Intelligent elevator operation box

Abstract

The invention discloses an intelligent elevator operation box which comprises a user identification module used for obtaining and identifying identity information of a user in a multi-mode mode; the intelligent control module is connected with the user identification module and is used for generating a dynamically adjusted personalized control strategy according to the identified identity information and the environment parameters; the self-adaptive interaction module is connected with the intelligent control module and is used for providing a scenarized human-computer interaction interface and receiving user input; and the reinforcement learning module is connected with the intelligent control module and is used for continuously optimizing the personalized control strategy based on the user feedback and the environment change. According to the intelligent elevator operation box disclosed by the invention, high intelligence and individuation of elevator service are realized by fusing a multi-mode identification technology, a reinforcement learning algorithm and a multi-system collaborative architecture.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of elevator intelligent control, and particularly relates to an intelligent elevator operating box. BACKGROUND

[0002] With the rapid development of building intelligence, as the core tool of vertical transportation, the intelligent level of elevators directly affects the operation efficiency of buildings and user experience. The traditional elevator operating box has the following technical defects: 1. Single recognition method: mainly relying on IC cards or key operations, unable to achieve accurate user recognition and personalized services; 2. Fixed interaction mode: using a unified operation interface and interaction method, unable to adapt to the needs of users of different ages and abilities; 3. Lack of learning ability: unable to optimize services according to user habits, long-term use experience has no improvement.

[0003] Therefore, the above problems are further improved. SUMMARY

[0004] The main purpose of the application is to provide an intelligent elevator operating box, which realizes high intelligence and personalization of elevator services through the fusion of multi-modal recognition technology, reinforcement learning algorithm and multi-system collaborative architecture.

[0005] To achieve the above purpose, the application provides an intelligent elevator operating box, comprising: A user identification module for multi-modal acquisition and identification of user identity information; An intelligent control module connected with the user identification module, for generating a dynamically adjusted personalized control strategy according to the identified identity information and environmental parameters; An adaptive interaction module connected with the intelligent control module, for providing a scenario-based human-computer interaction interface and receiving user input; A reinforcement learning module connected with the intelligent control module, for continuously optimizing the personalized control strategy based on user feedback and environmental changes.

[0006] As a further preferred technical solution of the above technical solution, the user identification module comprises: A biological feature acquisition unit for acquiring at least one biological feature of a user's facial feature, fingerprint feature, iris feature or voiceprint feature; A behavior feature analysis unit for identifying the gait feature and operation habit feature of the user; A physiological state detection unit for detecting the physiological parameters of the user including heart rate and breathing rate through a non-contact method; The recognition fusion unit obtains the biological features transmitted by the behavior feature analysis unit and the physiological parameters transmitted by the physiological state detection unit, and performs weighted fusion on the multi-modal recognition results (i.e., the biological features and the physiological parameters), so as to accurately recognize the identity information of the user.

[0007] As a further preferred technical solution of the above technical solution, the personalized control strategy comprises: A destination intelligent prediction sub-strategy generates a plurality of priority-ordered predicted floors based on user data including historical behavior data, current time, weather conditions, and building function distribution; An interaction mode adaptation sub-strategy automatically switches the interaction mode according to user data including age and operation habits, including dynamic adjustment of touch sensitivity, voice response speed, and interface layout; A service priority adjustment sub-strategy dynamically adjusts the elevator response priority in combination with user physiological state and reservation information; An information push sub-strategy pushes personalized information content and timing control according to user interest tags and travel characteristics.

[0008] As a further preferred technical solution of the above technical solution, the adaptive interaction module comprises: A multi-modal input unit supports seamless switching of multiple input modes including touch, voice, gesture, and eye tracking; A contextual display unit automatically adjusts display parameters including brightness, contrast, font size, and interface layout according to light intensity, user distance, and use scenario; A feedback output unit provides multi-channel feedback of vision, hearing, and touch; An emergency interaction unit is equipped with a one-key distress function and can automatically identify user emergency state and trigger corresponding services.

