Wristband device-based elder falling alarm system and alarm method

Abstract

The invention provides an old man tumbling alarm system and alarm method based on bracelet equipment, and belongs to the technical field of intelligent wearable equipment. Comprising an intelligent bracelet, a mobile terminal and a cloud alarm platform. An accelerometer and a low-power-consumption wireless communication module are arranged in the intelligent bracelet, an accelerometer, a GPS positioning module, a wide area network communication module and a near field communication module are arranged in the mobile terminal, and the intelligent bracelet and the mobile terminal are in wireless communication and are used for jointly monitoring posture changes of a user. And the cloud alarm platform is used for transmitting the alarm information of the smart bracelet and the mobile terminal to a preset emergency contact. According to the invention, accurate position information is provided, the smart bracelet is assisted in tumble detection, the accurate position information is ensured, a redundant GPS module and a wide area network communication module of the smart bracelet are reduced, the power consumption of the smart bracelet is reduced, and through multi-device cooperative detection, machine learning algorithm optimization and a composite positioning system, the accuracy of tumble detection of the smart bracelet is improved. And the accuracy and the reliability of fall detection are obviously improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of smart wearable devices, and particularly relates to an elderly fall alarm system and an alarm method based on a bracelet device. Background Art

[0002] With the increase of age, the physical and mental functions gradually degenerate, and the incidence of falls among the elderly will gradually increase. If the elderly fall into a coma after falling, the situation will be even more critical, which has also become the leading cause of injury or death for the elderly at high ages. Existing technologies generally include camera-based determination, infrared detection, or protection devices after falling. Although these technologies have their own characteristics, due to high costs or inconvenient wearing, it is difficult to apply them to the market. At the same time, there are also smart bracelets that use acceleration sensors to detect the falls of the elderly. The smart bracelets themselves have complete GPS, wide-area network communication modules, and short-range communication modules. However, such smart bracelets have high power consumption, and in scenarios such as swinging the hand or quickly placing an item while wearing the bracelet, the false alarm rate is relatively high. Therefore, there is an innovation to provide a simple, low-power, and low-cost elderly fall alarm system and an alarm method based on a bracelet device. Summary of the Invention

[0003] (1) Technical Problems to be Solved Aiming at the deficiencies of the existing technology, the purpose of the present invention is to provide an elderly fall alarm system and an alarm method based on a bracelet device to solve the problems of high power consumption of existing smart bracelets using acceleration sensors to detect the falls of the elderly and relatively high false alarm rates in scenarios such as swinging the hand or quickly placing an item while wearing the bracelet.

[0004] (2) Technical Solutions To solve the above technical problems, the present invention provides such an elderly fall alarm system based on a bracelet device, including a smart bracelet, a mobile terminal, and a cloud alarm platform; The smart bracelet is internally provided with an accelerometer and a low-power wireless communication module. The mobile terminal is internally provided with an accelerometer, a GPS positioning module, a wide-area network communication module, and a short-range communication module. The smart bracelet and the mobile terminal communicate wirelessly and are used to jointly monitor the posture changes of the user. The cloud alarm platform is used to convey the alarm information of the smart bracelet and the mobile terminal to a preset emergency contact.

[0005] Preferably, the accelerometer is a three-axis accelerometer. The low-power wireless communication module of the smart bracelet supports Bluetooth and Wi-Fi networks. The mobile terminal is a smartphone for daily use. The wide-area network communication module and the short-range communication module support Bluetooth, mobile networks, or Wi-Fi networks.

[0006] An old person fall alarm method based on a bracelet device, the alarm method comprising the following steps: Step 1: The smart bracelet and the mobile terminal detect acceleration data through an accelerometer. When the detected acceleration data exceeds a preset threshold, it is determined that the user has fallen; Step 2: The mobile terminal provides accurate location information and wirelessly transmits it to the cloud alarm platform for primary alarm. At the same time, the smart bracelet performs vibration and buzzer sound reminders; Step 3: The cloud alarm platform conveys the primary alarm information to the preset emergency contacts.

[0007] Preferably, in step 1, the smart bracelet and the mobile terminal use an acceleration data fusion algorithm to determine whether the user has fallen.

[0008] Preferably, the acceleration data fusion algorithm comprises the following steps: S1: Denoise the acceleration data collected by the smart bracelet 1 and the mobile terminal, and then synchronize the data in time to ensure that the data collected by the smart bracelet 1 and the mobile terminal at the same time point can correspond; S2: Assign weights to the detection data of the smart bracelet 1 and the mobile terminal, and calculate the fused acceleration data according to the weights.

