosed with an infectious disease or identified as a close contact, the relevant person can log in to the health management system on the server through two-factor authentication and submit encrypted and signed diagnostic information (which includes the passenger's tracker ID, timestamp, and disease code). After the server verifies the relevant person's operating permissions and the integrity of the submitted diagnostic information, the server generates a one-time token and pushes a high-risk status update instruction to the passenger's tracker through an encrypted channel. Furthermore, after receiving the update instruction, the tracker updates its own status to high-risk and records the timestamp of the status update, as well as the duration of the high-risk status initialization based on the disease type.

[0056] In this application, for ease of explanation, the tracker in a high-risk state is referred to as the target tracker. It should be noted that after updating its state, the target tracker will enable an anonymous broadcast mechanism and periodically transmit a first Bluetooth signal to its surroundings. This first Bluetooth signal carries the anonymous identifier of the target tracker, so that when other trackers in the vicinity scan the first Bluetooth signal, they can identify that the first Bluetooth signal comes from a high-risk device.

[0057] It's important to note that each tracker has a unique device identifier. When a tracker transmits a Bluetooth signal, other devices can identify it using this identifier carried within the signal. When broadcasting a Bluetooth signal, the target tracker can create an encryption key based on the device identifier using an encryption algorithm, and then anonymize the identifier by incorporating a current timestamp, generating an anonymous identifier. This anonymous identifier can be changed every 15 minutes using the timestamp, ensuring that the target tracker can be tracked by other trackers without being able to identify its true identity.

[0058] In addition, the Manufacturer Data field of the Bluetooth signal emitted by the target tracker includes string markers such as "INFECTED" or "CLOSE_CONTACT". These string markers are predefined special identification information. "INFECTED" indicates that the tracker user is an infected person, and "CLOSE_CONTACT" indicates that the tracker user is a close contact. Using this information, other trackers can identify the tracker as a high-risk tracker when they receive the Bluetooth signal.

[0059] Optionally, in one possible implementation, the target tracker switches its own status information to a normal state when it detects a state switching instruction; wherein, the state switching instruction is generated when the duration of the high-risk state exceeds a preset duration, or, the state switching instruction is issued by the server to the target tracker when it receives a health recovery notification for a passenger holding the target tracker; or, the state switching instruction is generated when the passenger holding the target tracker confirms that the risk has been eliminated, and the target tracker outputs confirmation information as to whether the risk has been eliminated to the passenger holding the target tracker when the duration of the high-risk state exceeds the preset duration, so that the passenger can confirm whether the risk has been eliminated.

[0060] It should be noted that when a passenger recovers their health or completes a risk assessment update, the target tracker will remove its risk marker, revert to a normal tracker status, and cease broadcasting its initial Bluetooth signal. The preset duration can be 48 hours or a duration determined based on medical guidance.

[0061] In summary, the ship close contact visualization tracking method provided in this application allows the tracker to broadcast short-range data to the surrounding environment via Bluetooth signals, without relying on Wi-Fi or internet connections. This avoids the leakage of users' sensitive data on public networks. Furthermore, the data broadcast by the tracker via Bluetooth signals does not contain any information that can directly identify an individual; it only contains anonymized identifiers and special identification information used to indicate the status. This further reduces the risk of privacy leakage. In addition, when a passenger recovers their health or completes a risk assessment update, the target tracker's status will be reset to its initial state, ensuring timely management of high-risk statuses.

[0062] S102. When the other tracker receives the first Bluetooth signal, and determines that the other tracker and the target tracker are in close contact based on the first Bluetooth signal, the other tracker records and saves the contact event.

[0063] Specifically, when other trackers receive the first Bluetooth signal, they record the RSSI value of that Bluetooth signal, and the distance between the other tracker and the target tracker can be calculated using the following formula:

[0064] ;

[0065] in, This refers to the distance between the other tracker and the target tracker; The RSSI value (empirical value) of the Bluetooth signal received by the other tracker at a reference distance of 1m. is the RSSI value of the Bluetooth signal actually received by the other tracker from the target tracker; n is the environmental attenuation factor.

[0066] Furthermore, when the distance between the other tracker and the target tracker is less than a preset threshold, it is considered a close-range contact state. Simultaneously, the timestamp of the start of this close-range contact is recorded, and the duration of the current close-range contact state is continuously included in the "contact duration" as long as the calculated distance between the other tracker and the target tracker remains less than the preset threshold. The preset threshold can be set according to actual needs; in this application, it is not limited.

[0067] Furthermore, when it is determined from the first Bluetooth signal that the other tracker is in close contact with the target tracker, the anonymous identifier of the target tracker, the device identifier of the other tracker, the timestamp used to characterize the contact start point, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the location information of the other tracker during the contact duration are recorded by the other tracker as a contact event and stored locally.

[0068] It should be noted that this process ensures that all contact events are continuously recorded even when offline, providing complete data for subsequent data synchronization and tracking.

[0069] The location information of other trackers is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the vessel. Figure 3 Please refer to the schematic diagram of Embodiment 2 of the visual tracking method for close contact between ships provided in this application. Figure 3 In one possible implementation, the method for determining the position information of any other tracker at any given time includes:

[0070] S301. For any beacon detected by the other tracker within a preset time period before that moment, perform Kalman filtering and average filtering sequentially on the second Bluetooth signal from the beacon collected within the preset time period to obtain the filtered signal of the beacon.