[0009] As a further preferred technical solution of the above technical solution, the reinforcement learning module comprises: An online learning unit is used for real-time collection of user operation data and incremental learning; An offline optimization unit is used for large-scale data training and model optimization during night low-peak periods; A user feedback collection unit acquires implicit and explicit feedback of users through active inquiry and operation behavior analysis; A transfer learning unit is used for applying behavior patterns of similar user groups to new users; A strategy evaluation unit evaluates and optimizes the control strategy through multi-dimensional indicators including accuracy, response speed, and user satisfaction.

[0010] As a further preferred technical solution of the above technical solution, the control method of the intelligent elevator operating box comprises: Step S1: acquiring and fusing the identity information of the user by the user identification module multi-modal; Step S2: the intelligent control module combines the environmental parameters and the identity information to generate a personalized control strategy containing destination prediction, interactive mode adaptation, and service priority; Step S3: the adaptive interaction module provides scenario-based human-computer interaction services based on the personalized control strategy; Step S4: the reinforcement learning module collects user feedback data in real time and continuously optimizes the personalized control strategy through the reinforcement learning algorithm.

[0011] As a further preferred technical solution of the above technical solution, step S2 is specifically implemented as: Step S2.1: constructing a three-dimensional user portrait (including basic attribute dimension, behavior pattern dimension, and preference feature dimension); Step S2.2: establishing an environment-user service association model to analyze the mapping relationship between environmental parameters and user needs; Step S2.3: predicting user short-term and long-term needs based on time series analysis and deep learning algorithms.

[0012] The beneficial effects of the present application are: 1. Accurate identification: multi-modal fusion identification technology improves the accuracy rate, far exceeding the performance of single identification method; 2. Personalized service: three-dimensional user portrait and dynamic strategy generation mechanism realize truly personalized service; 3. Continuous evolution: the reinforcement learning mechanism continuously optimizes the system performance over time; 4. Interactive friendliness: multi-modal interaction and adaptive interface ensure the convenience of various user groups. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 is a schematic diagram of the present application. DETAILED DESCRIPTION

[0014] The following description is used to disclose the present application so that those skilled in the art can implement the present application. The preferred embodiments in the following description are only as examples, and other obvious modifications can be thought of by those skilled in the art. The basic principles of the present application defined in the following description can be applied to other embodiments, modifications, improvements, equivalents, and other technical solutions without departing from the spirit and scope of the present application.

[0015] In the preferred embodiments of the present application, those skilled in the art should note that the elevators, users, etc. involved in the present application can be regarded as prior art.

[0016] Preferred embodiments.

[0017] like Figure 1 As shown, the present invention discloses an intelligent elevator control panel, comprising: The user identification module is used to acquire and identify user identity information in a multimodal manner; The intelligent control module, connected to the user identification module, is used to generate dynamically adjusted personalized control strategies based on the identified identity information and environmental parameters, specifically as follows: User profile construction: A three-dimensional profile is built based on basic user information (age, gender, occupation), behavioral data (usage time, frequency, destination), and preference settings; Destination prediction: Using an LSTM time series model combined with a random forest algorithm, the destination is predicted based on the user's historical data and the current context (time, weather, building activity), generating a priority list containing 3 floors; Strategy Generation: Based on user profiles and system status (elevator load, direction of travel, number of waiting passengers), a multi-objective optimization algorithm is used to generate control strategies that balance individual experience with overall efficiency. Dynamic adjustment: The user status and system parameters are evaluated once every preset period. When a significant change is detected (such as a sudden increase in the user's heart rate), the service strategy is adjusted immediately. An adaptive interaction module, connected to the intelligent control module, is used to provide a scenario-based human-computer interaction interface and receive user input; The reinforcement learning module, connected to the intelligent control module, is used to continuously optimize personalized control strategies based on user feedback and environmental changes.