[0009] Preferably, the weight assignment method is: W1 + W2 = 1; wherein, W1 represents the weight of the smart bracelet 1, W2 represents the weight of the mobile terminal, and 1 represents the total weight of the smart bracelet 1 and the mobile terminal; The acceleration data calculation method is: Afused = W1 * Abracelet + W2 * Aphone; wherein, Afused represents the fused acceleration data, Abracelet represents the acceleration data collected by the smart bracelet 1, and Aphone represents the acceleration data collected by the mobile terminal.

[0010] Preferably, the preset threshold in step 1 is: the acceleration value detected by the smart bracelet exceeds the preset value and lasts for a certain period of time, and the morphological change conforms to the falling characteristics. At the same time, the mobile terminal detects that the acceleration change exceeds the preset value within the preset time threshold (for example, 3 seconds), and the spatial attitude suddenly changes. The preset value will be adjusted by deep learning according to user behavior analysis.

[0011] Preferably, a barrier-free help background service is installed on the mobile terminal and shakes hands with the APP installed on the mobile terminal through AIDL in real time. The alarm method further comprises the following steps: Step 1: After the APP disconnects the handshake with the accessibility service, the accessibility service of the mobile terminal turns on Bluetooth or Wi-Fi scanning and monitors Bluetooth or Wi-Fi broadcasts in real time. Step 2: After the accessibility service of the mobile terminal monitors the distress broadcast of the smart bracelet, it immediately reports to the cloud alarm platform and establishes a connection communication (such as Bluetooth) with the smart bracelet. At the same time, it feeds back the successful result of reporting to the cloud alarm platform. When the report to the cloud alarm platform is successful, the vibration and buzzer sound reminder of the smart bracelet are cancelled.

[0012] Preferably, an emergency button is installed on the smart bracelet, and long pressing the emergency button is used to trigger secondary manual help seeking.

[0013] Preferably, the cloud alarm platform notifies the preset emergency contacts by means of text messages, phone calls, APP push, etc.

[0014] (3) Beneficial effects Compared with the prior art, the beneficial effects of the present invention are as follows: In the above solution, the acceleration changes of the user can be monitored in real time through the smart bracelet and the elderly's common mobile terminal. And an accelerometer and a GPS positioning module are built into the elderly's common mobile terminal, which not only provides accurate location information, but also assists the smart bracelet in fall detection. At the same time, the alarm information is timely conveyed to the preset emergency contacts through the cloud alarm platform, so as to ensure accurate location information, reduce the redundant GPS module and wide-area network communication module of the smart bracelet, reduce the power consumption of the smart bracelet, and make the battery of the smart bracelet last longer.

[0015] In the above solution, a dual-device collaborative detection algorithm is adopted. Only when the smart bracelet and the mobile terminal simultaneously detect specific acceleration changes and spatial attitude mutations, will the first-level automatic alarm be triggered, which can significantly reduce the false alarm rate caused by misjudgment of a single device. Moreover, the system also sets up a manual emergency mechanism. When the mobile terminal is not carried with or fails to synchronously detect the fall signal, the user can trigger the secondary manual alarm by long pressing the emergency button on the smart bracelet (such as: the set time is 5 seconds) to ensure that an alarm can be issued in time in case of emergency.

[0016] In the above solution, a machine learning algorithm is adopted to learn and analyze the user's daily behavior data, so as to optimize the detection accuracy. The algorithm can identify and filter routine actions, such as walking, running, sitting down, etc., to reduce the occurrence of false alarms. At the same time, the system also combines an acceleration data fusion algorithm to fuse the acceleration data of the smart bracelet and the mobile terminal to improve the accuracy of fall detection.

[0017] In the above solution, the Bluetooth Low Energy protocol is adopted for data transmission between the smart bracelet and the mobile terminal to extend the device battery life. The alarm information is first sent from the smart bracelet to the mobile terminal via Bluetooth, and then uploaded to the cloud alarm platform by the mobile terminal through 4G / 5G or Wi-Fi network. After receiving the alarm information, the cloud alarm platform will immediately notify the preset emergency contacts by means of SMS, phone call, APP push, etc., to ensure that the user can get rescue in the shortest time.

[0018] In the above solution, through multi-device collaborative detection, machine learning algorithm optimization, and a composite positioning system, the accuracy and reliability of fall detection have been significantly improved, providing a safer and more convenient life guarantee for the elderly user group. Brief Description of the Drawings

[0019] Figure 1 It is a schematic structural diagram of an elderly fall alarm system based on a smart bracelet device; Figure 2 It is a schematic structural diagram of the smart bracelet of an elderly fall alarm system based on a smart bracelet device; Figure 3 It is a schematic structural diagram of the dual-device detection process of an elderly fall alarm method based on a smart bracelet device; Figure 4 It is a schematic structural diagram of the alarm signal transmission timing diagram of an elderly fall alarm method based on a smart bracelet device.