[0071] It should be noted that the preset time period is set according to actual needs, and is not limited in this application. For example, in one embodiment, the preset time period can be 5 seconds.

[0072] Optionally, in one possible implementation, the step of performing Kalman filtering on the second Bluetooth signal from the beacon acquired within the preset time period may include:

[0073] Step 1: Obtain the current movement speed of the other tracker, as well as the sway information used to characterize the current degree of swaying of the ship.

[0074] An IMU (Inertial Measurement Unit) can be installed inside the tracker to obtain the tracker's movement speed.

[0075] Specifically, the accelerometer in the IMU collects the tracker's raw acceleration along the spatial coordinate axes according to a preset sampling rate. ( The gyroscope in the IMU samples the angular velocity of the tracker on the spatial coordinate axes according to a preset sampling rate. Then, using a sensor fusion algorithm, combining acceleration and angular velocity data, the tracker's attitude quaternion q is calculated to describe the tracker's rotation relative to the Earth coordinate system. Further, the gravitational acceleration [0, 0, -9.8] in the Earth coordinate system is transformed to the device coordinate system to obtain the gravity components. Using the measured original acceleration Subtracting the gravitational component, we obtain the tracker's pure acceleration: .

[0076] Furthermore, by integrating the pure motion acceleration over time according to the following formula, the current moving speed of the tracker can be calculated:

[0077] ;

[0078] in, This represents the tracker's current movement speed; The tracker's speed at the previous moment; The pure motion acceleration of the tracker; The sampling interval is denoted as .

[0079] Furthermore, within a preset time period, the accelerometer collects multiple sets of data from the tracker at multiple sampling time points according to a preset sampling rate, where each set of data includes the tracker's raw acceleration in the spatial coordinate axes. ( Then, the magnitude of the tracker's raw acceleration in the spatial coordinate system at each sampling time point is calculated. Then, based on the acceleration amplitude sequence at all sampling time points, the variance of the acceleration amplitude is calculated, and the calculated variance is used to characterize the current degree of ship rolling. That is, the calculated variance is the rolling information.

[0080] In practice, the variance can be calculated using the following formula:

[0081] ;

[0082] in, The variance of acceleration amplitude at multiple sampling time points within a preset time period; This refers to the number of sampling time points within a preset time period; Let be the acceleration amplitude at the i-th sampling time point; This is the average of the acceleration amplitudes at all sampling time points within the preset time period.

[0083] Step 2: Update the process noise covariance in the Kalman filter based on the moving speed.

[0084] In this step, the process noise covariance in the Kalman filter can be updated according to the following formula:

[0085] ;

[0086] in, The updated process noise covariance; This is a proportionality coefficient used to adjust the effect of speed on the process noise covariance. This represents the tracker's current movement speed; Based on the noise covariance.

[0087] Step 3: Update the measurement noise covariance in the Kalman filter based on the shaking information and the standard deviation of the second Bluetooth signal from the beacon collected within the preset time period.

[0088] In this step, the standard deviation of the second Bluetooth signal collected from the beacon within a preset time period can be calculated using the following formula:

[0089] ;

[0090] in, The standard deviation of the second Bluetooth signal from the beacon collected within a preset time period; The RSSI value of the i-th second Bluetooth signal from the beacon, collected within a preset time period; The value is the average RSSI value of N second Bluetooth signals from the beacon collected within a preset time period; N is the number of second Bluetooth signals from the beacon collected within the preset time period.

[0091] It should be noted that the RSSI (Received Signal Strength Indicator) value is an indicator used to measure signal strength and can reflect the quality of signal transmission. The RSSI value uses a logarithmic scale rather than a linear scale. By using a logarithmic function, the actual value of the signal strength can be compressed into a numerical range that is easier to process and compare.

[0092] Furthermore, the measurement noise covariance in the Kalman filter can be updated using the following formula:

[0093] ;

[0094] in, The updated measurement noise covariance; The fluctuation weight of the standard deviation of the second Bluetooth signal from the beacon collected within a preset time period; Shaking weights are used to indicate the degree of shaking of the tracker. Based on the fundamental noise covariance; The standard deviation of the second Bluetooth signal from the beacon collected within a preset time period; This is information that is being shaken.

[0095] Step 4: Perform Kalman filtering on the second Bluetooth signal from the beacon collected within the preset time period based on the updated process noise covariance and measurement noise covariance.

[0096] In this step, the second Bluetooth signal from the beacon collected within a preset time period can be Kalman filtered based on the updated process noise covariance and measurement noise covariance.

[0097] It should be noted that in this embodiment, by dynamically adjusting the process noise covariance based on the real-time movement speed of the tracker, the positioning offset caused by the high-speed movement of the tracker can be suppressed. At the same time, by dynamically adjusting the measurement noise covariance based on the fluctuation of the second Bluetooth signal and the degree of ship rolling, the impact of noise on Kalman filtering can be reduced when the signal is interfered with or the ship is rolling violently. In summary, by dynamically adjusting the process noise covariance and measurement noise covariance in Kalman filtering by combining movement speed and rolling information, efficient noise reduction of the second Bluetooth signal can be achieved, improving the accuracy and stability of positioning in complex ship environments.

