Automatic falling alarm and emergency door opening rescue system for elderly living alone
By using wireless AI cameras and millimeter-wave radar to detect falls, the system automatically opens doors and sounds a loud alarm, enabling multi-party rescue coordination. This solves the problem of elderly people living alone having difficulty entering their homes after a fall, and improves rescue efficiency and accuracy.
Patent Information
- Application Number
- CN202511119747.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-11
- Publication Date
- 2025-11-11
AI Technical Summary
When elderly people living alone fall, current technology makes it difficult to quickly open doors for rescue, delaying the golden rescue time and hindering communication, which can lead to the worsening of injuries.
By combining wireless AI cameras and millimeter-wave radar, it automatically detects falls and opens doors, and integrates loud alarms with multi-party rescue coordination to achieve real-time information sharing and collaborative rescue.
Quickly removing obstacles to entry, providing initial assistance through neighborly mutual aid, and coordinating rescue efforts from multiple parties significantly shorten rescue time and improve the safety of elderly people living alone.
Smart Images
Figure CN120932371A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the fields of smart home, health monitoring and emergency rescue technology, and specifically to an automatic alarm and emergency door opening rescue system for elderly people living alone that is linked to a multimodal perception and AI anti-fall smart door lock. Background Technology
[0002] As the aging society continues to intensify, the number of elderly people living alone has increased significantly, making their safety a growing concern and a pressing social issue. Among the many threats to the health and safety of the elderly, falls are particularly common and have serious consequences, being a major cause of disability and even death among the elderly. Especially for those over 65, a single accidental fall can often lead to serious injuries such as hip fractures, significantly reducing their quality of life and potentially endangering their lives. Therefore, when such accidents occur, the ability to receive timely and effective assistance within the critical golden rescue time directly impacts the elderly person's safety and subsequent rehabilitation. This underscores the extreme importance of establishing efficient and reliable emergency response mechanisms and promoting the widespread use of smart assistive devices such as emergency call devices to ensure that elderly people living alone can receive necessary assistance in emergencies.
[0003] The existing technology has the following main problems:
[0004] Question 1. Fall detection alarm: Existing smartwatches, wearable devices or indoor sensors can detect falls and issue alarms (such as notifying family members or platforms).
[0005] Problem 2. Barriers to Entry: When an elderly person falls and loses mobility or consciousness, rescuers (ambulances, neighbors, community workers) often cannot immediately enter the house upon arrival. They need to wait for family members with keys, contact a locksmith, or force the door open, which seriously delays the precious "golden rescue time" (generally considered to be within 4-6 minutes), potentially leading to worsening injuries or even death.
[0006] Problem 3. Poor or unclear information communication: Before arriving at the scene, rescuers lack understanding of the situation (such as the elderly person's posture, whether there is bleeding, and level of consciousness), making it difficult to formulate the optimal rescue plan. Communication between on-site personnel and remote medical guidance or family members may also be difficult. Summary of the Invention
[0007] To address the problems existing in the prior art, this invention provides a system that utilizes a wireless AI camera, millimeter-wave radar, and a smart door lock working in tandem. Upon detecting a fall by an elderly person, the system automatically opens the door, triggers a loud alarm, and coordinates with multiple rescue teams. The aim is to:
[0008] (1) After confirming that the elderly person has suffered a serious fall (such as being unable to get up on their own or call for help), the door will be automatically unlocked and opened to remove the physical obstacles for rescuers to enter.
[0009] (2) The smart lock immediately triggers a loud alarm on site, attracting nearby people (neighbors, passersby) to provide initial assistance as soon as possible.
[0010] (3) At the same time, it automatically notifies and coordinates multiple rescue forces (120 emergency center, community hospital, family members, social workers / caregivers) through wireless network, and shares real-time on-site information (audio and video, intercom), and views and communicates to achieve efficient collaborative rescue.
[0011] (4) Minimize the time from falling to receiving effective assistance and improve the success rate of rescue.
[0012] The technical solution adopted by this invention to solve its technical problem is:
[0013] An automatic alarm and emergency door-opening rescue system for elderly people living alone who have fallen includes a sensing layer, a core processing and execution layer, and an alarm and communication linkage layer. Among them:
[0014] The perception layer includes sensors deployed and installed in the main activity areas of the elderly. These sensors include wireless AI cameras and / or millimeter-wave radar sensors.
[0015] The core processing and execution layer includes core devices, which include an AI processing unit, a communication module, and a smart lock with a door lock controller. The AI processing unit includes a data processing and AI analysis module, a fall status judgment module, a stillness judgment module, a standing failure judgment module, an active SOS judgment module, a high-volume alarm module, and an emergency action execution module. The communication module includes a network connection module. The smart lock includes an AI anti-fall smart lock automatic opening module.
