Height-adjustable medical robot carrying identification electronic assembly

Abstract

The invention relates to the technical field of medical robots, and discloses a height-adjustable medical robot carrying an identification electronic component, which comprises the following modules: a mobile navigation module; a self-adaptive man-machine interaction module; a physical support unit; integrating a sensing unit; an interaction feedback unit; a central processing module; a multi-machine cooperative communication module; according to the invention, the posture of a user is sensed in real time through an integrated vision and depth sensor, and the physical support unit and the interactive interface are automatically driven to carry out height and angle self-adaptive adjustment based on a classification decision rule; the robot can accurately adapt to the optimal interaction view and operation distance of users in different states such as standing of adults and children, riding of wheelchairs and even lying of sickbeds, and the interaction friendliness and service accessibility of patients with mobility difficulties and special body positions are greatly improved; secondly, voice, touch control and visual gesture signals are collected in parallel, fusion decision making is carried out, and dialect recognition and sign language animation output are combined.

Description

Technical Field

[0001] This invention relates to the field of medical robot technology, specifically to a highly adjustable medical robot equipped with identification electronic components. Background Technology

[0002] With the deepening of smart healthcare construction, medical service robots with functions such as guidance, consultation, and transportation have been gradually applied in hospital settings, aiming to optimize the medical process and alleviate repetitive labor for medical staff. Existing medical robots are mostly equipped with touchscreens and basic voice modules for human-computer interaction and utilize SLAM technology for autonomous navigation. However, these robots still have significant limitations in practical applications: First, the height and angle of their interactive interfaces (such as screens and cameras) are usually fixed, making it difficult to simultaneously adapt to the natural interactive field of vision and comfortable operating distance of users in different postures, such as standing adults, children, wheelchair users, or those lying in bed, which is not user-friendly for patients with mobility impairments or special positions. Second, the interaction modality is relatively simple, lacking flexible adaptability to situations where users may experience operational inconveniences (such as inability to touchscreen) or language barriers (such as using dialects or being deaf or mute) in complex medical scenarios. Third, they are mostly independent working units, lacking real-time information sharing and task collaboration between robots. Their intelligence level is insufficient when dealing with scenarios requiring multi-person collaboration, such as emergency transport and peak-hour crowd control, making it difficult to achieve systematic efficiency improvements and emergency response.

[0003] To address this, we propose a highly adjustable medical robot equipped with recognition electronic components. Summary of the Invention

[0004] The purpose of this invention is to provide a highly adjustable medical robot equipped with identification electronic components to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a highly adjustable medical robot equipped with identification electronic components, comprising the following modules: The mobile navigation module is used to enable the robot to move autonomously and avoid obstacles. It runs a synchronous localization and mapping algorithm to generate and update the environment map in real time, and at the same time receives control commands output by the path planning algorithm to drive the robot chassis to move. The adaptive human-computer interaction module includes an electrically adjustable physical support unit, an integrated sensing unit, and an interactive feedback unit. The physical support unit includes a height-adjustable support structure for supporting the integrated sensing unit and the interactive feedback unit. The integrated sensing unit includes at least a visual sensor for acquiring color images, a depth sensor for acquiring depth information, and a microphone array for acquiring speech. The interactive feedback unit includes at least a touchscreen for information display and a speaker for voice output; The central processing module is connected to the mobile navigation module, the adaptive human-computer interaction module, the multi-machine collaborative communication module, and the data interface module, respectively, and is used to execute control logic and information processing. The multi-robot collaborative communication module is used to wirelessly exchange data with other robots or a central dispatch system in the hospital environment, and share status and map information. The data interface module is used for secure data communication with the hospital information system to retrieve and transmit relevant medical information.

[0006] Preferably, the central processing module is configured to execute a user state adaptive adjustment process, which includes: S201: User perception sub-process, which periodically collects RGB-D data of the scene in front through the visual sensor and depth sensor in the integrated sensing unit; S202: Pose estimation and classification sub-process, which performs the following processing on the acquired RGB-D data; S203: Interactive parameter mapping sub-process, based on the obtained pose category Query the preset mapping table M to obtain the corresponding target interaction parameter set. ; At least including: the target height of the physical support unit The target pitch angle of the display in the interactive feedback unit ; S204: Drive control sub-process, according to... Generate control commands to drive the physical support unit to adjust to the desired height. And drive the interactive feedback unit to adjust to the pitch angle. .