[0098] Furthermore, in this step, after performing Kalman filtering on the second Bluetooth signal collected by any beacon within a preset time period, an intermediate filtered sequence is output. By averaging this intermediate filtered sequence, the filtered signal for the beacon can be generated. .

[0099] Specifically, the arithmetic mean of the intermediate filtered sequences can be calculated using the following formula to obtain the filtered signal of the beacon:

[0100] ;

[0101] in, This is the filtered signal for the beacon; This represents the number of samples in the intermediate filtered sequence. For the first The filtered RSSI value of each sampling point.

[0102] It should be noted that after the second Bluetooth signal undergoes Kalman filtering, an intermediate filtered sequence is generated. Averaging filtering then eliminates residual high-frequency fluctuations in this intermediate filtered sequence, generating a stable filtered signal. This addresses the instability of RSSI values ​​caused by metal reflections, obstructions, and rolling in a marine environment. The combined use of Kalman filtering and averaging filtering preserves the dynamic response of Kalman filtering while using averaging filtering to suppress fluctuations, thus improving the stability of positioning and signal interpretation. Furthermore, Kalman filtering has adaptive suppression capabilities for single-point jumps, and averaging filtering, used subsequently, can further compress the duration of abrupt changes, preventing them from affecting subsequent processing.

[0103] In summary, in this embodiment, by first performing Kalman filtering on the second Bluetooth signal collected from any beacon within a preset time period, the dynamic adaptability of Kalman filtering is used to eliminate complex interference in the ship environment. Then, the intermediate filtered sequence obtained after Kalman filtering is averaged and filtered using the smoothing ability of average filtering to smooth the signal processing, finally generating a high-quality filtered signal. By sequentially processing the second Bluetooth signal through Kalman filtering and average filtering, the stability and reliability of the filtered signal can be significantly improved.

[0104] S302. When the other tracker detects fewer than three beacons within the preset time period, the position of the other tracker is determined based on the filtered signal of any one beacon and the position information of that beacon.

[0105] Specifically, when the other tracker detects fewer than three beacons within the preset time period, the other tracker can determine its location using the filtered signal of any one beacon and the beacon's known location information.

[0106] Specifically, the distance between the other tracker and the beacon can be calculated using the following formula:

[0107] ;

[0108] in, The distance between the other tracker and the beacon. The strength of the filtered signal for this beacon; The RSSI value of the Bluetooth signal received by the other tracker at a reference distance of 1m; As an environmental degradation factor, The value is between 2.2 and 2.8.

[0109] Furthermore, the location of the other tracker can be determined by the distance between the other tracker and the beacon, as well as the location information of the beacon.

[0110] S303. When the other tracker detects at least three beacons within the preset time period, for each of the at least three beacons, a confidence score for the beacon is determined based on the signal strength of the beacon's filtered signal and the duration for which the other tracker continuously detects the beacon.

[0111] Specifically, the confidence score for each beacon can be calculated using the following formula:

[0112] ;

[0113] in, Calculate the confidence score for each beacon; It is the first weight; This is the normalized scoring function for the signal strength of the filtered signal of the beacon; As the second weight; A normalized scoring function for the duration during which the other trackers continuously detect the beacon.

[0114] Furthermore, the normalized scoring function for the signal strength of the beacon's filtered signal can be determined according to the following formula:

[0115] ;

[0116] in, This is the normalized scoring function for the signal strength of the filtered signal of the beacon; This represents the strength value of the filtered signal for the beacon. It is the minimum value among the intensity values ​​of the filtered signals from at least three beacons; It is the maximum value among the intensity values ​​of the filtered signals from at least three beacons.

[0117] Furthermore, the normalized scoring function for the duration during which other trackers continuously detect the beacon can be determined according to the following formula:

[0118] ;

[0119] in, A normalized scoring function for the duration during which the other trackers continuously detect the beacon; The duration for which other trackers continuously detect the beacon; Duration of at least three beacons The maximum value.

[0120] S304. Select three target beacons from the at least three beacons in descending order of confidence, and determine the position information of the other tracker based on the filtered signals of the three target beacons and the position information of the three target beacons.

[0121] In this step, based on the confidence score of each beacon calculated in the above formula, the three target beacons with the highest confidence scores are selected from high to low, and the location information of the other trackers is determined using the triangulation method.

[0122] Specifically, given the coordinates of the three target beacons are as follows: , and In this case, the estimated distance between each target beacon and the other trackers is: , and And construct three circle equations according to the following formula, with the position of each target beacon as the center and the distance between the target beacon and the other trackers as the radius:

[0123] ;

[0124] ;

[0125] ;

[0126] Furthermore, based on these three circle equations, the common intersection point of the three circle equations is solved, and the position of this common intersection point is the position information of the other tracker.

[0127] It should be noted that, due to errors in actual measurements, the three circular equations typically do not intersect at a single point. Therefore, in this application, weighted least squares or nonlinear least squares optimization can be used to solve for the best-fit positions of the three circular equations. Specifically, during the solution process, the weight of each beacon's position information relative to the best-fit position (i.e., the weight of each beacon) can be allocated based on the confidence level of each beacon, ensuring that more reliable measurements have a greater weight in the final best-fit position.