[0016] The alarm and communication linkage layer includes a multi-party alarm notification module and a real-time information sharing platform. Specifically, when the door opens and a loud alarm is triggered, the smart lock automatically sends emergency alarm information to multiple pre-set recipients via its network connection. This is typically done through a dedicated app push notification, SMS, or telephone call. Recipients include 120 emergency services, linked family members' mobile phones, community hospitals / community health service centers, social workers / caregivers, and property management / community management platforms.
[0017] Upon alarm triggering, the real-time information sharing platform automatically activates the status display of cameras or radar in the relevant area. All notified rescue parties can log in to the platform via a dedicated mobile app or computer web interface.
[0018] The present invention also has the following additional technical features:
[0019] As a further specific optimization of the technical solution of the present invention: In the perception layer, the wireless AI camera is used to provide visual information, and the millimeter-wave radar sensor is used to provide motion, breathing, and posture information. Both sensors are unaffected by light, and they can be used complementaryly. The sensors are connected to the core device wirelessly.
[0020] As a further specific optimization of the technical solution of the present invention: in the core processing and execution layer, the data processing and AI analysis module is used to receive real-time data streams from the camera / radar, and uses a fall detection and status judgment AI algorithm.
[0021] As a further specific optimization of the technical solution of the present invention: in the core processing and execution layer, the fall state judgment module, the fall detection and state judgment AI algorithm, based on sensor fusion data, after identifying a "fall" event, needs to further determine whether it is a serious fall requiring emergency intervention. The triggering conditions include:
[0022] The static judgment module indicates that if an elderly person falls and there is no significant movement of their body for more than 5 seconds, the situation will be assessed.
[0023] The module for determining failure to stand up indicates that if an elderly person falls and fails to stand up on their own or move effectively to a safe position within 10 seconds, it indicates that the elderly person has not been able to do so.
[0024] The active distress call detection module allows the system to recognize clear verbal distress calls issued by an elderly person after they have fallen.
[0025] The high-volume alarm module, either built into or connected to the smart lock, immediately activates and plays a preset distress message in a loop, attracting the attention of nearby neighbors or passersby and encouraging them to enter and check on the elderly person as soon as possible.
[0026] The emergency action execution module activates the smart lock of the door lock controller immediately upon confirming a serious fall.
[0027] The network connectivity module supports Wi-Fi and / or cellular networks, ensuring reliable communication even when the network environment changes.
[0028] The AI-powered anti-fall smart lock's automatic door opening module automatically unlocks and drives the smart lock motor to open the door, ensuring unobstructed access for rescue personnel.
[0029] As a further specific optimization of the technical solution of the present invention: in the alarm and communication linkage layer,
[0030] The functions of the real-time information sharing platform include sharing real-time video / status, real-time two-way intercom, and location and status updates.
[0031] In the shared real-time video / status, authorized users can view video footage of the fall area or attitude diagrams / vital sign data generated by radar in real time to understand the situation on site.
[0032] The real-time two-way intercom platform supports multi-party real-time voice communication, allowing rescue personnel to enter the building immediately to provide comfort and initial injury assessment. Remote rescue teams can communicate with on-site personnel to provide professional guidance. All parties can collaborate and communicate to develop the optimal rescue and medical transport plan.
[0033] During the location and status updates, the App / platform can display the estimated arrival time of the rescue unit.
[0034] Compared with the prior art, the advantages of this invention are:
[0035] Advantage 1. "Automatic door opening" eliminates key obstacles, and a loud alarm is sounded immediately on-site to alert surrounding people to participate in the rescue: This is the most core innovation that distinguishes it from existing technologies. After confirming a serious fall, the system automatically and physically opens the door, completely solving the fatal delay problem of rescuers being unable to enter the room upon arrival. The immediate alarm on-site wins the most precious "golden time" for rescue.
[0036] Advantage 2. High-volume on-site alarm drives neighborhood mutual assistance: The automatically triggered high-volume voice alarm can effectively attract the attention of nearby people. Taking advantage of the "geographical proximity", it can drive neighbors or passers-by to become the first responders and provide the fastest initial assistance and psychological comfort, which is crucial before professional rescue arrives.
[0037] Advantage 3. Multimodal perception improves accuracy: Combining AI cameras (visual behavior recognition) and millimeter-wave radar (motion / vital sign monitoring, privacy friendly), sensor fusion technology improves the accuracy and reliability of fall detection and status judgment, reducing false alarms and missed alarms.
