A multi-mode integrated elevator accessibility intelligent human-computer interaction system
By introducing millimeter-wave sensing modules and biometric technology into elevators, combined with IC cards, an accessible elevator experience for people with mobility impairments and visual impairments has been achieved, improving elevator safety and home integration capabilities, and solving the problems of contactless elevator calling and home integration in existing elevators.
Patent Information
- Application Number
- CN202611190527.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-06
- Publication Date
- 2026-11-17
AI Technical Summary
The existing human-computer interaction methods of elevators are not user-friendly for people with mobility impairments and visual impairments, lack proactive safety precautions, cannot achieve contactless elevator calling and home integration, and safety risks are difficult to prevent.
By employing millimeter-wave sensing modules combined with IC cards and biometrics, environmental and behavioral recognition is achieved. This, combined with historical habits for adaptive elevator calling and linkage with home appliances, forms a fully intelligent and barrier-free system.
It has enabled barrier-free elevator access for people with mobility impairments and visual impairments, improved safety and home integration, and reduced safety risks and energy waste.
Smart Images

Figure CN123113449A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent human-computer interaction technology, specifically to a multi-mode integrated elevator barrier-free intelligent human-computer interaction system. Background Technology
[0002] Current elevator human-machine interaction methods heavily rely on physical buttons or touchscreens, requiring passengers to have normal physical and visual recognition abilities. This presents significant obstacles for people with mobility impairments, the visually impaired, and other vulnerable groups. While some elevators have introduced IC cards for access control or employed biometric recognition technology, their functions are typically limited to simple identity verification and floor locking, lacking the ability to proactively adapt to passengers' past elevator usage habits. Passengers still need to manually select floors, failing to achieve truly contactless elevator calling. Furthermore, the elevator car is a closed and unattended space, making it difficult for existing safety measures to proactively detect abnormal situations such as falls, salt spray buildup, pets entering, or electric vehicles entering. Potential safety risks are often only detected after an accident occurs, hindering early prevention and immediate response. In addition, although millimeter-wave radar technology has been used for fall detection in some areas, it has not yet been deeply integrated with elevator control systems and identity recognition systems, failing to form a proactive, barrier-free elevator calling and risk management closed loop. In the field of home elevators, there is a lack of effective linkage mechanism between elevators and home appliances in the house. When passengers return home or leave home, they still need to manually turn on or off the air conditioner, lights and other equipment one by one, and cannot enjoy the integrated smart home experience with the elevator as the hub.
[0003] To address this, a multi-mode integrated intelligent human-computer interaction system for elevator accessibility is proposed. Summary of the Invention
[0004] The purpose of this invention is to provide a multi-mode integrated elevator accessibility intelligent human-computer interaction system that combines millimeter wave sensing, IC card and biometrics to achieve full-chain intelligence and accessibility, including environmental and behavioral recognition, adaptive elevator calling based on historical habits, access control and linkage with home appliances.
[0005] The specific technical solution is as follows: A multi-mode integrated elevator accessibility intelligent human-computer interaction system includes: A millimeter-wave sensing module is used to transmit millimeter-wave signals to the elevator car and / or elevator hall area and receive echoes. By analyzing the echo signals, the module identifies the behavioral state and environmental characteristics of the target. The behavioral state includes at least a falling action, and the environmental characteristics include at least the presence of salt spray, pet targets, and electric vehicle targets. The identity recognition module integrates an IC card reader and a biometric recognition unit to obtain passenger identity information; The control module is communicatively connected to the millimeter-wave sensing module, the identity recognition module, the elevator controller, and the smart home gateway. The control module includes a historical record database and a linkage strategy unit. The control module is configured to: when the identity recognition module obtains a passenger's identity information, determine the passenger's frequently used floors based on historical elevator records associated with that identity information in the historical record database, and generate floor recommendation information including the frequently used floors in the no-elevator-control mode to push to the interactive interface; in the elevator-control mode, verify the floor permissions corresponding to the identity information and automatically generate an elevator call command to register the authorized floors corresponding to the permissions; and, When applied to home elevator scenarios, the linkage strategy unit, through the smart home gateway, sends control commands to the home appliances associated with the passenger when the control module generates an elevator call command or detects that a passenger has arrived at the target floor, so as to automatically adjust the operating status of the home appliances.
[0006] The above solution deeply integrates millimeter-wave environmental sensing, two-factor authentication, and elevator travel history, enabling the system to proactively detect various anomalies such as falls, salt spray, pets, and electric vehicles. Based on these anomalies, it triggers alarms, ventilation, or door closure prevention measures, significantly improving elevator safety and equipment protection. In elevator-free environments, passengers do not need to operate the system; it automatically pushes frequently used floors. In elevator-controlled environments, permission verification and floor registration are completed automatically, avoiding the tediousness of manual floor selection and permission verification, providing a truly barrier-free elevator experience for people with mobility impairments. In home elevators, calling or arriving events can instantly trigger home appliances, creating a comfortable environment before arriving home and automatically saving energy when leaving, making the elevator a natural entry point to smart homes.