[0128] The following example uses the weighted least squares method for illustration:

[0129] When solving the three circle equations using the weighted least squares method, the best-fit position can be found using the following formula. This minimizes the error between the best-fit position and the actual position of the other trackers.

[0130] ;

[0131] in, Let be the weight of the i-th beacon; Let i be the distance between the i-th target beacon and the other trackers; Let be the coordinates of the i-th target beacon.

[0132] Furthermore, the weight of each beacon can be mapped by the confidence score of each beacon; in one possible implementation, the mapping can be performed according to the following formula:

[0133] ;

[0134] in, Let be the weight of the i-th beacon; The confidence score for the i-th beacon.

[0135] It should be noted that the exponent k can amplify the weight of beacons with high confidence scores while suppressing the influence of beacons with low confidence scores; k > 1, and the specific value of k can be chosen according to actual needs, and is not limited in this application. For example, in one possible implementation, k = 2.

[0136] In addition, the error between the best-fit position and the actual position of the other trackers should be controlled within ±0.5 meters. If the error is large, the positioning result should be discarded, or additional triangulation and weighted least squares methods should be used for correction.

[0137] S103. After successfully establishing a connection with the server, the other trackers will report all saved contact events to the server.

[0138] In this step, after successfully establishing a connection with the server, the other trackers will report all saved contact events to the server.

[0139] It should be noted that due to the instability of the network on board the ship, the tracker attempts to connect to the server at preset intervals. For example, the preset interval is 5 minutes. The tracker attempts to connect to the server every 5 minutes. If the connection fails, it continues to remain offline and waits for the next connection attempt. If the connection is successful, the tracker will receive a response from the server.

[0140] Furthermore, after the tracker successfully connects to the server, it integrates all locally stored contact events into a data batch according to a predetermined format, and can use compression algorithms such as GZIP to compress the data to reduce its size. Then, the tracker sends the data batch to the server via an HTTP POST request. Upon receiving the data, the server verifies it to ensure its integrity and decrypts it. Additionally, after the server processes and integrates the uploaded data, it returns a confirmation response to the tracker. Upon receiving this confirmation response, the tracker deletes the successfully uploaded data from its local storage and logs the successful upload.

[0141] It should be noted that if a network interruption occurs or the server returns an error code during the process of the tracker uploading data to the server, the tracker will retain the complete offline data that was not fully uploaded and will re-attempt the upload after a preset time interval to ensure that all offline data can eventually be stably uploaded to the server.

[0142] S104. The server determines the activity trajectory of the target tracker based on the received contact event, and for each trajectory point in the activity trajectory, determines the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker.

[0143] Optionally, in one possible implementation, the specific implementation process of this step may include:

[0144] Step 1: For each contact event, extract the location information of other trackers corresponding to that contact event.

[0145] Specifically, if a target tracker comes into close contact with another tracker "Tracker_001" in area A of deck, the other tracker "Tracker_001" records a contact event. Furthermore, when the server receives a record containing this contact event, it can directly extract the location information of the other tracker in that contact event.

[0146] Step 2: Cluster all location information extracted from all contact events to obtain clustering results.

[0147] In this step, clustering algorithms such as DBSCAN or K-means can be used to cluster all extracted locations to obtain clustering results. It is understood that the clustering results include multiple clusters, each representing an active region.

[0148] Step 3: Treat the spatial range represented by each cluster in the clustering results as an activity area of ​​the target tracker, and take the location of the cluster center as the trajectory point of the target tracker in the activity area.

[0149] In this step, the spatial range represented by each cluster is defined as an activity region of the target tracker, and the center point of each cluster is taken as the trajectory point of the target tracker within that activity region. For example, if the clustering result contains 5 clusters, for ease of distinction, these 5 clusters are denoted as region A, region B, region C, region D, and region E, and these 5 regions are respectively used as 5 activity regions of the target tracker. Furthermore, for example, for region A, if the coordinates of the cluster center are (X=50, Y=30), then these coordinates are determined as the trajectory point of the target tracker within region A.

[0150] Step 4: Connect all trajectory points of the target tracker in all activity areas in chronological order to obtain the activity trajectory of the target tracker.

[0151] In this step, connecting multiple trajectory points in chronological order yields the target tracker's activity trajectory. Using the example above, if the class center of region A is A1, the class center of region B is B1, the class center of region C is C1, the class center of region D is D1, and the class center of region E is E1, and the chronological order is A1-C1-B1-E1-D1, then the target tracker's activity trajectory is A1-C1-B1-E1-D1.

[0152] Step 5: Identify other trackers within each activity area as close contact trackers that are in close contact with the target tracker at the trajectory point corresponding to that activity area, and determine the contact duration of the close contact tracker based on the contact event corresponding to the close contact tracker.

[0153] For example, if the target tracker and tracker f3 are both in area A for 10 minutes, then tracker f3 is identified as a close contact, and the contact duration of the close contact tracker is 600 seconds.

[0154] It should be noted that by clustering the location information contained in contact events, the system can reconstruct the movement trajectory within the ship without requiring the target tracker to have independent positioning capabilities. This effectively reduces the system's dependence on hardware and improves deployment flexibility. Furthermore, by extracting the target tracker's dwell positions in each activity area through clustering and constructing trajectory points accordingly, the system achieves accurate reconstruction of its actual movement path. Simultaneously, identifying trackers with close contact relationships at each trajectory point and combining this with contact duration information can more accurately reflect the true close contact situation, avoiding misjudging brief contacts as risk events. This method not only improves the spatial accuracy and temporal reliability of close contact determination but also provides excellent visualization capabilities for the final generated trajectory and close contact relationships, offering efficient and traceable technical support for epidemic prevention and control, safety management, and emergency response within the ship.