[0038] Advantage 4. AI intelligent status judgment: Introducing multi-condition judgment logic (stationary timeout, failure to stand, active call for help) to effectively distinguish between minor trips (which can be handled by oneself) and serious falls that require emergency intervention, avoiding unnecessary door opening and alarms.
[0039] Advantage 5. Highly efficient multi-party collaborative rescue:
[0040] One-click multi-channel alarm: Simultaneously notify all key parties such as 120, family members, community hospitals, and social workers, avoiding long information transmission chains and low efficiency.
[0041] Real-time transparent information sharing: Real-time video (or status information) and two-way communication between multiple parties are shared via App / Web platforms, enabling: remote professionals (120 doctors) to provide "on-site" guidance to rescue personnel; family members to stay informed and reduce anxiety; and all rescue forces to collaborate efficiently, optimizing the rescue process.
[0042] Advantage 6. Dual network redundancy ensures communication: The smart lock supports Wi-Fi and cellular networks, ensuring that critical alarm and communication functions remain available during network fluctuations or Wi-Fi failures, thus improving system reliability.
[0043] Advantage 7. Privacy protection considerations: Prioritize the use of millimeter-wave radar or restrict camera use in non-essential areas / times, and explicitly authorize access when the App accesses the camera, thus balancing security and privacy. Attached Figure Description
[0044] Figure 1 This is a schematic diagram of the logical structure of the present invention.
[0045] Figure 2 This is a schematic diagram of the rescue process according to the present invention. Detailed Implementation
[0046] Exemplary embodiments of the present invention will now be described in more detail with reference to the accompanying drawings.
[0047] Example 1
[0048] An automatic alarm and emergency door opening rescue system for elderly people living alone who fall is based on the organic combination of four elements: multimodal perception, AI intelligent judgment, intelligent door lock linkage, and multi-party collaborative communication.
[0049] This system comprises a perception layer, a core processing and execution layer, and an alarm and communication linkage layer. The perception layer includes sensors, which are deployed and installed in the main activity areas of the elderly (living room, bedroom, kitchen, and bathroom).
[0050] The sensors include wireless AI cameras and / or millimeter-wave radar sensors.
[0051] The wireless AI camera provides visual information, while the millimeter-wave radar sensor provides motion, breathing, and posture information. Both sensors are unaffected by light conditions and can be used complementaryly. The wireless AI camera and / or millimeter-wave radar sensor connect to the core device wirelessly (e.g., via Wi-Fi, Zigbee, Bluetooth).
[0052] The core processing and execution layer includes core equipment, which comprises an AI processing unit, a communication module, and a smart lock (i.e., an "AI anti-fall smart lock") with a door lock controller. The AI processing unit includes a data processing and AI analysis module, a fall status judgment module, a stillness judgment module, a standing failure judgment module, an active SOS judgment module, a high-volume alarm module, and an emergency action execution module. The communication module includes a network connection module. The smart lock includes an AI anti-fall smart lock automatic door opening module.
[0053] The data processing and AI analysis module receives real-time data streams from cameras / radar and uses fall detection and status assessment AI algorithms.
[0054] The fall status assessment module uses an AI algorithm based on sensor fusion data to detect and assess falls. After identifying a "fall" event, it further determines whether the fall is a serious one requiring emergency intervention. Triggering conditions include (any one of these conditions will trigger the assessment):
[0055] The static judgment module determines whether the elderly person has not made any significant body movement (such as attempting to get up) for more than 5 seconds after falling (the time can be set).
[0056] The module for determining failure to stand up indicates that if an elderly person falls and fails to stand up on their own or move to a safe position within 10 seconds (the time can be set), the system will detect this.
[0057] The active distress call judgment module allows the system to recognize clear verbal distress commands issued by the elderly after a fall (such as preset keywords like "help", "I fell down", "need help").
[0058] The high-volume alarm module, either built into or connected to the smart lock, immediately activates and plays a preset distress message (such as "Emergency! An elderly person in XX household has fallen and urgently needs help!") to attract the attention of nearby neighbors or passersby, encouraging them to enter and check on the elderly person as soon as possible.
[0059] The emergency action execution module, once a serious fall is confirmed, immediately activates the smart lock on the door lock controller (i.e., the "AI anti-fall smart lock"):
[0060] The network connectivity module supports Wi-Fi and / or cellular mobile networks (such as 4G / 5G via SIM card) to ensure reliable communication even when the network environment changes.