[0007] The aforementioned multi-mode fusion elevator barrier-free intelligent human-computer interaction system includes a millimeter-wave sensing module comprising a millimeter-wave radar sensor and a signal processing unit. The signal processing unit generates a time-spectrum map or point cloud data from the echo signal and uses a trained deep learning model to classify the time-spectrum map or point cloud data to identify the fall action, salt spray presence, pet target, and electric vehicle target.
[0008] The above solution utilizes millimeter-wave radar sensors that do not rely on visible light, thus protecting privacy and enabling all-weather operation. By converting echoes into time-spectrum maps or point clouds, and combining them with deep learning models, it is possible to extract micro-motion features and spatial structure information of targets. This allows for reliable differentiation of human fall movements, scattering attenuation caused by salt spray, pet movement patterns, and the contour features of electric vehicles, significantly improving the recognition accuracy and anti-interference capabilities in enclosed metal environments such as elevators.
[0009] The aforementioned multi-mode integrated elevator barrier-free intelligent human-computer interaction system includes an IC card reading unit of the identity recognition module that supports contactless IC cards, and a biometric recognition unit that includes at least one of a fingerprint reader, a face recognition camera, and an iris reader. The identity recognition module is configured to output the identity information after dual authentication by the IC card reading unit and the biometric recognition unit.
[0010] The above solution combines contactless IC cards with biometric features such as fingerprints, faces, or irises to form a two-factor authentication system, effectively preventing the risk of card theft or counterfeiting and significantly improving the security of identity verification. At the same time, the contactless method reduces hygiene risks associated with physical contact, making the identity verification process more convenient and hygienic, aligning with the concept of accessibility design.
[0011] The aforementioned multi-mode integrated elevator barrier-free intelligent human-computer interaction system includes a historical record database that stores the elevator travel time, starting floor, and target floor for each passenger, and determines the frequently used floors based on elevator travel frequency or recent usage frequency statistics; the interactive interface includes a display screen and / or voice broadcast device installed in the elevator car.
[0012] The above solution involves recording detailed elevator travel time and floor information, and dynamically calculating frequently used floors based on frequency or recency. This ensures that the recommended floors closely match passengers' actual daily routines and can adaptively adjust even if travel habits change. The combination of display screens and voice announcements allows passengers with visual or auditory limitations to clearly obtain floor information, improving the inclusiveness and effectiveness of information delivery.
[0013] In the aforementioned multi-mode integrated elevator accessibility intelligent human-computer interaction system, in elevator control mode, the control module compares the identity information with a pre-stored floor permission table. When the permission corresponds to multiple authorized floors, the system automatically selects the most frequently used authorized floor based on the historical record database to generate the elevator call command, or provides the multiple authorized floors for passengers to choose from through the interactive interface.
[0014] The above solution allows the system to automatically determine the most frequently accessed floor based on a passenger's historical elevator usage habits and directly initiate an elevator call, saving the time spent searching floor by floor and pressing buttons for confirmation, thus significantly improving passage efficiency. At the same time, the manual selection function is retained, which satisfies daily habits and also takes into account the need to temporarily go to other authorized floors, making permission management both efficient and flexible.
[0015] The aforementioned multi-mode integrated elevator barrier-free intelligent human-computer interaction system, wherein the linkage strategy unit is configured as follows: when the control module generates an elevator call command or detects that a passenger has arrived at the target floor, if it is determined to be a homecoming scenario, an activation command is sent to the home appliances associated with the passenger through the smart home gateway; if it is determined to be a departure scenario, a deactivation command is sent; wherein, the homecoming scenario is determined based on the passenger entering the home elevator through the identity recognition module and the elevator starting operation, and the departure scenario is determined based on the passenger leaving the residence and the elevator reaching the designated exit floor; the home appliances include at least one of air conditioners, lamps, and curtains.
[0016] The above solution clearly distinguishes elevator usage into "coming home" and "leaving home" scenarios, ensuring that the triggering conditions for home appliance linkage accurately match the passenger's true intentions. When arriving home, air conditioning, lights, and other appliances are turned on in advance to create a comfortable welcoming environment; when leaving home, appliances are automatically turned off to reduce energy waste. Scenario determination directly utilizes identity recognition and elevator operating status, eliminating the need for additional sensors in the home, making the solution simple and highly reliable.
