Internet of Things-Based Companion Robot System and Its Remote Monitoring Method

Through the Internet of Things-based escort robot system, data interaction, analysis and execution modules are used, and escort strategy adjustments are combined with machine learning, which solves the problem of insufficient interaction and remote monitoring of existing escort robots, and improves the user's sense of security and satisfaction.

CN118721225BActive Publication Date: 2025-07-25SHENZHEN MUCHY INTERNET OF THINGS
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202410891364.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-04
Publication Date
2025-07-25
Estimated Expiration
2044-07-04

AI Technical Summary

Technical Problem

The existing accompanying robots lack the interactive functions and remote monitoring capabilities with other devices, resulting in inflexible adjustment of accompanying strategies and low user security and satisfaction.

Method used

Design a companion robot system based on the Internet of Things, including data interaction module, data analysis module, remote monitoring module and robot execution module. By obtaining multi-dimensional user data for comprehensive analysis, combining machine learning to determine the companion strategy, and transmitting it to the remote monitoring module through the Internet of Things for adjustment and execution.

Benefits of technology

Accurate adjustment of accompanying strategies has been achieved, users' sense of security and satisfaction have been improved, and the accuracy and flexibility of accompanying strategies have been ensured through multi-dimensional data analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118721225B_ABST
    Figure CN118721225B_ABST
Patent Text Reader

Abstract

The present invention provides an escort robot system based on the Internet of Things and its remote monitoring method. By means of a data interaction module, physical sign data and posture data of a user in a wearable device, living environment data of the user and language data of the user in an environment detection device are acquired, multi-dimensional user information is obtained, providing a data basis for precise escort. Through a data analysis module, the physical sign data, posture data, living environment data and language data of the user are comprehensively analyzed to ensure the accuracy of the obtained user requirements. Through a remote monitoring module, the obtained user requirements and the current state of the user are acquired and transmitted, and combined with machine learning, an escort robot strategy for the user is determined, and the escort strategy is flexibly adjusted to ensure the accuracy of the obtained escort robot strategy. Through a robot execution module, according to the escort robot strategy, an operation instruction of the escort robot is determined and executed according to the operation instruction, improving the sense of security and satisfaction of the user.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of the Internet of Things, and particularly to an escort robot system based on the Internet of Things and a remote monitoring method thereof. Background Art

[0002] As one of the greatest inventions in the 20th century, robot technology has made great progress in industrial production and manufacturing. With the development of society and technology, it has gradually moved from industrial robots in production practice to all aspects of human life. Service robots mainly refer to robots that provide services for humans, and they should have the characteristics of getting along harmoniously with humans, being small in size, flexible, and beautiful. As a member of the robot family, the escort service robot mainly provides necessary services for those who have no self-care ability or the elderly, such as pouring water, getting medicine, chatting, checking the body, etc. in daily life, and can replace young people in these aspects to relieve their pressure. Most of the escort robots on the market mainly complete various functions after the user actively issues instructions.

[0003] Currently, most of the escort robots on the market do not have the function of interacting with other devices and cannot perform remote monitoring, so they cannot flexibly adjust the escort strategy and achieve more accurate escort services, and the sense of security and satisfaction of users need to be improved. Summary of the Invention

[0004] The present invention provides an escort robot system based on the Internet of Things and a remote monitoring method thereof to solve the problems raised in the background art.

[0005] An escort robot system based on the Internet of Things includes:

[0006] A data interaction module, configured to obtain the physical sign data and posture data of the user in the wearable device, obtain the living environment data of the user in the environment detection device, and obtain the language data of the user;

[0007] A data analysis module, configured to comprehensively analyze the physical sign data, posture data, living environment data and language data of the user to determine the user's needs;

[0008] A remote monitoring module, configured to obtain the transmitted user needs and the current state of the user, and combine machine learning to determine the robot escort strategy for the user;

[0009] A robot execution module, configured to determine the operation instructions of the escort robot according to the robot escort strategy and execute according to the operation instructions.

[0010] Preferably, it further includes: a transmission module, configured to transmit the user needs and the current state of the user to the remote monitoring module based on the Internet of Things;

[0011] The transmission module includes:

[0012] A transmission packet acquisition unit, configured to pack the user requirements and the user's current status according to a preset data packing method to obtain a data transmission packet;

[0013] A transmission unit, configured to determine a transmission method based on the Internet of Things, and transmit the data transmission packet to a remote monitoring module through the Internet of Things according to the transmission method.