[0007] Preferably, the central processing module is further configured to execute a multimodal fusion interaction process, which includes: S301: Parallel input acquisition subprocess, synchronously starting the following channels; S302: Intent understanding and decision-making sub-process, which integrates and understands multi-channel input; S303: Adaptive output sub-process, selects the output mode based on user status or instruction content.

[0008] Preferably, the processing flow of the gesture recognition model includes: Detect the hand region from image sequence I(t); Key points were extracted from the hand region image, resulting in a set of spatiotemporal key point coordinate sequences. ,in Represents the coordinates of the i-th keypoint in the image. and its corresponding timestamp ; Calculate the dynamic features describing a gesture, such as the velocity sequence of keypoint motion trajectories. ,in ; Combine dynamic features with a pre-stored gesture template library The matching process is performed, and the matching degree is calculated using a dynamic time warping algorithm to determine the minimum cumulative distance. .

[0009] Preferably, the central processing module is configured to execute an emergency priority coordination process, which includes: S501: Emergency Trigger Subprocess, continuously monitors preset emergency trigger signals; S502: Status broadcast and collaborative sub-process: Once an emergency signal is triggered, the emergency protocol message is immediately broadcast through the multi-machine collaborative communication module, and an emergency alarm is sent to the hospital information system through the data interface module. S503: Navigation and Follow Sub-processes: The path planning algorithm of the mobile navigation module immediately switched to "emergency mode".

[0010] Preferably, the path planning algorithm in the mobile navigation module, in non-emergency mode, is a dynamic optimization process, including: S601: Basic path planning, static map built based on SLAM Given the user's target G, calculate the initial path using a search algorithm. ; S602: Dynamic information fusion, receiving real-time position and speed information shared by other robots through the multi-machine collaborative communication module; S603: Congestion Prediction and Replanning. Based on the received information, predict the traffic density of each segment of the path in the future time interval Δt and trigger dynamic replanning.

[0011] Preferably, the central processing module is configured to execute a load-balanced multi-machine service scheduling process, which includes: S701: Load status self-assessment, periodically counting the number of service requests received within a unit time window T. and average request processing time Calculate current load metrics : ; in, The calculated current load index is a dimensionless value; This represents the total number of service requests received within the statistical period T. This is the preset maximum service request processing capacity per unit time. This represents the average time required to process each service request within the statistical period T. The maximum average processing time acceptable for a single service request (preset). , The weighting coefficients for request quantity load and processing time load satisfy... + =1; S702: Load status broadcasting and listening, broadcasting its own load metrics The location information is periodically broadcast through the multi-machine collaborative communication module, and the broadcast information of other robots is monitored; S703: Scheduling decision and execution, if itself Continuously exceeding the high load threshold Then, the robot with the lowest available load is selected from the listening information to initiate a collaboration request.

[0012] Preferably, the diagnostic assistance linkage process implemented by the data interface module includes: S801: Information retrieval, after the user has been authenticated, retrieves their historical medical records from the hospital information system through a secure link; S802: Symptom correlation analysis, which analyzes the correlation between the user's current symptom description and historical records to calculate a risk warning score; S803: Report generation and push. Based on the analysis results, it automatically generates initial diagnosis linkage suggestion reports and pushes them to the corresponding department doctors' terminals.

