Intelligent learning accompanying robot

Abstract

The invention discloses an intelligent learning accompanying robot which comprises a Bluetooth earphone and a robot body. The Bluetooth earphone has a voice obtaining function; the robot body comprises a touch module, a voice receiving module, a voice learning system, a central controller, an instruction storage unit and a wireless information transmission module; the robot body further comprises a life accompanying module, a health management module, an emergency module, a power module and an electric quantity detection module. The life accompanying module comprises a voice interaction module, a voice communication module and an AI emotion accompanying module; the health management module is used for measuring health data of specified hospitalized old people and managing the health data; and the emergency module is used for calling for help through voice or touch operation and sending a calling signal to the supervisor end and the medical staff end. According to the invention, intelligent voice learning can be carried out, voice instructions of a user can be accurately recognized, functions are rich, and health monitoring and accompanying are both considered.

Description

Technical Field

[0001] The present invention relates to the technical field of companion robots, and in particular to an intelligent learning companion robot. Background Art

[0002] In the current market environment, companion robots are gradually gaining widespread attention as an innovative solution to cope with an aging society. With the advancement of technology and the improvement of the quality of life requirements of the elderly, the types and functions of companion robots are constantly enriched, forming a diversified product ecosystem. Companion robots can be roughly divided into several categories:

[0003] Emotional companion robots, such as Paro (seal robot) and Aibo (Sony robot dog), mainly provide emotional support through cute appearance and natural behavior patterns. These products focus on emotional interaction with users to help the elderly reduce loneliness. Although their functions are relatively simple, they play a positive role in meeting the emotional needs of the elderly. Health monitoring robots, such as ElliQ and Robear, combine health monitoring functions to collect and analyze the health data of the elderly in real time. This type of robot not only provides health advice, but also reminds the elderly to take medicine on time or do appropriate exercises to help them better manage their health. This type of product is especially favored by family caregivers because it can effectively reduce the burden of care. Multifunctional intelligent robots, such as Buddy and Leka, integrate multiple functions such as emotional companionship, health monitoring, entertainment and social interaction, and can interact with users in a rich way. These robots can not only play music and videos, but also connect through social media to enhance the connection between the elderly and their families and promote social interaction. Service robots, such as Care-O-bot and Pepper, are mainly used to provide services and support, including information query, smart home control and emergency calls. These products interact with users through voice and have certain emotional communication capabilities, but their core functions still focus on service.

[0004] Although the companion robot market has broad prospects, it still faces some challenges. The companion robots on the market have a single function and cannot take into account both health monitoring and companionship. Companion robots often work by recognizing the user's voice commands. However, when some users speak with an accent or slur their words, and when different environments and users' physical conditions produce different sounds, companion robots often cannot recognize these sounds, making the companion robots unusable. Especially when the user is in danger, if the companion robot is far away from the user and cannot distinguish the voice command for help, it is easy to cause danger to the user. Summary of the invention

[0005] In order to overcome the above-mentioned shortcomings of the prior art, the present invention proposes an intelligent learning companion robot that can perform intelligent voice learning, accurately recognize the user's voice commands, has rich functions, and takes into account both health monitoring and companionship.

[0006] The technical solution adopted by the present invention to solve the technical problem is: an intelligent learning companion robot, including a Bluetooth headset and a robot body;

[0007] The Bluetooth headset has a voice acquisition function;

[0008] The robot body includes: a touch module, a voice receiving module, a voice learning system, a central controller, an instruction storage unit and a wireless information transmission module; the touch module is used to receive the user's touch signal; the Bluetooth communication module is wirelessly connected to the Bluetooth headset to receive the sound signal emitted by the user through the Bluetooth headset; the voice learning system includes a voice receiving system, a voice recognition module, an A / D converter and a memory; the voice receiving system includes an external collection device and a voice receiver; the collection device is provided with a voice learning button and a learning module; the voice learning button can transmit the sound signal collected by the collection device to the voice receiver and can perform voice learning through the voice learning module; the voice receiver is respectively connected to the Bluetooth communication module and the central controller; the central controller is respectively connected to the voice recognition module and the learning module; the voice recognition module is connected to the A / D converter; the A / D converter is connected to the memory; the instruction storage unit is used to store instructions and the touch signals and sound control signals corresponding to the instructions; the central controller controls the robot body according to the sound control signal or the touch signal compared with the instruction storage unit output corresponding instructions;