[0061] The AI anti-fall smart lock's automatic door opening module automatically unlocks and drives the smart lock motor to open the door (e.g., to 90 degrees or a preset safe angle), ensuring unobstructed access for rescue personnel.
[0062] The alarm and communication linkage layer includes a multi-party alarm notification module and a real-time information sharing platform. Specifically, when the door opens and a loud alarm is triggered, the smart lock automatically sends emergency alarm information to multiple preset recipients via its network connection. This is typically done through a dedicated app push notification, SMS, or telephone call. Recipients include 120 emergency medical services (providing precise address and homeowner information), linked family member mobile phones (one or more), community hospitals / community health service centers, social workers / caregivers, and property / community management platforms (optional).
[0063] Upon alarm triggering, the real-time information sharing platform (App / Web) automatically activates the status display of cameras (within privacy limits) or radar in the relevant area. All notified rescue parties (120 ambulance, community doctor, family members, social workers) can log in to the platform via a dedicated mobile app or computer web interface.
[0064] The functions of the real-time information sharing platform include sharing real-time video / status, real-time two-way intercom, and location and status updates.
[0065] In the shared real-time video / status, authorized users can view the video footage of the fall area in real time (if enabled and compliant) or the attitude diagram / vital sign data generated by radar to understand the situation on site.
[0066] The real-time two-way intercom platform supports multi-party real-time voice communication, allowing rescue personnel to enter the house immediately (such as arriving neighbors or passersby) to provide comfort and initial injury assessment. Remote rescue parties (120 command center, community doctor, family members) can communicate with on-site personnel (neighbors, passersby) to provide professional guidance (such as "do not move the elderly person unnecessarily" or "check for bleeding"). All parties can collaborate and communicate to develop the optimal rescue and medical transport plan.
[0067] During the location and status update, the App / platform can display the estimated arrival time of rescue units (such as 120 vehicles).
[0068] Example 2
[0069] Taking an elderly person living alone who falls in the kitchen as an example, the rescue process is as follows:
[0070] 1. An elderly person slipped and fell in the kitchen, hitting the back of their head on the ground. They immediately lost the ability to move and became confused.
[0071] 2. The millimeter-wave radar (or AI camera) in the kitchen detects sudden changes in height and abnormal falling postures, and transmits the data in real time to the AI anti-fall smart lock on the door.
[0072] 3. The smart lock's AI algorithm analyzes the data and determines it as a "fall" event. Within the subsequent 10 seconds (the time can be set), no signal is detected indicating that the elderly person attempted to get up or move effectively (meeting the "failure to stand" condition).
[0073] 4. The smart lock immediately and automatically activates the door lock motor to unlock and open the entrance door. Simultaneously, it triggers the built-in loudspeaker, repeatedly broadcasting: "Emergency! Mr. Zhang from apartment 201 has fallen and urgently needs help! Please ask neighbors for assistance!"
[0074] 5. The neighbor across the hall heard a continuous high-pitched alarm and went out to check. They found the door to apartment 201 wide open. They quickly went into the kitchen to check on the elderly person, offered some comfort, and took precautions to prevent further injury.
[0075] 6. The smart lock synchronizes via 4G network:
[0076] Send an alarm message to the 120 emergency center (including address, homeowner's name, age, and preset medical history).
[0077] The system sends strong alerts to the mobile apps of the elderly person's children.
[0078] An alarm message was sent to the community hospital and the social worker responsible for the area.
[0079] 7. Children of the elderly, community doctors, 120 emergency operators, social workers, etc. can immediately access the system platform through a mobile app.
[0080] Once authorized (or only displaying the radar attitude map), you can see the real-time situation in the kitchen area (a neighbor is comforting an elderly person).
[0081] The 120 doctor established a two-way call with the neighbors on site through the platform: "This is Dr. Li from 120. Please do not move the elderly man. Check if there is any bleeding in his head and if he is conscious. Please tell me how he is breathing." The neighbors observed and answered according to the instructions.
[0082] The community doctor also connected to the call and provided the elderly person's past medical history information.
[0083] Family members can hear the conversations at the scene and speak with doctors or neighbors to understand the situation and comfort the elderly.
[0084] The platform shows that the 120 ambulance is expected to arrive in 5 minutes.
[0085] 8. Under professional guidance, neighbors stayed with the elderly until the ambulance arrived. Upon arrival, since the door was already open, medical personnel entered directly, quickly examined and treated the elderly person during this critical time, and then transported them to the hospital.