[0013] Preferably, the privacy protection process of the adaptive human-computer interaction module includes: S901: Sensor status control, the vision sensor and depth sensor in the integrated sensing unit are provided with a physical shielding mechanism in front of them; when the robot is in standby or moving state, the physical shielding mechanism is controlled to be closed. S902: Controlled activation, the physical shielding mechanism is activated and the vision and depth perception functions are started only when a valid wake word is recognized by the microphone array or when a close human body is detected by the low-power infrared sensor. S903: Session end processing: After the interactive session has clearly ended, delay for a preset time and automatically control the physical shielding mechanism to return to the closed state.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This invention utilizes integrated vision and depth sensors to perceive user posture in real time and automatically drives the physical support unit and interactive interface to adaptively adjust height and angle based on classification decision rules. This allows the robot to precisely adapt to the optimal interactive field of view and operating distance for users in different states, from standing adults and children to those in wheelchairs and even those lying in hospital beds, greatly improving the interactivity and service accessibility for patients with mobility impairments and special postures. Secondly, by collecting voice, touch, and visual gesture signals in parallel and performing fusion decision-making, combined with dialect recognition and sign language animation output, a flexible and complementary multimodal interaction channel is constructed. This effectively overcomes the limitations of traditional single interaction methods when users have difficulty operating or face language barriers, significantly enhancing the robot's universality and inclusiveness in complex medical scenarios. Finally, relying on the collaborative scheduling algorithm integrated in the multi-machine collaborative communication module and the central processing module, real-time information sharing among robot groups, global path planning prioritizing emergency care, and dynamic load balancing are achieved. This upgrades the robot from an independent working unit to an intelligent collaborative network, thereby systematically improving the overall efficiency and emergency response capabilities of medical services in key scenarios such as emergency response and peak-hour triage. Detailed Implementation

[0015] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0016] This invention provides a technical solution: a highly adjustable medical robot equipped with recognition electronic components, comprising the following modules: The mobile navigation module is used to enable the robot to move autonomously and avoid obstacles. It runs a synchronous localization and mapping algorithm to generate and update the environment map in real time, and at the same time receives control commands output by the path planning algorithm to drive the robot chassis to move. The adaptive human-computer interaction module includes an electrically adjustable physical support unit, an integrated sensing unit, and an interactive feedback unit. The physical support unit includes a height-adjustable support structure for supporting the integrated sensing unit and the interactive feedback unit. The integrated sensing unit includes at least a visual sensor for acquiring color images, a depth sensor for acquiring depth information, and a microphone array for acquiring speech. The interactive feedback unit includes at least a touchscreen for information display and a speaker for voice output; The central processing module is connected to the mobile navigation module, the adaptive human-computer interaction module, the multi-machine collaborative communication module, and the data interface module, respectively, and is used to execute control logic and information processing. The multi-robot collaborative communication module is used to wirelessly exchange data with other robots or a central dispatch system in the hospital environment, and share status and map information. The data interface module is used for secure data communication with the hospital information system to retrieve and transmit relevant medical information.

[0017] Furthermore, the central processing module is configured to execute a user state adaptive adjustment process, which includes: S201: User perception sub-process, which periodically collects RGB-D data of the scene in front by integrating the vision sensor and depth sensor in the sensing unit; S202: Pose estimation and classification sub-process, which performs the following processing on the acquired RGB-D data: The pre-trained human detection model is invoked to locate human regions in the image and obtain their two-dimensional bounding boxes. ; By combining depth data, the 2D bounding box is mapped to the robot's 3D coordinate system, and the 3D coordinates of key points on the user's torso are calculated. ; Based on the key point coordinates, calculate the feature vector representing the user's posture. The feature vector includes at least: the estimated angle θ between the torso axis and the direction of gravity, and the estimated height of the user's eyes relative to the robot's base coordinate system. ; eigenvectors Input a predefined classification function, output user pose category The classification function is defined as follows: Let the set of posture categories be Ω = {"standing adult", "child", "sitting in wheelchair", "lying in bed"}; given the feature vector ,in for The vertical coordinates; the preset threshold parameter is: angle tolerance. , Height threshold Vertical coordinate threshold The classification decision rule is as follows: ; in, The output user pose category belongs to set Ω; The estimated angle between the torso axis and the direction of gravity (vertically downward), in radians; To estimate the vertical height of the user's eyes in the robot's base coordinate system; The vertical coordinates of the user's torso key points in the robot's base coordinate system; , The preset angle tolerance threshold is a small positive value used to define the boundary between "basically upright" and "significantly leaning backward". A preset height threshold is used to distinguish the approximate height of adults and children; The preset vertical coordinate threshold is used to determine whether the user is in a sitting or lying position (i.e., the torso height is significantly lower than when standing). | is the logical AND operator; |·| is the absolute value operator. This means "much smaller than"; S203: Interactive parameter mapping sub-process, based on the obtained pose category Query the preset mapping table M to obtain the corresponding target interaction parameter set. ; At least including: the target height of the physical support unit The target tilt angle of the display in the interactive feedback unit ; S204: Drive control sub-process, according to... Generate control commands to drive the physical support unit to adjust to the desired height. And drive the interactive feedback unit to adjust to the pitch angle. .