[0018] Specifically, the user identification module includes: A biometric acquisition unit is used to acquire at least one of the following biometric features of a user: facial features (via an infrared camera), fingerprint features (via an optical fingerprint sensor), iris features, or voiceprint features (via a microphone array); The behavioral feature analysis unit is used to identify the user's gait characteristics and operating habits. The physiological state detection unit is used to detect the user's physiological parameters, including heart rate and respiratory rate, in a non-contact manner (for example, using a non-contact heart rate sensor to detect changes in the user's heart rate to determine whether the user is in a state of tension or emergency. If the user suddenly feels unwell, the system will automatically identify the emergency and immediately display on the screen: "We have detected that you may be unwell. Do you need help?" and other similar messages. It can also activate the one-button emergency call function through buttons, voice, etc.). The identification fusion unit obtains the biometrics transmitted by the behavior feature analysis unit and the physiological parameters transmitted by the physiological state detection unit, and performs weighted fusion on the multimodal identification results (i.e., biometrics and physiological parameters) to accurately identify the user's identity information.

[0019] More specifically, The personalized control strategy includes: A destination intelligent prediction sub-strategy generates a plurality of priority-ordered predicted floors based on user data including historical behavior data, current time, weather conditions, and building function distribution; An interaction mode adaptation sub-strategy automatically switches interaction modes, including touch sensitivity, voice response speed, and dynamic adjustment of interface layout, according to user data including age and operation habits; A service priority adjustment sub-strategy dynamically adjusts elevator response priority in combination with user physiological state and reservation information; An information push sub-strategy pushes personalized information content and timing control according to user interest tags and travel characteristics.

[0020] Further, the adaptive interaction module includes: A multi-modal input unit supports seamless switching of multiple input modes including touch, voice, gesture, and eye tracking; A contextual display unit automatically adjusts display parameters, including brightness, contrast, font size, and interface layout, according to light intensity, user distance, and use scenarios; A feedback output unit provides multi-channel feedback of vision, hearing, and touch; An emergency interaction unit is equipped with a one-key distress function and can automatically identify user emergency states and trigger corresponding services.

[0021] Further, the reinforcement learning module includes: An online learning unit is used for real-time collection of user operation data and incremental learning; An offline optimization unit is used for large-scale data training and model optimization during night low-peak periods; A user feedback collection unit obtains implicit and explicit feedback of users through active inquiry and operation behavior analysis; A transfer learning unit is used for applying behavior patterns of similar user groups to new users; A strategy evaluation unit evaluates and optimizes the control strategy through multi-dimensional indicators including accuracy, response speed, and user satisfaction.

[0022] Preferably, the control method of the intelligent elevator operating box includes: Step S1: The user identification module acquires and fuses the identity information of the user through multi-modal identification; Step S2: The intelligent control module generates a personalized control strategy containing destination prediction, interaction mode adaptation, and service priority in combination with environmental parameters and identity information; Step S3: The adaptive interaction module provides a scenario-based human-computer interaction service based on the personalized control strategy. Step S4: The reinforcement learning module collects user feedback data in real time, and continuously optimizes the personalized control strategy through a reinforcement learning algorithm.

[0023] Preferably, step S2 is implemented as: Step S2.1: Construct a three-dimensional user portrait (including basic attribute dimensions, behavior pattern dimensions, and preference feature dimensions); Step S2.2: Establish an environment-user service association model to analyze the mapping relationship between environmental parameters and user needs; Step S2.3: Predict user short-term and long-term needs based on time series analysis and deep learning algorithms.

[0024] In actual application scenarios, as an example, the workflow is as follows: In the morning, the user enters the elevator car, and the user recognition module completes face recognition and voiceprint verification in a short time to confirm the user's identity; The intelligent control module retrieves the user portrait, combines the current time and historical data, and predicts that the user's destination is the 16th floor (office); The adaptive interaction module highlights the 16th floor button on the screen and issues a voice prompt: Good morning, do you need to go to the 16th floor? The user nods to confirm (or gesture recognition, etc.), and the system registers the 16th floor and displays the estimated arrival time; The reinforcement learning module records the user's confirmation behavior and increases the weight of the 16th floor prediction for this period.