[0020] The marks in the drawings are: 1, smart bracelet; 101, emergency button. Detailed Embodiment

[0021] An embodiment of the present invention provides an elderly fall alarm system based on a smart bracelet device, including a smart bracelet 1, a mobile terminal, and a cloud alarm platform; The smart bracelet 1 is built-in with an accelerometer and a low-power wireless communication module, and the mobile terminal is built-in with an accelerometer, a GPS positioning module, a wide-area network communication module, and a short-range communication module. Specifically, the smart bracelet 1 is built-in with a three-axis accelerometer, and the low-power wireless communication module of the smart bracelet supports Bluetooth and Wi-Fi networks. The mobile terminal is a commonly used smartphone of the user. The GPS positioning module adopts a redundant design of Beidou positioning. The wide-area network communication module and the short-range communication module of the mobile terminal support mobile networks such as Bluetooth, 4G / 5G, or Wi-Fi direct connection. The smart bracelet 1 and the mobile terminal jointly monitor the posture changes of the user. The user is an elderly person. The cloud alarm platform is used to convey the alarm information of the smart bracelet 1 and the mobile terminal to the preset emergency contacts; Due to the high penetration rate of mobile terminals as necessities of life, when designing the smart bracelet 1, the communication capabilities of mobile terminals are fully utilized. Since mobile terminals are products that everyone carries with them, most of the functions of the smart bracelet 1 are placed on the mobile terminal as much as possible. The smart bracelet 1 only retains the acceleration sensor and the low-power wireless communication module to communicate with the mobile terminal, which reduces the cost and power consumption of the smart bracelet 1 itself, and at the same time improves the universality of the system.

[0022] In this embodiment, the low-power wireless communication module of the smart bracelet supports Bluetooth and Wi-Fi networks, and the mobile terminal is a smartphone for daily use. The wide-area network communication module and the short-range communication module of the mobile terminal support mobile networks, Bluetooth or Wi-Fi networks.

[0023] The present invention also provides an elderly fall alarm method based on a bracelet device. The alarm method includes the following steps: Step 1: The smart bracelet 1 and the mobile terminal detect acceleration data through the accelerometer. When the detected acceleration data exceeds a preset threshold (this preset threshold will be dynamically adjusted by the system in combination with user behavior analysis), it is determined that the user has fallen; Step 2: The mobile terminal provides accurate location information and wirelessly transmits it to the cloud alarm platform for first-level alarm. At the same time, the smart bracelet 1 vibrates and the buzzer makes a sound for reminder; Step 3: The cloud alarm platform conveys the first-level alarm information to the preset emergency contacts.

[0024] The alarm information is first sent from the smart bracelet 1 to the mobile terminal via Bluetooth. The mobile terminal uploads the alarm information to the cloud alarm platform through 4G / 5G or Wi-Fi network. Then, after receiving the alarm information, the cloud alarm platform immediately notifies the preset emergency contacts by means of text messages, phone calls, APP push, etc. During the transmission process of the alarm information, it does not pass through a third-party server to ensure data security. The cloud alarm platform notifies the preset emergency contacts by means of text messages, phone calls, APP push, etc.

[0025] In this embodiment, in Step 1, the smart bracelet 1 and the mobile terminal use an acceleration data fusion algorithm to determine whether the user has fallen.

[0026] A machine learning algorithm module is set in the smart bracelet 1 and the mobile terminal. The acceleration algorithm uses a machine learning algorithm to optimize the detection accuracy based on the user's daily behavior data and reduce false alarms. The algorithm can identify and filter regular actions, such as walking, running, sitting, etc.; It can also use fusion algorithms such as Kalman filtering and weighted average to process the data of both to obtain a more accurate fall detection result.

[0027] In this embodiment, the acceleration data fusion algorithm includes the following steps: S1. Denoise the acceleration data collected by the smart bracelet 1 and the mobile terminal to reduce interference, and then synchronize the data in time to ensure that the data collected by the smart bracelet 1 and the mobile terminal at the same time point can correspond; S2. Assign weights to the detection data of the smart bracelet 1 and the mobile terminal, and calculate the fused acceleration data according to the weights.

[0028] In this embodiment, the method of assigning weights is: W1 + W2 = 1; Among them, W1 represents the weight of the smart bracelet 1, W2 represents the weight of the mobile terminal, and 1 represents the total weight of the smart bracelet 1 and the mobile terminal; The calculation method of the acceleration data is: Afused = W1 * Abracelet + W2 * Aphone; Among them, Afused represents the fused acceleration data, Abracelet represents the acceleration data collected by the smart bracelet 1, and Aphone represents the acceleration data collected by the mobile terminal.