[0009] The robot body also includes a life companionship module, a health management module, an emergency module, a power module, and a power detection module; the life companionship module includes a voice interaction module, a voice communication module, and an AI emotional companionship module; the health management module is used to measure the health data of designated hospitalized elderly people and manage the health data; the emergency module is used to call for help through voice or touch operation and send a distress signal to the supervisor and medical staff; the power module is rechargeable and used to power the robot body; the power detection module is used to detect the power of the robot body; the health management module includes a health monitoring module, and the health monitoring module is connected to several health monitoring devices; the health monitoring devices include a heart rate measurement device, a blood oxygen measurement device, a blood pressure measurement device, and a microcirculation measurement device.

[0010] Furthermore, the speech learning specifically includes the following steps:

[0011] (1) Create a source data domain in the memory The source data domain is a data domain obtained by converting the standard voice of the preset voice command through an A / D converter; each source data domain data corresponds to a voice command; wherein the nth source sample data is

[0012] (2) Source data domain V S Obtain clustering function through offline learning

[0013] (3) When the user presses the voice learning button and issues a voice command to the companion robot, the central controller receives the command and obtains the learning sample domain data through the voice receiving system, voice recognition module, and A / D converter. The learning sample domain data is compared with the source data domain V S If there is exactly the same data, stop learning; if not exactly the same, proceed to step (4);

[0014] (4) Learning sample domain data Divided into two parts and in, The source data domain V S Learning sample domain data with the same characteristics; The learning sample domain data has different characteristics from the source sample data;

[0015] (5) Predicting learning sample domain data The tags are: Among them l t To learn the sample domain dataset, To learn the clustering function of sample domain data, It is a data label that expresses the characteristics of the sample domain data;

[0016] (6) Develop label correction strategies;

[0017] (7) According to The same feature clustering function is calculated as in step (5) according to And step (5) calculates different feature clustering functions

[0018] (8) Design the same feature loss function and the same feature loss condition, calculate the same feature loss function, and if the same feature loss condition is met, Clustering functions with the same features as training samples Conduct training;

[0019] (9) Design different feature loss functions and different feature loss conditions, calculate different feature loss functions, and if different feature loss conditions are met, As training samples, clustering functions of different features Conduct training;

[0020] (10) Perform label correction on the training set data according to the label correction strategy;

[0021] (11) Assign weight values ​​α, β, γ according to the clustering function obtained in step (2) Same feature clustering function Different feature clustering functions Construct a clustering function:

[0022]

[0023] (12) The clustering function of step (11) is trained to obtain the weight δ; the training model of the sample labels is obtained:

[0024]

[0025] (13) The sample label f t (l t ) is stored in the memory, and the next time the user issues a voice command, it can be converted into a corresponding sound control signal through the sample label.

[0026] Furthermore, the robot body also includes an audio-visual entertainment module; the audio-visual entertainment module can play music, animations, and TV series.

[0027] Furthermore, the health management module also includes sleep monitoring. The sleep monitoring module is connected to the sleep monitoring equipment to collect the sleep data of the elderly and generate a sleep report.

[0028] Furthermore, the robot body also includes a data analysis module; the data analysis module performs analysis reports based on the health data and obtains the cycle and examination items for regular health examinations for the elderly.

[0029] Furthermore, the emergency module also includes a fall detection and warning module, which is connected to the millimeter wave radar device. When the user falls, it immediately detects and triggers the generation of an alarm signal which is sent to a designated terminal through a wireless information transmission module.

[0030] Furthermore, the natural voice interaction module is used to process natural language, so that the user can interact with the robot through voice, can identify the user's basic needs from the voice, and conduct conversations based on the preset robot personality that the user likes. The voice communication module is used to conduct real-time voice interaction with the supervisor, make two-way voice calls, and support the robot body to initiate voice calls to designated terminals. The AI ​​emotional companionship module is used to analyze the user's voice, expression and behavior patterns, monitor the user's emotional state in real time, and issue reminders in time.