[0014] Preferably, the data interaction module includes:

[0015] A wearable device interaction unit, configured to acquire the collected sensing data about the user in the wearable device, remove abnormal data from the collected sensing data, and perform classification processing according to the data type to obtain the user's physical sign data and posture data;

[0016] An environmental device interaction unit, configured to acquire the environmental sensing data about the user from the environmental detection device, remove abnormal data from the environmental sensing data, and perform classification processing according to the data type to obtain the user's living environment data;

[0017] A voice interaction unit, configured to acquire the voice sensing data about the user from the voice acquisition device, and parse the voice sensing data to obtain the language data about the user.

[0018] Preferably, the data analysis module includes:

[0019] A requirement determination unit, configured to obtain keywords from the language data, match the keywords with a requirement database, and obtain the number of requirements with a matching degree greater than a preset matching degree;

[0020] Determine whether the number of requirements is greater than 1;

[0021] If so, obtain the corresponding requirement set;

[0022] Otherwise, obtain the corresponding initial requirement;

[0023] An action analysis unit, configured to establish a physical sign and posture structure of the user based on the physical sign data and the posture data, and perform dynamic action restriction on the physical sign and posture structure based on the physical sign data and the posture data to obtain action restriction information;

[0024] A requirement screening unit, configured to obtain the standard action features corresponding to the requirements in the requirement set, judge the standard action features based on the action restriction information, and select a target requirement set that meets the action restriction information;

[0025] A reliability determination unit, configured to determine the current user state based on physical sign data, posture data, and language data, and assign a first reliability to the physical sign data and the posture data, and a second reliability to the language data based on the state difference between the current user state and the historical average user state;

[0026] An evaluation unit, configured to evaluate the target requirement set based on the first reliability and the second reliability, and select the requirement with the highest evaluation value as the initial requirement;

[0027] An environment analysis unit, configured to obtain the standard environment data required by the initial requirement, compare the standard environment data with the living environment data to obtain an environment difference, and determine whether the environment difference affects the execution of the initial requirement;

[0028] If so, based on the environment difference, finely adjust the initial requirement to obtain the user requirement;

[0029] Otherwise, use the initial requirement as the user requirement.

[0030] Preferably, the requirement screening unit includes:

[0031] A judgment unit, configured to judge whether the standard action feature contains the action feature in the action restriction information;

[0032] If so, determine that the requirement corresponding to the standard action feature does not meet the action restriction information;

[0033] Otherwise, determine that the requirement corresponding to the standard action feature meets the action restriction information;

[0034] An integration unit, configured to integrate the requirements that meet the action restriction information to obtain a target requirement set.

[0035] Preferably, the remote monitoring module includes:

[0036] A label configuration unit, configured to divide the user requirement based on the requirement unit of machine learning to obtain multiple minimum sub-requirements, and configure a first specified label for the minimum sub-requirements, divide the current user state based on the state unit of machine learning to obtain multiple minimum sub-states, and configure a second specified label for the minimum sub-states;

[0037] A model training unit, configured to perform model training based on machine learning based on the previously obtained first label and second label to obtain a policy matching model;

[0038] A policy determination unit, configured to input the first specified label into the policy matching model to obtain a first escort policy, and input the second specified label into the policy matching model to obtain a second escort policy;

[0039] A key determination unit for comparing the first escort strategy and the second escort strategy, determining the first escort key points of the first escort strategy and the second escort key points of the second escort strategy, and integrating the first escort key points and the second escort key points to obtain the target escort key points;

[0040] A selection and integration unit for selecting escort strategy points from the first escort strategy and the second escort strategy respectively based on the target escort key points, and integrating the escort strategy points to obtain a robot escort strategy for the user.

[0041] Preferably, the model training unit includes:

[0042] A training unit for obtaining the escort strategy points corresponding to the first label and the second label, inputting the first label and its corresponding escort strategy points into the initial machine learning model for training to obtain a demand matching model, and inputting the second label and its corresponding escort strategy points into the initial machine learning model for training to obtain a state matching model;

[0043] A fusion unit for fusing the demand matching model and the state matching model based on a preset model fusion method to obtain a strategy matching model.

[0044] Preferably, the robot execution module includes:

[0045] An instruction determination unit for parsing the robot escort strategy, determining the escort process of the escort robot, and matching the operation instructions related to the escort process from the operation instruction library;

[0046] An execution unit for obtaining the execution resources required for the operation instructions and executing the escort robot in combination with the execution resources.