[0155] Furthermore, in another possible implementation, the specific implementation process of this step may include:

[0156] Step 1: Classify all received contact events according to their timestamps to obtain the first classification result.

[0157] In this step, a fixed time interval is preset, and all contact events within a certain time interval are grouped into the same category. For example, contact events within the time interval of 10:00-10:05 are grouped into the same category.

[0158] Step 2: For each class in the first classification result, fit the trajectory point of the target tracker under the corresponding timestamp of the class based on the location information recorded in all contact events in that class.

[0159] Specifically, the trajectory points of the target tracker under the corresponding timestamp of the class can be fitted according to the following formula based on the location information recorded in all contact events of that class:

[0160] ;

[0161] in, For the target tracker, the trajectory points under the corresponding timestamp of this category; N represents the location data recorded for all contact events in this class; N is the number of all contact events in this class. The weight of each location data is determined based on the contact duration recorded in the contact event corresponding to that location data; the longer the contact duration, the greater the weight.

[0162] Step 3: According to the chronological order, the line connecting the trajectory points of the target tracker under all timestamps is determined as the activity trajectory of the target tracker.

[0163] Step 4: Identify the other trackers corresponding to all contact events under each category as close contact trackers that are in close contact with the target tracker at the trajectory point, and determine the contact duration of the close contact tracker based on the contact events corresponding to the close contact tracker.

[0164] For details on the implementation process and principles of steps 3 and 4, please refer to the previous descriptions, which will not be repeated here.

[0165] It should be noted that by classifying received contact events by timestamp and combining this with the location information of other trackers within each time period, the trajectory position of the target tracker at each time point is fitted, thus achieving activity trajectory reconstruction based on time-series information. This method not only more accurately reconstructs the dynamic movement process of the target tracker but also effectively eliminates noise and abnormal data, improving the accuracy and stability of trajectory fitting. Simultaneously, identifying the contact object within each time period as the close contact tracker at the corresponding trajectory point and calculating the contact duration based on the contact events helps to accurately identify high-risk contact personnel. This not only achieves simultaneous modeling of the spatiotemporal behavior of the target tracker and interpersonal contact, providing higher timeliness and reliability for close contact analysis and risk tracing, but also further enhances the practicality and safety control capabilities of the tracking system in the enclosed environment of a ship.

[0166] S105. The server overlays the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, and displays the visual tracking interface to the passengers.

[0167] Specifically, in one possible implementation, the process of implementing this step may include:

[0168] (1) Loading ship plan map and establishing coordinate system

[0169] The server pre-stores or loads electronic planar map layers corresponding to the ship's structure and establishes a unified spatial coordinate system for receiving and displaying various positioning and trajectory data. This coordinate system is consistent with the beacon deployment location or tracker location information to ensure accurate spatial information mapping.

[0170] (2) Trajectory point coordinate transformation and trajectory line generation

[0171] Based on the target tracker's trajectory point location information at different time points, the server maps these spatial points to corresponding coordinate positions on the ship's plan view. Subsequently, the trajectory points are connected in chronological order to form a polyline trajectory, creating a complete activity path.

[0172] (3) Close contact tracker information labeling

[0173] For each trajectory point, the server retrieves other trackers that have a history of contact with the target tracker at that location and identifies them as close contact trackers. The locations of close contact trackers are also mapped onto the floor plan and marked with icons or numbers near the corresponding trajectory points.

[0174] (4) Overlay of contact duration data

[0175] For each pair of close contacts, the server displays the contact duration on a visual interface using text, heatmaps, color gradients, or labels. For example, it may display "Contact Duration: 3 minutes 20 seconds" next to the close contact tracker icon, or use color depth to reflect contact intensity.

[0176] (5) Generation and display of interactive interface

[0177] Finally, the server integrates the generated trajectory lines, close contact locations, and contact duration data into the ship's plan view in an overlay format, forming a unified, dynamic, and visual tracking interface. This interface supports interactive operations such as zooming, dragging, and timeline playback, allowing passengers to promptly understand relevant contact risk information.

[0178] The method provided in this embodiment achieves accurate tracking and visualization of close contact between high-risk individuals while fully protecting the privacy of passengers. It anonymizes the identifier of the target tracker device to ensure that the individual's true identity is not exposed during the entire data collection and interaction process, effectively preventing privacy leaks. Simultaneously, leveraging Bluetooth signal strength and beacon-assisted positioning technology, the system can accurately identify close contact events between personnel and construct the dynamic activity trajectory of the target tracker based on spatiotemporal information. Through analysis of contact events, the server not only reconstructs the target personnel's movement path within the ship but also accurately identifies their close contacts and the duration of contact at each trajectory point, ultimately displaying this information in an intuitive and interactive manner on the ship's map. This solution achieves highly efficient and accurate contact tracing and risk identification without relying on personal identification data, providing intelligent and low-intrusion technical support for epidemic prevention and control and safety management.