[0018] Furthermore, the central processing module is also configured to execute a multimodal fusion interaction process, which includes: S301: Parallel input acquisition sub-process, synchronously starting the following channels: Voice channel: Audio stream A(t) is acquired through a microphone array and then subjected to noise reduction and enhancement processing; Touch channel: Monitors the touch event flow T(t) of the touch screen; Visual channel: Acquire image sequence I(t) through a designated gesture recognition camera or the main visual sensor in the integrated sensing unit, and run the gesture recognition model; S302: Intent understanding and decision-making sub-process, which integrates and understands multi-channel input: Automatic speech recognition is performed on audio stream A(t) to obtain text candidates. ; Gesture recognition is performed on the image sequence I(t) to obtain the gesture symbol; Design a fusion decision function to determine the final machine instruction (Action) to be executed based on predefined priorities and complementary rules; S303: Adaptive output sub-process, selects the output mode based on user status or instruction content.

[0019] Furthermore, the processing flow of the gesture recognition model includes: Detect the hand region from image sequence I(t); Key points were extracted from the hand region image, resulting in a set of spatiotemporal key point coordinate sequences. ,in Represents the coordinates of the i-th keypoint in the image. and its corresponding timestamp ; Calculate the dynamic features describing a gesture, such as the velocity sequence of keypoint motion trajectories. ,in ; Combine dynamic features with a pre-stored gesture template library The matching process is performed, and the matching degree is calculated using a dynamic time warping algorithm to determine the minimum cumulative distance. : Let the observed gesture trajectory sequence be A template sequence in the template library is Define a local distance metric (Usually Euclidean distance); Dynamic time warping finds a normalized path. This minimizes the cumulative distance along the path: ; in, For sequence and The minimum dynamic time warp distance between them; , Observation sequences and template sequence Length; Sequences and The Middle , 1 eigenvector; This is a local distance function that calculates the distance between two feature vectors; the Euclidean distance is commonly used. ; To normalize the path, use index pairs ( , )composition, =1,2,..., , This represents the path length. and They represent the first and second digits on the path, respectively. During the step, in the sequence and The corresponding index in; For all possible regularized paths Take the minimum value; If a template exists Make If the distance threshold is set, then the corresponding gesture symbol is identified. .

[0020] Furthermore, the central processing module is configured to execute an emergency response priority coordination process, which includes: S501: Emergency Trigger Subprocess, continuously monitors preset emergency trigger signals; S502: Status broadcast and collaborative sub-process. Once an emergency signal is triggered, the emergency protocol message is immediately broadcast through the multi-machine collaborative communication module, and an emergency alarm is sent to the hospital information system through the data interface module. S503: Navigation and Follow Sub-processes: The path planning algorithm of the mobile navigation module immediately switches to "emergency mode," and its cost function... Adjusted to: ; in, To evaluate candidate routes in emergency mode The total cost; This is a candidate path from the starting point to the target point; For path The estimated travel time; For path Roughness or comfort indicators (such as mean curvature, number of sharp turns); , These are the weighting coefficients for time cost and comfort cost, respectively; in emergency rescue mode, they meet the following requirements. ≫ In other words, time efficiency takes a much higher priority than comfort. Actively scan and lock onto the nearest medical personnel or emergency cart target using integrated sensing units; Calculate its relative position to the Target and generate a following trajectory.

[0021] Furthermore, the path planning algorithm in the mobile navigation module, in non-emergency mode, is a dynamically optimized process, including: S601: Basic path planning, static map built based on SLAM Given the user's target G, calculate the initial path using a search algorithm. ; S602: Dynamic information fusion, which receives real-time position and speed information shared by other robots through a multi-robot collaborative communication module; S603: Congestion Prediction and Replanning. Based on the received information, predict the traffic density of each segment of the path in the future time interval Δt and trigger dynamic replanning.