[0086] In summary, through Embodiments 1 and 2, this invention creatively integrates multimodal perception technology, AI decision-making, intelligent door lock control, and a multi-party collaborative communication platform, fundamentally solving the critical bottleneck problem of "difficulty in getting into the door" for rescue after an elderly person falls. By automatically opening the door, providing immediate rescue with a high-volume on-site alarm, enabling one-click multi-channel notifications, and facilitating real-time information sharing and intercom, a closed-loop rescue system with extremely fast response, transparent information, and high collaborative efficiency is constructed. This significantly shortens rescue time, improves the safety of elderly people living alone, and has significant social value and market potential.
[0087] The above detailed description of embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention.
Claims
1. An automatic alarm and emergency door-opening rescue system for elderly people living alone who have fallen, characterized in that: It includes a perception layer, a core processing and execution layer, and an alarm and communication linkage layer; among which: The perception layer includes sensors deployed and installed in the elderly person's main activity areas. These sensors include wireless AI cameras and / or millimeter-wave radar sensors. The core processing and execution layer includes core devices, which include an AI processing unit, a communication module, and a smart lock with a door lock controller. The AI processing unit includes a data processing and AI analysis module, a fall status judgment module, a stillness judgment module, a standing failure judgment module, an active SOS judgment module, a high-volume alarm module, and an emergency action execution module. The communication module includes a network connection module. The smart lock includes an AI anti-fall smart lock automatic door opening module. The alarm and communication linkage layer includes a multi-party alarm notification module and a real-time information sharing platform. When the door opening and high-volume alarm are triggered, the smart lock automatically sends emergency alarm information to multiple preset recipients through its network connection. These recipients are typically notified via a dedicated app push, SMS, or telephone. Recipients include 120 emergency centers, linked family members' mobile phones, community hospitals / community health service centers, social workers / caregivers, and property / community management platforms. After an alarm is triggered, the real-time information sharing platform automatically activates the status display of cameras or radar in the relevant area; all notified rescue parties can log in to the platform via a dedicated mobile app or computer web interface.
2. The automatic alarm and emergency door opening rescue system for elderly people living alone who have fallen, as described in claim 1, is characterized in that: In the perception layer, the wireless AI camera provides visual information, and the millimeter-wave radar sensor provides motion, breathing, and posture information. It is unaffected by light, and the two sensors can be used in a complementary manner. The sensors are connected to the core device wirelessly.
3. The automatic alarm and emergency door opening rescue system for elderly people living alone who have fallen, as described in claim 1, is characterized in that: In the core processing and execution layer, the data processing and AI analysis module is used to receive real-time data streams from cameras / radar, and uses fall detection and status judgment AI algorithms.
4. The automatic alarm and emergency door opening rescue system for elderly people living alone who have fallen, as described in claim 3, is characterized in that: The core processing and execution layer includes a fall status judgment module. The fall detection and status judgment AI algorithm, based on sensor fusion data, identifies a "fall" event and then further determines whether it is a serious fall requiring emergency intervention. Triggering conditions include: The static judgment module detects cases where an elderly person's body does not move significantly for more than 5 seconds after falling. The module for determining failure to stand up indicates that if an elderly person falls and fails to stand up on their own or move effectively to a safe position within 10 seconds; The active distress call judgment module allows the system to recognize clear verbal distress calls issued by an elderly person after falling. The high-volume alarm module, which is built into or connected to the smart lock, immediately activates and plays a preset distress message in a loop to attract the attention of nearby neighbors or passersby and encourage them to enter and check on the elderly person as soon as possible. The emergency action execution module activates the smart lock of the door lock controller immediately upon confirming a serious fall. A network connectivity module that supports Wi-Fi and / or cellular networks ensures reliable communication even when the network environment changes. The AI-powered anti-fall smart lock's automatic door opening module automatically unlocks and drives the smart lock motor to open the door, ensuring unobstructed access for rescue personnel.
5. The automatic alarm and emergency door opening rescue system for elderly people living alone who have fallen, as described in claim 1, is characterized in that: In the alarm and communication linkage layer The functions of the real-time information sharing platform include sharing real-time video / status, real-time two-way intercom, and location and status updates. In the shared real-time video / status, authorized users can view video footage of the fall area or attitude diagrams / vital sign data generated by radar in real time to understand the situation on site; The real-time two-way intercom platform supports real-time voice communication between multiple parties, allowing rescue personnel to enter the house immediately to provide comfort and initial injury assessment; remote rescue parties can communicate with on-site personnel to provide professional guidance; and all parties can collaborate to develop the optimal rescue and medical transport plan. During the location and status updates, the App / platform can display the estimated arrival time of the rescue unit.