[0017] In the aforementioned multi-mode integrated elevator accessibility intelligent human-computer interaction system, the control module is further configured to perform at least one of the following actions: When the millimeter-wave sensing module detects the fall, it generates an alarm signal and sends it to a preset property management terminal or emergency contact terminal via the wireless communication module. When the presence of salt spray is detected to exceed a preset salt spray threshold, the elevator ventilation device or drying device is activated. The preset salt spray threshold is set between 10dB and 20dB based on the attenuation of the scattering intensity of the echo signal. When the pet or electric vehicle is detected, a voice warning message is output, and a command is sent to the elevator controller to prevent the elevator door from closing.
[0018] The above solution enables immediate alerts to property management or pre-set contacts after a fall, significantly shortening rescue response time and protecting the lives of elevator passengers. When the scattering attenuation caused by salt spray exceeds a set threshold, ventilation and drying automatically activate, effectively inhibiting moisture and salt deposition inside the elevator, delaying corrosion of metal components and electrical faults, and extending equipment lifespan. When a pet or electric vehicle is detected, a voice alert and prevents the door from closing prevent accidents caused by pets being left alone in the elevator or electric vehicles entering, achieving proactive risk control.
[0019] The aforementioned multi-mode integrated elevator barrier-free intelligent human-computer interaction system, wherein the millimeter wave sensing module operates in the frequency range of 57GHz to 64GHz, has a transmission power of less than 10dBm, and an effective detection distance of 0.1 meters to 10 meters; when the millimeter wave sensing module identifies the fall action, it determines it by detecting the height change rate and posture change of the human target within a time window of 0.5 seconds to 3 seconds.
[0020] The above solution utilizes millimeter-wave frequencies that provide sufficiently high distance and velocity resolution, allowing multiple targets within the elevator space to be clearly distinguished, and is less affected by ambient light and temperature. Its extremely low transmission power is harmless to humans and suitable for long-term operation. The detection range covers the area around the elevator car and landing doors, ensuring comprehensive monitoring. By combining the rate of height change with posture changes within a limited time frame to determine if a fall has occurred, it effectively distinguishes between a genuine fall and everyday actions such as slow squatting or bending over, reducing false alarms and improving alarm accuracy.
[0021] The aforementioned multi-mode integrated elevator barrier-free intelligent human-computer interaction system also includes a voice interaction module. When the millimeter-wave sensing module detects the fall action and the identity recognition module fails to detect valid identity information within a predetermined time, the voice interaction module initiates a voice inquiry and confirms whether rescue needs to be initiated based on the voice feedback.
[0022] The above solution allows the system to proactively question the person in distress via voice when their identity cannot be verified, thus promptly confirming their level of consciousness and need for assistance. The system then uses this voice feedback to determine whether to call the emergency services. This approach avoids delays in treatment due to uncertainty about the situation and reduces false alarms caused by minor slips or uninjured individuals, making the emergency response process more humane and precise.
[0023] This invention also provides a multi-mode fusion elevator accessibility intelligent human-computer interaction method, applied to the above-mentioned system, comprising the following steps: S1. Transmit millimeter wave signals and receive echoes. By analyzing the echo signals, identify the target's behavioral state and environmental characteristics. The behavioral state includes at least a falling action, and the environmental characteristics include at least the presence of salt spray, pet targets, and electric vehicle targets. S2. Obtain passenger identity information; S3. Based on the identity information and associated historical elevator usage records, automatically push frequently used floors in non-escalator control mode, and automatically generate elevator call commands and register authorized floors in elevator control mode; and... When applied to home elevator scenarios, when a call command is generated or a passenger is detected to have arrived at the target floor, a linkage control command is sent to the home appliances associated with the passenger through the smart home gateway.
[0024] The above-mentioned solution organically integrates millimeter-wave detection, multimodal identity recognition, and historical data utilization, enabling elevators to possess intelligent perception and decision-making capabilities throughout the entire elevator ride. From warnings of dangerous behaviors and handling of environmental anomalies, to hands-free floor selection, automatic access control, and cross-space home appliance control, a complete barrier-free proactive service system is formed, significantly improving elevator safety, ease of operation, and home integration experience.
[0025] The present invention has the following beneficial effects: By integrating millimeter-wave environmental sensing, two-factor authentication, and adaptive historical habits, this invention enables elevators to proactively detect and respond to various potential hazards. Fall detection and instant alarms significantly shorten emergency response time, protecting the lives of passengers; the combination of salt spray monitoring and automatic ventilation and drying helps slow down corrosion of metal components and aging of electrical wiring, reducing maintenance costs and extending elevator lifespan; the identification of pets and electric vehicles and the prevention of door closing actions shift accident prevention from passive to proactive, reducing safety hazards caused by pets being left alone in the elevator or electric vehicles hitting hall doors.