[0047] Preferably, the execution unit includes:

[0048] A monitoring unit for monitoring the execution process of the robot to obtain monitoring data;

[0049] An early warning unit for performing corresponding early warning reminders based on the relationship between the monitoring data and the monitoring threshold.

[0050] A remote monitoring method for an escort robot system based on the Internet of Things, including:

[0051] S1: Obtaining the physical sign data and posture data of the user in the wearable device, obtaining the living environment data of the user in the environment detection device, and obtaining the language data of the user;

[0052] S2: Comprehensively analyzing the physical sign data, posture data, living environment data and language data of the user to determine the user's needs;

[0053] S3: Obtain the user requirements and the user's current status obtained through transmission, and combine with machine learning to determine the robot escort strategy for the user;

[0054] S4: According to the robot escort strategy, determine the operation instructions for the escort robot and execute them according to the operation instructions.

[0055] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0056] Obtain the physical sign data and posture data of the user in the wearable device, the living environment data of the user and the user's language data in the environmental detection device through the data interaction module, obtain multi-dimensional user information, and provide a data basis for accurate escort. Through the data analysis module, comprehensively analyze the user's physical sign data, posture data, living environment data and language data to determine the user requirements and ensure the accuracy of obtaining the user requirements. Through the remote monitoring module, obtain the transmitted user requirements and the user's current status, combine with machine learning to determine the robot escort strategy for the user, flexibly adjust the escort strategy, and ensure the accuracy of the obtained robot escort strategy. Through the robot execution module, according to the robot escort strategy, determine the operation instructions for the escort robot and execute them according to the operation instructions, providing the user's sense of security and satisfaction.

[0057] Other features and advantages of the present invention will be described in the subsequent specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0058] The following will further describe the technical solutions of the present invention in detail through the drawings and embodiments. Description of the Drawings

[0059] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation to the present invention. In the drawings:

[0060] Figure 1 It is a structural diagram of an escort robot system based on the Internet of Things in an embodiment of the present invention;

[0061] Figure 2 It is a structural diagram of the data interaction module in an embodiment of the present invention;

[0062] Figure 3 It is a flowchart of a remote monitoring method for an escort robot system based on the Internet of Things in an embodiment of the present invention. Detailed Embodiments

[0063] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only for the purpose of illustrating and explaining the present invention, and are not intended to limit the present invention.

[0064] Embodiment 1:

[0065] An embodiment of the present invention provides an Internet of Things-based companion robot system, as Figure 1 shown, including:

[0066] A data interaction module, configured to obtain the physical sign data and posture data of the user in the wearable device, obtain the living environment data of the user in the environmental detection device, and obtain the language data of the user;

[0067] A data analysis module, configured to comprehensively analyze the physical sign data, posture data, living environment data, and language data of the user to determine the user's needs;

[0068] A remote monitoring module, configured to obtain the transmitted user needs and the current state of the user, and combine machine learning to determine the robot companion strategy for the user;

[0069] A robot execution module, configured to determine the operation instructions of the companion robot according to the robot companion strategy, and execute according to the operation instructions.

[0070] In this embodiment, the robot companion strategy for the user is obtained through analysis by a machine learning algorithm in combination with machine learning.

[0071] The beneficial effects of the above design solution are as follows: The data interaction module obtains the physical sign data and posture data of the user in the wearable device, the living environment data of the user in the environmental detection device, and the language data of the user, obtaining multi-dimensional user information, providing a data basis for accurate companionship. The data analysis module comprehensively analyzes the physical sign data, posture data, living environment data, and language data of the user to determine the user's needs, ensuring the accuracy of obtaining the user's needs. The remote monitoring module obtains the transmitted user needs and the current state of the user, and combines machine learning to determine the robot companion strategy for the user, flexibly adjusting the companion strategy to ensure the accuracy of the obtained robot companion strategy. The robot execution module determines the operation instructions of the companion robot according to the robot companion strategy and executes according to the operation instructions, providing the user's sense of security and satisfaction.

[0072] Embodiment 2:

[0073] Based on Embodiment 1, an embodiment of the present invention provides an Internet of Things-based companion robot system, further including: a transmission module, configured to transmit the user needs and the current state of the user to the remote monitoring module based on the Internet of Things;

[0074] The transmission module includes:

[0075] A transmission packet acquisition unit, configured to pack the user requirements and the user's current status according to a preset data packing method to obtain a data transmission packet;

[0076] A transmission unit, configured to determine a transmission method based on the Internet of Things, and transmit the data transmission packet to the remote monitoring module through the Internet of Things according to the transmission method.