[0179] The method provided in this embodiment, based on the above embodiments, further includes:

[0180] Step 1: The server categorizes all contact events according to timestamp, device identifier, and location information to obtain the first, second, and third classification results.

[0181] In this step, the contact events are categorized according to their timestamps to obtain the first category result. Each hour can be recorded as a time period, and combined with the current date, a time-organized list or index can be created, facilitating the querying of contact events by time. For example, the time period "2025-04-27 10:00-11:00" contains 100 contact events.

[0182] Similarly, categorizing by device identifier allows you to associate all contact events for each tracker with its unique ID. For example, tracker "Tracker_001" might be associated with 50 contact events, while tracker "Tracker_002" might be associated with 30.

[0183] Similarly, based on location information, events can be grouped according to the area where the tracker is located, with nearby coordinates grouped into the same area. For example, "Deck A" contains 80 contact events, and "Cabin" contains 20 contact events.

[0184] Step 2: The server constructs a tracking database based on the first classification result, the second classification result, and the third classification result.

[0185] In this step, a tracking database is constructed based on the first, second, and third classification results obtained from the above steps. This tracking database stores all contact events and can support operators in efficient querying and analysis.

[0186] Step 3: In response to a display instruction triggered by a passenger based on the visual tracking interface, the server retrieves the target data to be displayed from the tracking database and displays the target data to the passenger; wherein, the display instruction is used to instruct the display of contact events within a specified time period or a specified area.

[0187] Figure 4 For a schematic diagram illustrating the visualization tracking interface provided in an exemplary embodiment of this application, please refer to... Figure 4 .from Figure 4 As can be seen, relevant personnel can input display commands and determine filtering conditions on the visual tracking interface, and the server can display the corresponding target data based on the display commands.

[0188] It should be noted that in this embodiment, by classifying contact events into multiple dimensions according to timestamps, device identifiers, and location information, a structured tracking database is constructed, making the storage and retrieval of contact information more efficient and flexible. With this database, the server can quickly respond to display commands issued by passengers in the visual interface, extracting contact events within a specified time period or area as needed, and dynamically presenting the contact status and trajectory information of relevant personnel. This mechanism not only significantly improves the real-time performance and accuracy of the system in query response and data visualization, but also supports users in flexibly filtering and analyzing contact risks across different dimensions, effectively enhancing epidemic prevention and emergency management capabilities in the enclosed environment of a ship.

[0189] Optionally, in one possible implementation, the method may further include:

[0190] (1) The server responds to the filtering conditions entered by passengers based on the visual tracking interface and filters the target data indicated by the filtering conditions from the tracking database.

[0191] Specifically, passengers can input query parameters through the visual tracking interface. For example, they can filter for contact events in the time period from 10:00 to 12:00 on April 27, 2025, in the area of ​​"Deck A", and specify the tracker "Tracker_001". In this way, a set of contact events that match the filtering conditions entered by the passengers can be obtained from the tracking database. For example, the set of contact events includes 100 records, and each record contains the timestamp of each contact event, its own device identifier, and location information.

[0192] (2) The server performs statistical processing on the target data to generate statistical indicators; the statistical indicators include: number of contact events, cumulative contact time, and risk area distribution.

[0193] Specifically, the server aggregates and analyzes the target data, extracts key information, and transforms it into quantitative indicators. This helps users quickly understand risk distribution and exposure characteristics. SQL aggregation functions or big data processing tools can be used to perform statistical processing on the target data.

[0194] In practice, based on the filtered target data, statistical fields are extracted and aggregated. Basic statistical methods such as counting, summation, and grouping are primarily used. For example:

[0195] (1) Statistics on the number of contact events:

[0196] Directly count the number of contact events in the target dataset.

[0197] If the display is based on a region or time distribution, then the count is grouped and grouped by the location_id or timestamp field.

[0198] (2) Cumulative contact time:

[0199] Iterate through the contact duration field of each contact event record in the target data and sum them.

[0200] If the statistics are further refined to be based on individuals or regions, then the data is aggregated separately using the device ID or location_id as the grouping key.

[0201] (3) Distribution of risk areas:

[0202] Based on the location information field (location_id or coordinates) of each contact event, count the number of events or the total contact time.

[0203] (4) The server organizes the statistical indicators to generate a structured report and displays the structured report to the passengers.

[0204] Structured reports can be designed in the form of tables or charts. Tables can display the raw data of statistical indicators, while charts can use bar charts to compare the number of contact events in different areas, line charts to show the trend of contact events over time, and heatmaps to represent the distribution density of risk areas on a map. Furthermore, Table 1 below provides an example of the content of a structured report:

[0205] Table 1

[0206] Number of contact events 100 items Number cards + bar chart Cumulative contact time 1000 seconds Progress bar + pie chart Distribution of risk areas Deck Area A Heat map + regional ranking table

[0207] The method provided in this embodiment responds to the filtering conditions input into the visual tracking interface via a server, efficiently filters target data from the tracking database, and generates multi-dimensional statistical indicators based on this data, including the number of contact events, cumulative contact time, and risk area distribution. This data is then used to construct structured reports and display them to passengers, achieving a comprehensive quantitative analysis of close contact. This solution not only improves the real-time nature and accuracy of responding to public health risks such as epidemics, but also allows passengers to intuitively understand their own and their environment's contact risk levels. Thus, while protecting personal privacy, it helps ship operators achieve data-driven health management and emergency response.