[0026] In terms of enhancing the accessibility experience, adaptive floor recommendations based on historical records allow passengers with mobility impairments or visual and auditory impairments to call elevators without pressing any buttons, truly achieving contactless and operation-free intelligent elevator calling. In elevator control management scenarios, permission verification and floor registration are integrated, eliminating the need for passengers to repeatedly switch between cards, fingerprints, and buttons, significantly improving passage efficiency. Moreover, permission allocation is flexible, taking into account both daily habits and temporary needs.
[0027] In the application scenario of home elevators, the act of riding the elevator itself serves as a trigger to achieve home appliance linkage, making the elevator a smart hub connecting the inside and outside of the home. Automatic welcome upon returning home and automatic energy saving upon leaving home not only improve living comfort but also reduce energy waste caused by human forgetfulness. All linkage decisions rely on identity recognition and elevator operating status, eliminating the need for additional sensors deployed in the home. The solution is simple, reliable, and easy to deploy.
[0028] In summary, the multi-mode fusion solution provided by this invention connects multiple previously independent functional chains such as environmental perception, identity authentication, elevator usage habits, access control, and home control, transforming the elevator from a single vertical transportation device into a comprehensive interactive platform integrating safety protection, accessibility services, and smart home access. Attached Figure Description
[0029] Figure 1 A schematic diagram of the connection relationship of a multi-mode fusion elevator barrier-free intelligent human-computer interaction system provided in an embodiment of the present invention; Figure 2A flowchart of a multi-mode fusion elevator accessibility intelligent human-computer interaction method provided in an embodiment of the present invention. Detailed Implementation
[0030] The technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.
[0031] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual images. They should not be construed as limiting the scope of this application. To better illustrate the embodiments of the present invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual dimensions of the product. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.
[0032] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "inner," and "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present application. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.
[0033] In the description of this invention, unless otherwise explicitly specified and limited, the term "connection" or similar designation indicating a connection between components should be interpreted broadly. For example, it can refer to a fixed connection, a detachable connection, or an integral part; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can refer to the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.
[0034] Example Reference Figure 1 The multi-mode fusion elevator barrier-free intelligent human-computer interaction system in this embodiment includes a millimeter wave sensing module, an identity recognition module, a control module, an interactive interface, a voice interaction module, and a wireless communication module. It also has a dedicated interface for connecting the elevator controller and the smart home gateway.
[0035] The millimeter-wave sensing module uses FMCW millimeter-wave radar, operating in the frequency range of 57GHz to 64GHz, with a standard nominal operating frequency of 60GHz. Its transmit power is less than 10dBm, and its effective detection range covers 0.1 meters to 10 meters. The module has a built-in signal processing unit capable of range-Doppler processing of the radar echo, outputting point clouds and time-spectrum maps. It relies on a pre-trained convolutional recurrent neural network to recognize human fall actions. The judgment logic is: if the human target's height decreases at a rate exceeding 1.2 meters per second within a time window of 0.5 to 3 seconds, and ultimately maintains a horizontal posture, it is determined as a fall event.
[0036] Salt spray status identification relies on the attenuation of environmental echo scattering intensity. First, a baseline scattering intensity value is collected in a clean environment. During equipment operation, the real-time echo intensity is compared. If the scattering intensity attenuation caused by salt spray particles reaches 10dB to 20dB and lasts for a certain period of time, it is determined that salt spray exists on site, and the control module immediately starts the elevator ventilation device or drying device.
[0037] The identification of pets and electric vehicles relies on the target's radar cross section, size, and micro-motion characteristics to classify them. After the corresponding target is detected, a warning icon pops up on the interactive interface, and a voice reminder is given through the speaker. At the same time, the control module sends a command to the elevator controller to prevent the door from closing.
[0038] The identity recognition module integrates a 13.56MHz contactless IC card reader and a binocular face recognition camera. The face recognition unit can be replaced by a fingerprint acquisition unit or an iris acquisition unit. The system adopts a two-factor authentication mechanism. The module will only output a valid identity identifier after both types of recognition verification have passed.
[0039] The control module is built on an embedded processor, and system history records are uniformly stored in local non-volatile memory. The system has two operating modes: no elevator control and elevator control. In no elevator control mode, the control module retrieves the floor with the highest elevator usage frequency for the user over the past 30 days as the frequently used floor, displays it on the in-car touchscreen, and announces it via voice. In elevator control mode, the control module matches the built-in floor permission table. If the user has only one authorized floor, the corresponding floor button is directly lit up. If there are multiple authorized floors, they are sorted according to historical usage frequency, and the floor with the highest usage frequency is set as the default call command. The remaining authorized floors are simultaneously displayed on the screen for passengers to manually select via touch.