[0029] In practical applications, the values of the weights W1 and W2 can be continuously adjusted according to the user's feedback and detection results to optimize the detection accuracy. At the same time, more feature information (such as angular velocity, attitude angle, etc.) can be introduced to improve the accuracy of fall detection.

[0030] In this embodiment, the preset threshold in step one is: the acceleration value detected by the smart bracelet exceeds the preset value and lasts for a certain period of time, and the morphological change conforms to the fall characteristics. At the same time, the mobile terminal detects that the acceleration change exceeds the preset value within the preset time threshold (for example, 3 seconds), and the spatial attitude changes suddenly. The time and preset value set here will be analyzed and adjusted by deep learning according to the user's behavior.

[0031] In this embodiment, a barrier-free assistance background service is installed on the mobile terminal. When the system is used, it is granted permission to run in the background. The barrier-free service will handshake with the APP installed on the mobile terminal through AIDL in real time. The alarm method further includes the following steps: Step one: After the APP disconnects the handshake with the barrier-free service, the barrier-free service of the mobile terminal turns on Bluetooth or Wi-Fi scanning and listens to Bluetooth or Wi-Fi broadcasts in real time; Step two: After the barrier-free service of the mobile terminal listens to the help broadcast of the smart bracelet, it immediately reports to the cloud alarm platform and establishes a connection communication (such as Bluetooth) with the smart bracelet, and at the same time feeds back the successful result of reporting to the cloud alarm platform; When the report to the cloud alarm platform is successful, the vibration and buzzer sound reminder of the smart bracelet 1 are cancelled.

[0032] Specifically as follows Method a: If the mobile terminal APP is in the open state, the smart bracelet 1 communicates with the mobile terminal APP via Bluetooth, directly sends a cloud-based help signal, the mobile terminal APP and the smart bracelet 1 synchronously display the help signal, and the buzzer reminder of the smart bracelet 1 is cancelled.

[0033] Method b: If the mobile terminal APP is not open, but the mobile terminal is in the powered-on state, use the mobile terminal's accessibility service to monitor the Bluetooth broadcast signal sent by the smart bracelet 1. Once a help broadcast is received, the accessibility service uploads the cloud-based help signal via other network methods (such as 4G / 5G, Wi-Fi). After success, the accessibility service establishes a Bluetooth connection with the smart bracelet 1, synchronizes the help signal, and cancels the buzzer reminder of the smart bracelet 1.

[0034] Method c: If the mobile terminal APP is not open, but the mobile terminal is in the powered-on state, when there is no response to the Bluetooth help signal sent by the smart bracelet 1 for 30 seconds, the smart bracelet 1 will establish a Wi-Fi direct connection with the mobile terminal's accessibility service and synchronize the help signal. The accessibility service uploads the cloud-based help signal via other network methods (such as 4G / 5G, Wi-Fi). After success, the accessibility service establishes a Bluetooth connection with the smart bracelet 1, synchronizes the help signal, and cancels the buzzer reminder of the smart bracelet 1.

[0035] Method d: If the mobile terminal is not around or is in a shutdown, flight mode (Bluetooth and Wi-Fi are off), no-network, etc. state, the smart bracelet 1 will continuously emit a buzzer reminder to attract the attention of people around and seek help; The alarm information is first received by the mobile terminal's accessibility service or APP from the help information of the smart bracelet 1. The mobile terminal uploads the alarm information to the cloud alarm platform via 4G / 5G or Wi-Fi network. Then, after the cloud alarm platform receives the alarm information, it immediately notifies the preset emergency contacts via text message, phone call, APP push, etc. During the transmission of the alarm information, it does not pass through a third-party server to ensure data security. Since the application-side APP is not blocked in the static state, the smart bracelet 1 can also communicate in the silent state of the mobile terminal to ensure that users can get timely rescue.

[0036] In this embodiment, an emergency button is installed on the smart bracelet 1. Long pressing the emergency button is used to trigger a secondary manual help request; specifically, a pressure sensor is linked with a timer, and continuous pressing for a preset duration triggers an alarm. A tactile feedback device is installed on the smart bracelet 1 to provide vibration feedback to confirm the alarm status. When the elderly person's mobile terminal is not carried with them or the fall signal is not synchronously detected, the secondary manual alarm can be triggered through the emergency button on the smart bracelet 1.

[0037] All technical features in this embodiment can be freely combined according to actual needs.