[0022] Furthermore, the central processing module is configured to execute a load-balanced multi-machine service scheduling process, which includes: S701: Load status self-assessment, periodically counting the number of service requests received within a unit time window T. and average request processing time Calculate current load metrics : ; in, The calculated current load index is a dimensionless value; This represents the total number of service requests received within the statistical period T. This is the preset maximum service request processing capacity per unit time. This represents the average time required to process each service request within the statistical period T. The maximum average processing time acceptable for a single service request (preset). , The weighting coefficients for request quantity load and processing time load satisfy... + =1; S702: Load status broadcasting and listening, broadcasting its own load metrics Location information is periodically broadcast via a multi-robot collaborative communication module, and the system listens for broadcast information from other robots. S703: Scheduling decision and execution, if itself Continuously exceeding the high load threshold Then, the robot with the lowest available load is selected from the listening information to initiate a collaboration request.

[0023] Furthermore, the diagnostic assistance linkage process implemented by the data interface module includes: S801: Information retrieval, after the user has been authenticated, retrieves their historical medical records from the hospital information system through a secure link; S802: Symptom correlation analysis, which analyzes the correlation between the user's current symptom description and historical records to calculate a risk warning score; S803: Report generation and push. Based on the analysis results, it automatically generates initial diagnosis linkage suggestion reports and pushes them to the corresponding department doctors' terminals.

[0024] Furthermore, the privacy protection process of the adaptive human-computer interaction module includes: S901: Sensor status control. The vision sensor and depth sensor in the integrated sensing unit are equipped with a physical shielding mechanism in front of them. When the robot is in standby or moving state, the physical shielding mechanism is controlled to be closed. S902: Controlled activation, the physical shielding mechanism is activated and the vision and depth perception functions are initiated only when a valid wake word is recognized by the microphone array or when a close-range human body is detected by the low-power infrared sensor. S903: Session end processing: After the interactive session has clearly ended, delay for a preset time and automatically control the physical shielding mechanism to return to the closed state.

[0025] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0026] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A highly adjustable medical robot equipped with recognition electronic components, characterized in that: Includes the following modules: The mobile navigation module is used to enable the robot to move autonomously and avoid obstacles. It runs a synchronous localization and mapping algorithm to generate and update the environment map in real time, and at the same time receives control commands output by the path planning algorithm to drive the robot chassis to move. The adaptive human-computer interaction module includes an electrically adjustable physical support unit, an integrated sensing unit, and an interactive feedback unit. The physical support unit includes a height-adjustable support structure for supporting the integrated sensing unit and the interactive feedback unit. The integrated sensing unit includes at least a visual sensor for acquiring color images, a depth sensor for acquiring depth information, and a microphone array for acquiring speech. The interactive feedback unit includes at least a touchscreen for information display and a speaker for voice output; The central processing module is connected to the mobile navigation module, the adaptive human-computer interaction module, the multi-machine collaborative communication module, and the data interface module, respectively, and is used to execute control logic and information processing. The multi-robot collaborative communication module is used to wirelessly exchange data with other robots or a central dispatch system in the hospital environment, and share status and map information. The data interface module is used for secure data communication with the hospital information system to retrieve and transmit relevant medical information.

2. The medical robot with highly adjustable identification electronic components according to claim 1, characterized in that: The central processing module is configured to execute a user state adaptive adjustment process, which includes: S201: User perception sub-process, which periodically collects RGB-D data of the scene in front through the visual sensor and depth sensor in the integrated sensing unit; S202: Pose estimation and classification sub-process, which performs the following processing on the acquired RGB-D data; S203: Interactive parameter mapping sub-process, based on the obtained pose category Query the preset mapping table M to obtain the corresponding target interaction parameter set. ; At least including: the target height of the physical support unit The target pitch angle of the display in the interactive feedback unit ; S204: Drive control sub-process, according to... Generate control commands to drive the physical support unit to adjust to the desired height. And drive the interactive feedback unit to adjust to the pitch angle. .