[0040] This system is compatible with home elevator scenarios. The smart home gateway connects to indoor air conditioner, lighting, and curtain controllers via Wi-Fi, and has a built-in linkage strategy unit. The linkage triggering rules are divided into two scenarios: "coming home" and "leaving home." When the control module generates a call command or identifies a passenger arriving at the target floor, the system determines it's a "coming home" scenario and issues a "coming home" linkage command. When the system determines the passenger has left the residence and the elevator has reached a designated exit floor, it issues a "leaving home" linkage command. The "coming home" scenario is determined by the passenger entering the home elevator through the identity recognition module and the elevator starting up; the "leaving home" scenario is determined by the passenger leaving the residence and the elevator reaching a preset exit floor. Typical linkage command examples: In the "coming home" scenario, the air conditioner can be set to 25 degrees Celsius, living room lights can be turned on, and curtains can be closed; in the "leaving home" scenario, the power to all preset controlled appliances can be cut off.
[0041] Reference Figure 2 The complete system workflow is as follows: After a passenger enters the elevator lobby, the millimeter-wave sensing module continuously scans the entire area; when a passenger approaches the recognition area, the identity recognition module completes two-factor authentication; the control module automatically pushes the target floor or directly initiates an elevator call in conjunction with the elevator control operation mode; if the device is a home elevator model, it synchronously links all the home appliances to execute corresponding scene commands; the millimeter-wave sensing module synchronously and in parallel monitors four types of targets: falls, salt spray, pets, and electric vehicles. Once the corresponding event is detected and triggered, the system immediately executes the corresponding safety measures or audible and visual alarm actions.
[0042] All parameter values in this solution have been verified through multiple sets of comparative experiments: the 57GHz to 64GHz millimeter wave band can meet the distance resolution requirements for indoor scenarios; the shortest response time window of 0.5 seconds can quickly capture sudden falls, while the longest observation window of 3 seconds can effectively filter out everyday irrelevant actions such as squatting and bending over; the 10dB to 20dB salt spray attenuation threshold can clearly distinguish between coastal saline and humid air and ordinary water mist, avoiding accidental triggering of dehumidification and ventilation equipment. All functional modules operate in parallel and collaboratively, realizing a barrier-free intelligent human-computer interaction function for elevators that integrates multiple sensing and verification modes.
[0043] In addition, the control module has preset fall determination equations and salt spray concentration determination equations, which are used to quantify and distinguish human target fall events and environmental salt spray conditions detected by the millimeter wave sensing module, respectively. The equation for determining a fall is: ; in, For the target center of mass of the human body within the time window The change in height within, in units of ; To enable sliding detection time windows, satisfying ; To preset the vertical velocity reference threshold, ; The change in angle between the principal axis of the human body target and the vertical direction, expressed in units of... ; To preset the angular velocity reference threshold, ; The instantaneous velocity of the human target in the vertical direction, in units of . , Its absolute value; For the weighting coefficients, satisfying ,and , , ; This represents the steepness coefficient of the S-shaped function. ; For the step decision function, when the comprehensive score within the parentheses... Time output Otherwise output ,in To preset the discrimination threshold, ; This indicates that the incident was determined to be a fall. This indicates that no fall occurred; The equation for determining salt spray concentration is: ; in, The power of the echo signal received at the current moment. The reference echo signal power calibrated in a clean environment; The distance at which millimeter-wave signals propagate within the elevator car, measured in millimeter units. ; The attenuation coefficient is related to the concentration of salt spray particles, and the unit is . ; The operating frequency of millimeter-wave signals, in units of . ; This is a frequency correction factor related to the salt spray particle size distribution, in units of... ; It is a dimensionless indicator for assessing salt spray concentration; when When the salt spray condition is deemed to exceed the standard, To preset the salt spray detection threshold, and The value of corresponds to the attenuation of the echo signal scattering intensity. to between.
[0044] Parameter description table: Example 1: Fall Assessment The system detects a human target at a certain moment and takes a time window. The following feature data were collected: Center of mass height from Down to ,but ; The angle between the principal axis and the vertical direction is from ( ) becomes ( ),but ; Vertical instantaneous velocity Take the absolute value .
[0045] Substitute into the fall determination equation and calculate step by step: because Step function output If the event is determined to be a fall, the control module will immediately generate an alarm signal.
[0046] In contrast, a slow squatting motion would be considered: , , ,but: It does not trigger alarms, effectively avoiding false alarms.
[0047] Example 2: Salt spray concentration determination The system is running inside a coastal building, and its current parameters are: Reference power Measured power (attenuation approximately) ); Transmission distance attenuation coefficient ; Operating frequency Frequency correction factor .
[0048] Substitute into the salt spray concentration equation: If the building's salt spray detection threshold (correspond (scattering attenuation), because If the salt spray level is found to be excessive, the control module immediately activates the elevator ventilation system to dry the car, continuing this process until three consecutive tests are conducted. Then stop.