[0025] It is worth mentioning that the elevator, user, and other technical features involved in the present patent application should be considered as prior art. The specific structure, working principle, and possible control method, spatial arrangement method of these technical features can be selected conventionally in the art, and should not be considered as the invention point of the present patent. The present patent will not be further detailed.

[0026] For those skilled in the art, the technical solutions described in the foregoing embodiments can still be modified or some technical features can be replaced by equivalents. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent elevator control box, characterized in that, Comprise: A user identification module for multi-modal acquisition and identification of user's identity information; An intelligent control module connected with the user identification module, for generating a dynamically adjusted personalized control strategy according to the identified identity information and environmental parameters; An adaptive interaction module connected with the intelligent control module, for providing a scenario-based human-computer interaction interface and receiving user input; A reinforcement learning module connected with the intelligent control module, for continuously optimizing the personalized control strategy based on user feedback and environmental changes.

2. The intelligent elevator control box according to claim 1, wherein, The user identification module comprises: A biological feature acquisition unit for acquiring at least one biological feature of the user's facial features, fingerprint features, iris features, or voiceprint features; A behavior feature analysis unit for identifying the user's gait features and operation habit features; A physiological state detection unit for detecting the user's physiological parameters including heart rate and breathing rate in a non-contact manner; An identification fusion unit that obtains the biological features transmitted by the behavior feature analysis unit and the physiological parameters transmitted by the physiological state detection unit, and performs weighted fusion on the multi-modal identification results to accurately identify the user's identity information.

3. The intelligent elevator control box of claim 2, wherein, The personalized control strategy comprises: A destination intelligent prediction sub-strategy that generates a plurality of priority-ordered predicted floors based on user data including historical behavior data, current time, weather conditions, and building function distribution; An interaction mode adaptation sub-strategy that automatically switches the interaction mode based on user data including age and operation habits, including dynamic adjustment of touch sensitivity, voice response speed, and interface layout; A service priority adjustment sub-strategy that dynamically adjusts the elevator response priority in combination with user physiological state and reservation information; An information push sub-strategy that pushes personalized information content and timing control based on user interest tags and trip characteristics.

4. The intelligent elevator control box of claim 3, wherein, The adaptive interaction module comprises: A multi-modal input unit that supports seamless switching of multiple input methods including touch, voice, gesture, and eye tracking; A contextual display unit that automatically adjusts display parameters including brightness, contrast, font size, and interface layout based on light intensity, user distance, and usage scenario; A feedback output unit that provides visual, auditory, and tactile multi-channel feedback; An emergency interaction unit equipped with a one-key distress function and capable of automatically identifying user emergency state and triggering corresponding services.

5. A smart elevator landing panel according to claim 4, wherein, The reinforcement learning module comprises: An online learning unit for real-time collection of user operation data and incremental learning; An offline optimization unit for large-scale data training and model optimization during night low-peak periods; A user feedback collection unit that obtains implicit and explicit feedback from users through active inquiry and operation behavior analysis; A transfer learning unit for applying behavior patterns of similar user groups to new users; A strategy evaluation unit that evaluates and optimizes the control strategy through multi-dimensional indicators including accuracy, response speed, and user satisfaction.

6. A smart elevator landing panel according to any of claims 1-5, characterized in that, The control method of the intelligent elevator operating box comprises: Step S1: Multi-modal acquisition and fusion of user's identity information by the user identification module; Step S2: The intelligent control module combines environmental parameters and identity information to generate a personalized control strategy containing destination prediction, interaction mode adaptation, and service priority; Step S3: The adaptive interaction module provides scenario-based human-computer interaction services based on the personalized control strategy; Step S4: The reinforcement learning module collects user feedback data in real time and continuously optimizes the personalized control strategy through reinforcement learning algorithms.

7. A smart elevator landing panel according to claim 6, wherein, Step S2 is implemented as follows: Step S2.1: Construct a three-dimensional user portrait; Step S2.2: Establish an environmental user service association model to analyze the mapping relationship between environmental parameters and user needs; Step S2.3: Predict short-term and long-term user needs based on time series analysis and deep learning algorithms.

Images