[0031] Furthermore, it also includes a management background; the management background includes a cloud data server, a digital health record module, and a health data analysis system; the data server is used to receive the user's health data; the digital health record module is used to establish a digital health record based on the user's health data; the health data analysis system is used to analyze the user's condition based on the health record and output a report.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. The present invention can provide intelligent emotional companionship and basic life care, effectively alleviate the user's loneliness, improve the quality of life, and take into account health monitoring and companionship. It can not only provide emotional companionship and reduce loneliness for users, but also combine health monitoring with emergency call functions to provide all-round care support. Through intelligent interaction, the elderly can enjoy more personalized entertainment and social experiences, improve mental health, reduce mental stress, and thus improve the overall quality of life.

[0034] 2. The present invention is provided with a voice learning function. Through the voice learning button, the user has the function of training the robot voice learning. Voice learning can prevent the user from speaking with an accent or slurred speech. In different environments and when the user's physical condition is different and different sounds are made, the companion robot can also recognize these sounds and be used normally. When the user is in danger, the companion robot will not be unable to distinguish the voice command for help because the user is far away, thereby avoiding danger to the user.

[0035] 3. The present invention is provided with an emergency module, which can call for help through voice or touch operation and send a distress signal to the supervisor and medical staff, and can also be combined with the Internet of Things to sense when a user falls. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for describing the embodiments are briefly introduced below.

[0037] Figure 1 This is a structural principle diagram of an intelligent learning companion robot according to an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all of the embodiments. Based on the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] In the description of the present invention, it should be noted that the terms "center", "up", "down", "left", "right", "vertical", "horizontal", "inside", "outside", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying 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 direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on the present invention.

[0040] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0041] like Figure 1 As shown, an embodiment of the present invention is an intelligent learning companion robot, including a Bluetooth headset, a robot body, and a management background;

[0042] The Bluetooth headset has the function of acquiring voice;

[0043] The robot body includes. A touch module, a voice receiving module, a voice learning system, a central controller, an instruction storage unit and a wireless information transmission module. The touch module is used to receive the user's touch signal. The Bluetooth communication module is wirelessly connected to the Bluetooth headset to receive the sound signal emitted by the user through the Bluetooth headset; the voice learning system includes a voice receiving system, a voice recognition module, an A / D converter and a memory. The voice receiving system includes an external collection device and a voice receiver. The collection device is provided with a voice learning button and a learning module. The voice learning button can transmit the sound signal collected by the collection device to the voice receiver and can perform voice learning through the voice learning module. The voice receiver is respectively connected to the Bluetooth communication module and the central controller. The central controller is respectively connected to the voice recognition module and the learning module. The voice recognition module is connected to the A / D converter. The A / D converter is connected to the memory; the instruction storage unit is used to store instructions and touch signals and sound control signals corresponding to the instructions. The central controller controls the robot body according to the sound control signal or the touch signal compared with the instruction storage unit output corresponding instructions;

[0044] The robot body also includes a life companion module, a health management module, an emergency module, a power module, and a power detection module. The life companion module includes a voice interaction module, a voice communication module, and an AI emotional companion module. The health management module is used to measure the health data of designated hospitalized elderly people and manage the health data; the emergency module is used to call for help through voice or touch operation and send a distress signal to the supervisor and medical staff; the power module is rechargeable and used to power the robot body. The power detection module is used to detect the power of the robot body. The health management module includes a health monitoring module, which is connected to several health monitoring devices. Health monitoring equipment includes a heart rate measurement device, a blood oxygen measurement device, a blood pressure measurement device, and a microcirculation measurement device.