[0049] The beneficial effects of the fall detection equation: Dimensional normalization enables legitimate fusion of multiple features: by introducing a reference threshold. and This unifies physical quantities with different dimensions into dimensionless scalars, fundamentally solving the dimensional contradiction problem in the weighted fusion of multi-source heterogeneous features, and making the equation strictly self-consistent mathematically.
[0050] S-shaped function achieves smooth decision boundary: S-shaped nonlinear mapping is used for the vertical velocity amplitude, when... exist When the surrounding area changes, the output shows a gradual trend rather than a hard jump, avoiding the sensitive fluctuations of the threshold feature at the critical point and significantly improving the robustness of the system under boundary conditions.
[0051] Multi-dimensional cross-validation significantly reduces the false alarm rate: three features mutually validate each other—a drop in height without a change in posture (such as a rapid squat) will not trigger a fall detection; a change in posture without a significant drop in height (such as leaning sideways) will not trigger a fall detection; and a speed that meets the criteria but both height and posture are normal (such as jumping and falling) will also not meet the triggering conditions. The synergy of these three features significantly improves the accuracy of fall detection compared to using only speed / height criteria.
[0052] Beneficial effects of the salt spray concentration determination equation: Quantitative continuous detection: elevates salt spray status from a qualitative assessment of presence or absence to a quantifiable, continuous indicator. This provides precise data for the start-up and shutdown control of ventilation and drying devices, avoiding frequent fluctuations of fixed thresholds under different humidity environments.
[0053] Distance adaptive compensation: through The term applies normalization compensation to the propagation distance, making the near end of the elevator car ( ) and remote ( The salt spray concentration assessment has a consistent judgment scale, eliminating systematic bias introduced by distance differences.
[0054] Frequency adaptive calibration: via The item compensates for different operating frequencies, so that the system can... to A consistent salt spray assessment standard across the entire frequency band eliminates the need for recalibration due to frequency adjustments, enhancing system compatibility and engineering applicability.
[0055] The workflow of the fall detection equation: Step 1: The millimeter-wave sensing module operates at a fixed frame rate (default). The signal processing unit collects point cloud data of a human target inside the elevator car and extracts the three-dimensional coordinates of the human centroid from the point cloud in real time. Given the direction vector of the human body's principal axis, calculate the angle between the principal axis and the vertical direction. ; Step 2: The control module operates within a sliding time window. (default Within ) calculate the change in centroid height. Change in the angle between the principal axes and the instantaneous vertical velocity at the current moment. (Obtained from the first-order difference of the vertical coordinates of the centroid of the point cloud); Step 3: Substitute the three characteristic quantities into the fall judgment equation and calculate the comprehensive score. ; Step 4: With preset threshold In comparison, if Step function output The control module immediately generates a fall alarm signal and pushes it to the property monitoring terminal via wireless communication; if Output The system continues to monitor, the time window slides forward, and steps two through four are repeated.
[0056] Workflow of the salt spray concentration determination equation: Step 1: During the cleanroom calibration and commissioning of the elevator, the system acquires the reference echo signal power in a confirmed clean environment (no salt spray). It is stored in the non-volatile memory of the control module; Step 2: During system operation, the millimeter-wave sensing module operates at a fixed interval (default every...). Collect the current echo signal power once per second. At the same time, the current millimeter-wave signal propagation distance is obtained. (Obtained by radar ranging) and operating frequency ; Step 3: The control module will Substitute into the salt spray concentration determination equation to calculate the current salt spray concentration assessment index. ; Step 4: With preset threshold In comparison, if The control module outputs a start command to the elevator ventilation / drying device until three consecutive tests are completed. Then stop; Step 5: If If the concentration of salt spray in the environment is determined to be within the normal range, the system will continue to monitor and repeat steps two through four.
[0057] The two judgment equations mentioned above operate independently and in parallel, working in conjunction with the identity recognition module, elevator controller, and smart home gateway to form a complete elevator accessibility safety protection and environmental adaptive control system.
[0058] In summary, the working principle of the multi-mode fusion elevator accessibility intelligent human-computer interaction system provided in this embodiment is as follows: The millimeter-wave sensing module continuously transmits millimeter-wave signals of a specific frequency band into the elevator car and lobby area, and receives echoes reflected from the target and the environment. The signal processing unit converts the echoes into time-spectrum maps or point cloud data, and uses a pre-trained deep learning model to classify and identify this data. For falls, the system detects the rate of descent and posture changes of the human target within a limited time window, effectively distinguishing between real falls and everyday actions such as slow squatting. For salt spray conditions, the system records the baseline scattering intensity in a clean environment and compares the attenuation of the echo scattering intensity in real time during operation. When the attenuation exceeds a preset threshold and remains so for a certain period of time, it is considered a positive salt spray accumulation. Pets and electric vehicles are distinguished based on radar cross-section, size, and micro-motion characteristics.