3. The medical robot with highly adjustable identification electronic components according to claim 1, characterized in that: The central processing module is also configured to execute a multimodal fusion interaction process, which includes: S301: Parallel input acquisition subprocess, synchronously starting the following channels; S302: Intent understanding and decision-making sub-process, which integrates and understands multi-channel input; S303: Adaptive output sub-process, selects the output mode based on user status or instruction content.

4. The medical robot with highly adjustable identification electronic components according to claim 1, characterized in that: The processing flow of the gesture recognition model includes: Detect the hand region from image sequence I(t); Key points were extracted from the hand region image, resulting in a set of spatiotemporal key point coordinate sequences. ,in Represents the coordinates of the i-th keypoint in the image. and its corresponding timestamp ; Calculate the dynamic features describing a gesture, such as the velocity sequence of keypoint motion trajectories. ,in ; Combine dynamic features with a pre-stored gesture template library The matching process is performed, and the matching degree is calculated using a dynamic time warping algorithm to determine the minimum cumulative distance. .

5. A highly adjustable medical robot equipped with identification electronic components according to claim 1, characterized in that: The central processing module is configured to execute an emergency priority coordination process, which includes: S501: Emergency Trigger Subprocess, continuously monitors preset emergency trigger signals; S502: Status broadcast and collaborative sub-process: Once an emergency signal is triggered, the emergency protocol message is immediately broadcast through the multi-machine collaborative communication module, and an emergency alarm is sent to the hospital information system through the data interface module. S503: Navigation and Follow Subprocess: The path planning algorithm of the mobile navigation module immediately switched to "emergency mode".

6. A highly adjustable medical robot equipped with identification electronic components according to claim 1, characterized in that: The path planning algorithm in the mobile navigation module, in non-emergency mode, is a dynamic optimization process, including: S601: Basic path planning, static map built based on SLAM Given the user's target G, calculate the initial path using a search algorithm. ; S602: Dynamic information fusion, receiving real-time position and speed information shared by other robots through the multi-machine collaborative communication module; S603: Congestion Prediction and Replanning. Based on the received information, predict the traffic density of each segment of the path in the future time interval Δt and trigger dynamic replanning.

7. A highly adjustable medical robot equipped with identification electronic components according to claim 1, characterized in that: The central processing module is configured to execute a load-balanced multi-machine service scheduling process, which includes: S701: Load status self-assessment, periodically counting the number of service requests received within a unit time window T. and average request processing time Calculate current load metrics : ; in, The calculated current load index is a dimensionless value; This represents the total number of service requests received within the statistical period T. This is the preset maximum service request processing capacity per unit time. This represents the average time required to process each service request within the statistical period T. The maximum average processing time acceptable for a single service request (preset). , The weighting coefficients for request quantity load and processing time load satisfy... + =1; S702: Load status broadcasting and listening, broadcasting its own load metrics The location information is periodically broadcast through the multi-machine collaborative communication module, and the broadcast information of other robots is monitored; S703: Scheduling decision and execution, if itself Continuously exceeding the high load threshold Then, the robot with the lowest available load is selected from the listening information to initiate a collaboration request.

8. A highly adjustable medical robot equipped with identification electronic components according to claim 1, characterized in that: The diagnostic assistance linkage process implemented by the data interface module includes: S801: Information retrieval, after the user has been authenticated, retrieves their historical medical records from the hospital information system through a secure link; S802: Symptom correlation analysis, which analyzes the correlation between the user's current symptom description and historical records to calculate a risk warning score; S803: Report generation and push. Based on the analysis results, it automatically generates initial diagnosis linkage suggestion reports and pushes them to the corresponding department doctors' terminals.

9. A highly adjustable medical robot equipped with identification electronic components according to claim 1, characterized in that: The privacy protection process of the adaptive human-computer interaction module includes: S901: Sensor status control, the vision sensor and depth sensor in the integrated sensing unit are provided with a physical shielding mechanism in front of them; when the robot is in standby or moving state, the physical shielding mechanism is controlled to be closed. S902: Controlled activation, the physical shielding mechanism is activated and the vision and depth perception functions are started only when a valid wake word is recognized by the microphone array or when a close human body is detected by the low-power infrared sensor. S903: Session end processing: After the interactive session has clearly ended, delay for a preset time and automatically control the physical shielding mechanism to return to the closed state.

Images