[0045] Speech learning specifically includes the following steps:

[0046] (1) Create a source data domain in the memory The source data domain is a data domain obtained by converting the standard voice of the preset voice command through the A / D converter; each source data domain data corresponds to a voice command; among which, the nth source sample data is

[0047] (2) Source data domain V S Obtain clustering function through offline learning

[0048] (3) When the user presses the voice learning button and issues a voice command to the companion robot, the central controller receives the command and obtains the learning sample domain data through the voice receiving system, voice recognition module, and A / D converter. The learning sample domain data is compared with the source data domain V S If there is exactly the same data, stop learning; if not exactly the same, proceed to step (4);

[0049] (4) Learning sample domain data Divided into two parts and in, The source data domain V S Learning sample domain data with the same characteristics; The learning sample domain data has different characteristics from the source sample data;

[0050] (5) Predicting learning sample domain data The tags are: Among them l t To learn the sample domain dataset, To learn the clustering function of sample domain data, It is a data label that expresses the characteristics of the sample domain data;

[0051] (6) Develop label correction strategies;

[0052] (7) According to And step (5) calculate the same feature clustering function according to And step (5) calculates different feature clustering functions

[0053] (8) Design the same feature loss function and the same feature loss condition, calculate the same feature loss function, and if the same feature loss condition is met, Clustering functions with the same features as training samples Conduct training;

[0054] (9) Design different feature loss functions and different feature loss conditions, calculate different feature loss functions, and if different feature loss conditions are met, As training samples, clustering functions of different features Conduct training;

[0055] (10) Perform label correction on the training set data according to the label correction strategy;

[0056] (11) Assign weight values ​​α, β, γ according to the clustering function obtained in step (2) Same feature clustering function Different feature clustering functions Construct a clustering function:

[0057]

[0058] (12) The clustering function of step (11) is trained to obtain the weight δ; the training model of the sample labels is obtained:

[0059]

[0060] (13) Set the sample label f t (l t ) is stored in the memory, and the next time the user issues a voice command, it can be converted into a corresponding sound control signal through the sample label.

[0061] The robot body also includes an audio-visual entertainment module, which can play music, animations, and TV series.

[0062] The health management module also includes sleep monitoring. The sleep monitoring module is connected to the sleep monitoring equipment to collect the sleep data of the elderly and generate sleep reports.

[0063] The robot body also includes a data analysis module. The data analysis module makes an analysis report based on the health data and obtains the cycle and examination items of regular health examinations for the elderly.

[0064] The emergency module also includes a fall detection and warning module, which is connected to the millimeter-wave radar equipment. When the user falls, it immediately detects and triggers the generation of an alarm signal which is sent to the designated terminal through the wireless information transmission module.

[0065] The natural speech interaction module is used for natural language processing, allowing users to interact with the robot through voice, identifying the user's basic needs from the voice, and communicating based on the preset robot personality that the user likes. The voice communication module is used for real-time voice interaction with the supervisor, two-way voice calls, and supports the robot to initiate voice calls to designated terminals. The AI ​​emotional companionship module is used to analyze the user's voice, expression and behavior patterns, monitor the user's emotional state in real time, and issue reminders in time.

[0066] The management backend includes a cloud data server, a digital health record module, and a health data analysis system. The data server is used to receive the user's health data. The digital health record module is used to create a digital health record based on the user's health data. The health data analysis system is used to analyze the user's condition based on the health record and output a report.

[0067] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed by the present invention should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. An intelligent learning companion robot, characterized in that: Including Bluetooth headset and robot body; The Bluetooth headset has a voice acquisition function; The robot body includes: a touch module, a voice receiving module, a voice learning system, a central controller, an instruction storage unit and a wireless information transmission module; the touch module is used to receive the user's touch signal; the Bluetooth communication module is wirelessly connected to the Bluetooth headset to receive the sound signal emitted by the user through the Bluetooth headset; the voice learning system includes a voice receiving system, a voice recognition module, an A / D converter and a memory; the voice receiving system includes an external collection device and a voice receiver; the collection device is provided with a voice learning button and a learning module; the voice learning button can transmit the sound signal collected by the collection device to the voice receiver and can perform voice learning through the voice learning module; the voice receiver is respectively connected to the Bluetooth communication module and the central controller; the central controller is respectively connected to the voice recognition module and the learning module; the voice recognition module is connected to the A / D converter; the A / D converter is connected to the memory; the instruction storage unit is used to store instructions and the touch signals and sound control signals corresponding to the instructions; the central controller controls the robot body according to the sound control signal or the touch signal compared with the instruction storage unit output corresponding instructions; The robot body also includes a life companionship module, a health management module, an emergency module, a power module, and a power detection module; the life companionship module includes a voice interaction module, a voice communication module, and an AI emotional companionship module; the health management module is used to measure the health data of designated hospitalized elderly people and manage the health data; the emergency module is used to call for help through voice or touch operation and send a distress signal to the supervisor and medical staff; the power module is rechargeable and used to power the robot body; the power detection module is used to detect the power of the robot body; the health management module includes a health monitoring module, and the health monitoring module is connected to several health monitoring devices; the health monitoring devices include a heart rate measurement device, a blood oxygen measurement device, a blood pressure measurement device, and a microcirculation measurement device.