[0059] The identity recognition module integrates a contactless IC card reader with biometric recognition units such as fingerprint, facial, or iris scans. It employs a two-factor authentication method, only outputting a passenger's identity identifier after both verifications are successful. This mechanism ensures that the identity cannot be misused.
[0060] The control module includes a historical record database and a linkage strategy unit. Upon acquiring a passenger's identity, the system retrieves records from the database associated with that identity, such as travel time, starting floor, and destination floor. By statistically analyzing travel frequency or recent usage frequency, it dynamically determines the passenger's frequently used floors. When elevator control is not enabled, the control module directly pushes recommended information including frequently used floors to the display screen and voice announcement device. When elevator control is enabled, the control module compares the identity information with a pre-stored floor permission table. If the permission corresponds to a unique floor, it directly registers that floor; if it corresponds to multiple authorized floors, it automatically selects the most frequently used authorized floor to generate an elevator call command, while retaining an interactive method for passengers to manually select other authorized floors.
[0061] When the system is applied to home elevators, the linkage strategy unit communicates with devices such as air conditioners, lights, and curtains within the residence via a smart home gateway. When the control module generates a call command or detects a passenger arriving at the target floor, it further determines whether the elevator ride is a homecoming or departure scenario. A homecoming scenario is determined by the passenger entering the elevator through identity verification and the elevator starting operation; a departure scenario is determined by the passenger leaving the residence and the elevator reaching the designated exit floor. If a homecoming scenario is determined, the system sends an activation command to the corresponding home appliances to create a comfortable environment in advance; if a departure scenario is determined, a deactivation command is sent to achieve automatic energy saving.
[0062] Furthermore, the millimeter-wave sensing module performs multiple safety checks in parallel throughout the passenger's elevator ride. Upon detecting a fall, the control module immediately generates an alarm signal, which is sent via wireless communication to the property management terminal or a pre-set emergency contact. When the salt spray level exceeds a set threshold, the elevator ventilation or drying system is automatically activated. If a pet or electric vehicle is detected entering, a voice warning is triggered, and a command is sent to the elevator controller to prevent the doors from closing, thus preventing accidents.
[0063] How to use: As passengers approach the elevator, the millimeter-wave sensing module continuously scans the environment. Passengers simply need to carry their registered contactless IC card as usual and face the biometric recognition unit to complete face, fingerprint, or iris scanning. The system then performs dual authentication in the background without any manual operation.
[0064] After identity verification, if the elevator does not have elevator control functionality, the in-car display screen will automatically highlight the passenger's most frequently visited floor, and the floor information will be announced via voice; the passenger only needs to confirm. If elevator control functionality is enabled, the system will automatically compare permissions and directly illuminate the authorized floor button. When multiple authorized floors exist, the most frequently used floor will be selected by default. Passengers can also temporarily change the target floor via touchscreen or voice command. In a home elevator scenario, from the moment a passenger enters the elevator and it starts running, home appliances such as air conditioners and lights will automatically adjust according to preset scene parameters; when the passenger leaves home and reaches the exit floor, the relevant appliances will automatically turn off.
[0065] Throughout the entire elevator ride, passengers do not need to perform any additional safety procedures; the system continuously and silently protects them. If someone falls inside the elevator car, an alarm will be triggered immediately; if there is a buildup of humid salt spray in the coastal area, the ventilation and drying system will automatically activate; if a pet or electric vehicle accidentally enters the elevator, the system will issue a warning and prevent the elevator doors from closing, protecting the safety of all passengers.
[0066] The above are merely preferred embodiments of the present invention and are not intended to limit the implementation methods and protection scope of the present invention. Those skilled in the art should recognize that any equivalent substitutions and obvious changes made based on the description and illustrations of the present invention should be included within the protection scope of the present invention.
Claims
1. A multi-modal, fusion-based, elevator accessibility, intelligent human-machine interaction system, characterized by, include: A millimeter-wave sensing module is used to transmit millimeter-wave signals to the elevator car and / or elevator hall area and receive echoes. By analyzing the echo signals, the module identifies the behavioral state and environmental characteristics of the target. The behavioral state includes at least a falling action, and the environmental characteristics include at least the presence of salt spray, pet targets, and electric vehicle targets. The identity recognition module integrates an IC card reader and a biometric recognition unit to obtain passenger identity information; The control module is communicatively connected to the millimeter-wave sensing module, the identity recognition module, the elevator controller, and the smart home gateway. The control module includes a historical record database and a linkage strategy unit. The control module is configured to: when the identity recognition module obtains the passenger's identity information, determine the passenger's frequently used floors based on the historical elevator records associated with the identity information in the historical record database, and generate floor recommendation information containing the frequently used floors in the no-elevator-control mode to push to the interactive interface; in the elevator-control mode, verify the floor permissions corresponding to the identity information, and automatically generate an elevator call command to register the authorized floors corresponding to the permissions. as well as, When applied to home elevator scenarios, the linkage strategy unit, through the smart home gateway, sends control commands to the home appliances associated with the passenger when the control module generates an elevator call command or detects that a passenger has arrived at the target floor, so as to automatically adjust the operating status of the home appliances.