[0038] The above embodiments are preferred implementation solutions of the present invention. In addition, the present invention can also be implemented in other ways. Any obvious replacement without departing from the concept of the technical solution of the present invention is within the protection scope of the present invention.

Claims

1. An elderly fall alarm system based on a bracelet device, characterized in that, It includes a smart bracelet, a mobile terminal, and a cloud alarm platform; The smart bracelet is built - in with an accelerometer and a low - power wireless communication module. The mobile terminal is built - in with an accelerometer, a GPS positioning module, a wide - area network communication module, and a short - range communication module. The smart bracelet and the mobile terminal communicate wirelessly and are used to jointly monitor the user's posture changes. The cloud alarm platform is used to convey the alarm information of the smart bracelet and the mobile terminal to the preset emergency contacts.

2. The elderly fall alarm system based on a bracelet device according to claim 1, characterized in that, The accelerometer is a three - axis accelerometer. The low - power wireless communication module of the smart bracelet supports Bluetooth and Wi - Fi networks. The mobile terminal is a smartphone for daily use. The wide - area network communication module and the short - range communication module support Bluetooth, mobile networks, or Wi - Fi networks.

3. The elderly fall alarm system based on bracelet devices according to claims 1 - 2 further includes the following steps: Step 1: The smart bracelet and the mobile terminal detect acceleration data through the accelerometer. When the detected acceleration data exceeds the preset threshold, it is determined that the user has fallen. Step 2: The mobile terminal provides accurate location information and wirelessly transmits it to the cloud alarm platform for primary alarm. At the same time, the smart bracelet gives vibration and buzzer sound reminders. Step 3: The cloud alarm platform conveys the primary alarm information to the preset emergency contacts.

4. The method for elderly fall alarm based on bracelet device according to claim 3, wherein, In Step 1, the smart bracelet and the mobile terminal use an acceleration data fusion algorithm to determine whether the user has fallen.

5. The method for elderly fall alarm based on bracelet device according to claim 4, characterized in that, The acceleration data fusion algorithm includes the following steps: S1: Denoise the acceleration data collected by the smart bracelet 1 and the mobile terminal, and then synchronize the data in time to ensure that the data collected by the smart bracelet 1 and the mobile terminal at the same time point can correspond. S2: Assign weights to the detection data of the smart bracelet 1 and the mobile terminal, and calculate the fused acceleration data according to the weights.

6. The method for alarming the fall of the elderly based on the bracelet device according to claim 5, wherein, The method of assigning weights is: W1 + W2 = 1; Where, W1 represents the weight of the smart bracelet 1, W2 represents the weight of the mobile terminal, and 1 represents the total weight of the smart bracelet 1 and the mobile terminal; The calculation method of the acceleration data is: Afused = W1 * Abracelet+W2 * Aphone; Where, Afused represents the fused acceleration data, Abracelet represents the acceleration data collected by the smart bracelet 1, and Aphone represents the acceleration data collected by the mobile terminal.

7. The method for alarming the fall of the elderly based on the bracelet device according to claim 6, wherein, The preset threshold in Step 1 is: The acceleration value detected by the smart bracelet exceeds the preset value and lasts for a certain period of time, and the morphological change conforms to the fall characteristics. At the same time, the mobile terminal detects that the acceleration change exceeds the preset value within the preset time threshold, and the spatial posture suddenly changes. The preset value will be adjusted through deep learning according to user behavior analysis.

8. The method for elderly fall alarm based on a bracelet device according to claim 3, characterized in that, An accessible assistance background service is installed on the mobile terminal and shakes hands with the APP installed on the mobile terminal in real - time through AIDL. The alarm method further includes the following steps: Step 1: After the APP disconnects the handshake with the accessible service, the accessible service of the mobile terminal turns on Bluetooth or Wi - Fi scanning and listens to Bluetooth or Wi - Fi broadcasts in real - time; Step 2: After the accessibility service of the mobile terminal monitors the help broadcast of the smart bracelet, it immediately reports to the cloud alarm platform, establishes a connection and communication with the smart bracelet, and at the same time feeds back the successful result of reporting to the cloud alarm platform; When the report to the cloud alarm platform is successful, the vibration and buzzer sound reminder of the smart bracelet are cancelled.

9. The method for elderly fall alarm based on a bracelet device according to claim 8, characterized in that, An emergency button is installed on the smart bracelet, and long pressing the emergency button is used to trigger secondary manual help.

10. The method for elderly fall alarm based on a bracelet device according to claim 9, characterized in that, The cloud alarm platform notifies the preset emergency contacts by means of text messages, phone calls, APP push, etc.

Images