2. The intelligent learning companion robot according to claim 1, characterized in that: The speech learning specifically comprises the following steps: (1) Create a source data domain in the memory The source data domain is a data domain obtained by converting the standard voice of the preset voice command through an A / D converter; each source data domain data corresponds to a voice command; wherein the nth source sample data is (2) Source data domain V S Obtain clustering function through offline learning (3) When the user presses the voice learning button and issues a voice command to the companion robot, the central controller receives the command and obtains the learning sample domain data through the voice receiving system, voice recognition module, and A / D converter. The learning sample domain data is compared with the source data domain V S If there is exactly the same data, stop learning; if not exactly the same, proceed to step (4); (4) Learning sample domain data Divided into two parts and in, V S Learning sample domain data with the same characteristics; The learning sample domain data has different characteristics from the source sample data; (5) Predicting learning sample domain data The tags are: Among them l t To learn the sample domain dataset, To learn the clustering function of sample domain data, It is a data label that expresses the characteristics of the sample domain data; (6) Develop label correction strategies; (7) According to The same feature clustering function is calculated as in step (5) according to And step (5) calculates different feature clustering functions (8) Design the same feature loss function and the same feature loss condition, calculate the same feature loss function, and if the same feature loss condition is met, Clustering functions with the same features as training samples Conduct training; (9) Design different feature loss functions and different feature loss conditions, calculate different feature loss functions, and if different feature loss conditions are met, As training samples, clustering functions of different features Conduct training; (10) Perform label correction on the training set data according to the label correction strategy; (11) Assign weight values ​​α, β, γ according to the clustering function obtained in step (2) Same feature clustering function Different feature clustering functions Construct a clustering function: (12) The clustering function of step (11) is trained to obtain the weight δ; the training model of the sample labels is obtained: (13) The sample label f t (l t ) is stored in the memory, and the next time the user issues a voice command, it can be converted into a corresponding sound control signal through the sample label.

3. The intelligent learning companion robot according to claim 1, characterized in that: The robot body also includes an audio-visual entertainment module; the audio-visual entertainment module can play music, animations, and TV series.

4. The intelligent learning companion robot according to claim 1, characterized in that: The health management module also includes sleep monitoring. The sleep monitoring module is connected to the sleep monitoring equipment to collect the sleep data of the elderly and generate a sleep report.

5. The intelligent learning companion robot according to claim 1, characterized in that: The robot body also includes a data analysis module; the data analysis module performs analysis reports based on the health data and obtains the cycle and examination items for regular health examinations for the elderly.

6. The intelligent learning companion robot according to claim 1, characterized in that: The emergency module also includes a fall detection and warning module, which is connected to the millimeter wave radar device. When a user falls, it immediately detects and triggers the generation of an alarm signal which is sent to a designated terminal via a wireless information transmission module.

7. The intelligent learning companion robot according to claim 1, characterized in that: The natural voice interaction module is used to process natural language, allowing users to interact with the robot through voice, identifying the user's basic needs from the voice, and conversing based on the preset robot personality that the user likes. The voice communication module is used to conduct real-time voice interaction with the supervisor, make two-way voice calls, and support the robot body to initiate voice calls to designated terminals. The AI ​​emotional companionship module is used to analyze the user's voice, expression and behavior patterns, monitor the user's emotional state in real time, and issue reminders in time.

8. The intelligent learning companion robot according to claim 1, characterized in that: It also includes a management backend; the management backend includes a cloud data server, a digital health record module, and a health data analysis system; the data server is used to receive the user's health data; The digital health record module is used to establish a digital health record based on the user's health data; the health data analysis system is used to analyze the user's condition based on the health record and output a report.

Images