2. The multi-modal, converged, elevator accessibility intelligent human-machine interaction system of claim 1, wherein, The millimeter-wave sensing module includes a millimeter-wave radar sensor and a signal processing unit; the signal processing unit generates a time-spectrum map or point cloud data from the echo signal, and uses a trained deep learning model to classify the time-spectrum map or point cloud data to identify the fall action, the presence of salt spray, pet targets, and electric vehicle targets.
3. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to claim 1, characterized in that, The IC card reading unit of the identity recognition module supports contactless IC cards, and the biometric recognition unit includes at least one of a fingerprint reader, a face recognition camera, and an iris reader; the identity recognition module is configured to output the identity information after the dual authentication by the IC card reading unit and the biometric recognition unit is successful.
4. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to claim 1, characterized in that, The historical record database stores the elevator travel time, starting floor, and destination floor for each passenger, and determines the frequently used floors based on elevator travel frequency or recent usage frequency statistics; the interactive interface includes a display screen and / or voice broadcast device installed in the elevator car.
5. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to claim 1, characterized in that, In elevator control mode, the control module compares the identity information with a pre-stored floor permission table. When the permission corresponds to multiple authorized floors, it automatically selects the most frequently used authorized floor according to the historical record database to generate the elevator call command, or provides the multiple authorized floors for passengers to choose from through the interactive interface.
6. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to claim 1, characterized in that, The linkage strategy unit is configured to: when the control module generates an elevator call command or detects that a passenger has arrived at the target floor, if it is determined to be a homecoming scenario, then send an activation command to the home appliances associated with the passenger through the smart home gateway; if it is determined to be a departure scenario, then send a deactivation command; wherein, the homecoming scenario is determined based on the passenger entering the home elevator through the identity recognition module and the elevator starting to run, and the departure scenario is determined based on the passenger leaving the residence and the elevator reaching the designated exit floor; the home appliances include at least one of air conditioners, lamps, and curtains.
7. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to any one of claims 1 to 6, characterized in that, The control module is also configured to perform at least one of the following actions: When the millimeter-wave sensing module detects the fall, it generates an alarm signal and sends it to a preset property management terminal or emergency contact terminal via the wireless communication module. When the presence of salt spray is detected to exceed a preset salt spray threshold, the elevator ventilation device or drying device is activated. The preset salt spray threshold is set between 10dB and 20dB based on the attenuation of the scattering intensity of the echo signal. When the pet or electric vehicle is detected, a voice warning message is output, and a command is sent to the elevator controller to prevent the elevator door from closing.
8. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to any one of claims 1 to 6, characterized in that, The millimeter-wave sensing module operates in the frequency range of 57GHz to 64GHz, with a transmission power of less than 10dBm and an effective detection range of 0.1m to 10m. When the millimeter-wave sensing module identifies the fall, it determines the fall by detecting the speed of height change and posture change of the human target within a time window of 0.5 seconds to 3 seconds.
9. The multi-mode fusion elevator barrier-free intelligent human-computer interaction system according to any one of claims 1 to 6, characterized in that, It also includes a voice interaction module. When the millimeter-wave sensing module detects the fall and the identity recognition module fails to detect valid identity information within a predetermined time, the voice interaction module initiates a voice inquiry and confirms whether rescue needs to be initiated based on the voice feedback.
10. A multi-mode fusion elevator accessibility intelligent human-computer interaction method, applied to the multi-mode fusion elevator accessibility intelligent human-computer interaction system according to any one of claims 1 to 9, characterized in that, Includes the following steps: S1. Transmit millimeter wave signals and receive echoes. By analyzing the echo signals, identify the target's behavioral state and environmental characteristics. The behavioral state includes at least a falling action, and the environmental characteristics include at least the presence of salt spray, pet targets, and electric vehicle targets. S2. Obtain passenger identity information; S3. Based on the identity information and associated historical elevator ride records, automatically push frequently used floors in the no-elevator-control mode, and automatically generate elevator call commands and register authorized floors in the elevator-control mode. as well as, When applied to home elevator scenarios, when a call command is generated or a passenger is detected to have arrived at the target floor, a linkage control command is sent to the home appliances associated with the passenger through the smart home gateway.