[0208] A second aspect of this application provides a tracking system, the tracking system comprising multiple beacons located at specific locations on a ship, a server installed on the ship, and a tracker held by each passenger on the ship; wherein:

[0209] The target tracker in a high-risk state among all trackers is used to periodically broadcast a first Bluetooth signal to notify the other trackers besides the target tracker of its own status information through the first Bluetooth signal; the first Bluetooth signal carries the anonymous identifier of the target tracker; the anonymous identifier is an identifier after anonymizing the device identifier of the target tracker.

[0210] The other tracker, upon receiving the first Bluetooth signal and determining, based on the first Bluetooth signal, that the other tracker and the target tracker are in close proximity, record and save a contact event; wherein the contact event includes the anonymous identifier of the target tracker, the device identifier of the other tracker, a timestamp representing the start point of the contact, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the location information of the other tracker during the contact duration; the location information of the other tracker is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the vessel;

[0211] The other trackers are also used to report all saved contact events to the server after a successful connection is established with the server;

[0212] The server is configured to determine the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determine the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker.

[0213] The server is also used to overlay the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, and to display the visual tracking interface to passengers.

[0214] The specific working principle of this tracking system is described in the previous embodiments and will not be repeated here.

[0215] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for visually tracking close contact between ships, characterized in that, The method is applied to a tracking system, which includes multiple beacons located at specific locations on a vessel, a server mounted on the vessel, and a tracker held by each passenger on board; the method includes: After receiving an update command, the tracker updates its own status to high-risk. Trackers in a high-risk state are designated as target trackers. After updating their status, target trackers in a high-risk state periodically broadcast a first Bluetooth signal to notify all other trackers except the target tracker of their status information. When a passenger recovers their health or completes a risk assessment update, the target tracker cancels its risk label, becomes a normal tracker, and stops broadcasting the first Bluetooth signal. The first Bluetooth signal carries the anonymous identifier of the target tracker; the anonymous identifier is an identifier obtained by anonymizing the device identifier of the target tracker. When the other trackers receive the first Bluetooth signal, they record the RSSI value of the first Bluetooth signal and can calculate the distance between the other trackers and the target tracker according to the following formula: ； in, The distance between the other trackers and the target tracker; The RSSI value of the Bluetooth signal received by the other trackers at a reference distance of 1m; The RSSI value is the actual value of the Bluetooth signal emitted by the target tracker received by the other trackers; n is the environmental attenuation factor. When it is determined, based on the first Bluetooth signal, that the other tracker is in close proximity to the target tracker, a contact event is recorded and saved; wherein, the contact event includes the anonymous identifier of the target tracker, the device identifier of the other tracker, a timestamp representing the start point of the contact, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the location information of the other tracker during the contact duration; the location information of the other tracker is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the vessel; After successfully establishing a connection with the server, the other trackers will report all saved contact events to the server. The server determines the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determines the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker. The server overlays the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, which is then displayed to the passengers.

2. The method according to claim 1, characterized in that, Based on the received contact events, the activity trajectory of the target tracker is determined, and the close contact trackers that are in close contact with the target tracker at each trajectory point, as well as the contact duration of the close contact trackers, are determined, including: For each contact event, extract the location information of other trackers corresponding to that contact event; Cluster all location information extracted from all contact events to obtain clustering results; The spatial range represented by each cluster in the clustering results is regarded as an activity area of ​​the target tracker, and the location of the cluster center is taken as the trajectory point of the target tracker in the activity area. Connect all trajectory points of the target tracker in all activity areas in chronological order to obtain the activity trajectory of the target tracker; Other trackers within each activity area are identified as close contact trackers that are in close contact with the target tracker at the corresponding trajectory point in that activity area, and the contact duration of the close contact tracker is determined based on the contact event corresponding to the close contact tracker.

3. The method according to claim 1, characterized in that, The step of determining the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determining the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker, includes: All received contact events are classified according to their timestamps to obtain the first classification result; For each category in the first classification result, the trajectory point of the target tracker under the corresponding timestamp of the category is fitted based on the location information recorded in all contact events in that category; The trajectory of the target tracker is determined by connecting the trajectory points of the target tracker under all timestamps in chronological order. For each category, identify the other trackers corresponding to all contact events as close contact trackers that are in close contact with the target tracker at the trajectory point, and determine the contact duration of the close contact tracker based on the contact events corresponding to the close contact tracker.

4. The method according to claim 1, characterized in that, The method further includes: The server categorizes all contact events according to timestamp, device identifier, and location information, resulting in a first category, a second category, and a third category. The server constructs a tracking database based on the first classification result, the second classification result, and the third classification result; In response to a display instruction triggered by a passenger based on the visual tracking interface, the server retrieves the target data to be displayed from the tracking database and displays the target data to the passenger; wherein, the display instruction is used to instruct the display of contact events within a specified time period or a specified area.

5. The method according to claim 4, characterized in that, The method further includes: The server responds to the filter criteria entered by passengers based on the visual tracking interface, and filters the target data indicated by the filter criteria from the tracking database; The server performs statistical processing on the target data to generate statistical indicators; the statistical indicators include: number of contact events, cumulative contact time, and distribution of risk areas. The server organizes the statistical indicators to generate a structured report, and then displays the structured report to the passengers.

6. The method according to claim 1, characterized in that, Methods for determining the location information of any other tracker at any given time include: For any beacon detected by the other tracker within a preset time period prior to that moment, the second Bluetooth signal from that beacon collected within the preset time period is sequentially subjected to Kalman filtering and average filtering to obtain the filtered signal of that beacon; If the other tracker detects fewer than three beacons within the preset time period, the location of the other tracker is determined based on the filtered signal of any one beacon and the location information of that beacon; When the other tracker detects at least three beacons within the preset time period, for each of the at least three beacons, a confidence score for the beacon is determined based on the signal strength of the beacon's filtered signal and the duration for which the other tracker continuously detects the beacon. Three target beacons are selected from the at least three beacons in descending order of confidence, and the location information of the other tracker is determined based on the filtered signals of the three target beacons and the location information of the three target beacons.

7. The method according to claim 6, characterized in that, Kalman filtering is performed on the second Bluetooth signal from the beacon collected within the preset time period, including: Obtain the current movement speed of the other tracker, as well as sway information used to characterize the current degree of swaying of the ship; Update the process noise covariance in the Kalman filter based on the moving speed; The measurement noise covariance in the Kalman filter is updated based on the shaking information and the standard deviation of the second Bluetooth signal from the beacon collected within the preset time period. Kalman filtering is applied to the second Bluetooth signal from the beacon acquired within the preset time period based on the updated process noise covariance and measurement noise covariance.

8. The method according to claim 7, characterized in that, Based on the moving speed, update the process noise covariance in the Kalman filter, including: The process noise covariance in the Kalman filter can be updated using the following formula: ； in, The updated process noise covariance; This is a proportionality coefficient used to adjust the effect of speed on the process noise covariance. This represents the tracker's current movement speed; Based on the fundamental noise covariance; Based on the shaking information and the standard deviation of the second Bluetooth signal from the beacon collected within the preset time period, the measurement noise covariance in the Kalman filter is updated, including: The measurement noise covariance in the Kalman filter can be updated using the following formula: ； in, The updated measurement noise covariance; The fluctuation weight of the standard deviation of the second Bluetooth signal from the beacon collected within a preset time period; Shaking weights are used to indicate the degree of shaking of the tracker. Based on the fundamental noise covariance; The standard deviation of the second Bluetooth signal from the beacon collected within a preset time period; This represents the variance of the acceleration amplitude at multiple sampling time points within a preset time period.

9. The method according to claim 1, characterized in that, The method further includes: When the target tracker detects a state switching command, it switches its own state information to a normal state. The state switching command is generated when the duration of the high-risk state exceeds a preset duration; or, the state switching command is issued by the server to the target tracker when it receives a health recovery notification for a passenger holding the target tracker; or, the state switching command is generated when the passenger holding the target tracker confirms that the risk has been eliminated, and the target tracker outputs confirmation information regarding whether the risk has been eliminated to the passenger holding the target tracker when the duration of the high-risk state exceeds the preset duration, allowing the passenger to confirm whether the risk has been eliminated.

10. A tracking system, characterized in that, The tracking system includes multiple beacons located at specific locations on the ship, a server installed on the ship, and a tracker held by each passenger on board; wherein: After receiving an update command, the tracker updates its own status to a high-risk state. Trackers in a high-risk state are designated as target trackers. After updating their status, target trackers in a high-risk state periodically broadcast a first Bluetooth signal to notify all other trackers except the target tracker of their status information. When a passenger recovers their health or completes a risk assessment update, the target tracker cancels its risk label, its status becomes a normal tracker, and it stops broadcasting the first Bluetooth signal. The first Bluetooth signal carries the anonymous identifier of the target tracker; the anonymous identifier is an identifier obtained by anonymizing the device identifier of the target tracker. The other trackers are configured to record the RSSI value of the first Bluetooth signal upon receiving the first Bluetooth signal, and can calculate the distance between the other trackers and the target tracker according to the following formula: ； in, The distance between the other trackers and the target tracker; The RSSI value of the Bluetooth signal received by the other trackers at a reference distance of 1m; The RSSI value is the actual value of the Bluetooth signal emitted by the target tracker received by the other trackers; n is the environmental attenuation factor. When it is determined, based on the first Bluetooth signal, that the other tracker is in close proximity to the target tracker, a contact event is recorded and saved; wherein, the contact event includes the anonymous identifier of the target tracker, the device identifier of the other tracker, a timestamp representing the start point of the contact, the contact duration, the strength of each first Bluetooth signal received by the other tracker during the contact duration, and the location information of the other tracker during the contact duration; the location information of the other tracker is determined based on a second Bluetooth signal received by the other tracker from a beacon deployed within the vessel; The other trackers are also used to report all saved contact events to the server after a successful connection is established with the server; The server is configured to determine the activity trajectory of the target tracker based on the received contact events, and for each trajectory point in the activity trajectory, determine the close contact tracker that is in close contact with the target tracker at that trajectory point, and the contact duration of the close contact tracker. The server is also used to overlay the activity trajectory, the close contact tracker, and the contact duration of the close contact tracker onto the ship's planar map to generate a visual tracking interface, and to display the visual tracking interface to passengers.

Images