[0077] In this embodiment, the transmission method is determined according to the Internet speed of the Internet of Things, etc., and includes single-group transmission and multi-group transmission.

[0078] The beneficial effect of the above design solution is that by determining the transmission method based on the Internet of Things and transmitting the data transmission packet to the remote monitoring module through the Internet of Things according to the transmission method, the function of remote monitoring is realized, providing a basis for flexibly adjusting the escort strategy and realizing more accurate escort services.

[0079] Embodiment 3:

[0080] Based on Embodiment 1, an embodiment of the present invention provides an escort robot system based on the Internet of Things, as Figure 2 shown, the data interaction module includes:

[0081] A wearing device interaction unit, configured to acquire the collected sensing data about the user in the wearing device, remove abnormal data from the collected sensing data, and perform classification processing according to the data type to obtain the physical sign data and posture data of the user;

[0082] An environmental device interaction unit, configured to acquire the environmental sensing data about the user from the environmental detection device, remove abnormal data from the environmental sensing data, and perform classification processing according to the data type to obtain the living environment data of the user;

[0083] A voice interaction unit, configured to acquire the voice sensing data about the user from the voice acquisition device, and parse the voice sensing data to obtain the language data about the user.

[0084] The beneficial effect of the above design solution is that by acquiring the physical sign data and posture data about the user in the wearing device, acquiring the living environment data about the user in the environmental detection device, and acquiring the language data about the user, the interaction with multiple other devices is realized, and multi-dimensional user information is obtained, providing a data basis for accurate escort.

[0085] Embodiment 4:

[0086] Based on Embodiment 1, an embodiment of the present invention provides an escort robot system based on the Internet of Things, the data analysis module includes:

[0087] A requirement determination unit, configured to obtain keywords from language data, match the keywords with a requirement database, and obtain the number of requirements with a matching degree greater than a preset matching degree;

[0088] Determine whether the number of requirements is greater than 1;

[0089] If so, obtain the corresponding requirement set;

[0090] Otherwise, obtain the corresponding initial requirement;

[0091] An action analysis unit, configured to establish a physical sign and posture structure of a user based on physical sign data and posture data, and perform dynamic action restriction on the physical sign and posture structure based on the physical sign data and the posture data to obtain action restriction information;

[0092] A requirement screening unit, configured to obtain standard action features corresponding to requirements in the requirement set, judge the standard action features based on the action restriction information, and select a target requirement set that meets the action restriction information;

[0093] A reliability determination unit, configured to determine the current user state based on physical sign data, posture data, and language data, and assign a first reliability to the physical sign data and the posture data and a second reliability to the language data based on the state difference between the current user state and the historical average user state;

[0094] An evaluation unit, configured to evaluate the target requirement set based on the first reliability and the second reliability, and select the requirement with the highest evaluation value as the initial requirement;

[0095] An environment analysis unit, configured to obtain standard environment data required for the initial requirement, compare the standard environment data with living environment data to obtain an environment difference, and judge whether the environment difference affects the execution of the initial requirement;

[0096] If so, perform fine-tuning on the initial requirement based on the environment difference to obtain the user requirement;

[0097] Otherwise, use the initial requirement as the user requirement.

[0098] In this embodiment, the action restriction information is specifically determined according to the user's physical condition and state. For example, the user cannot perform strenuous exercise currently, etc.

[0099] In this embodiment, the greater the relevance between the state difference and the physical sign and posture, the greater the first reliability, and the greater the relevance between the state difference and the language feature, the greater the second reliability.

[0100] In this embodiment, based on environmental differences, the initial requirements are fine-tuned. For example, when the environmental difference is a weather difference, the initial requirements can be transferred from outdoors to indoors.

[0101] The beneficial effects of the above design are as follows: By comprehensively analyzing from multiple aspects including language, physical signs, postures, and the surrounding environment to determine user requirements, the accuracy of the obtained user requirements is ensured, providing a basis for subsequent accurate companionship.

[0102] Embodiment 5:

[0103] Based on Embodiment 4, an Internet of Things-based companion robot system is provided in an embodiment of the present invention. The requirement screening unit includes:

[0104] A judgment unit for judging whether the standard action feature contains the action feature in the action limit information;

[0105] If so, it is determined that the requirement corresponding to the standard action feature does not meet the action limit information;

[0106] Otherwise, it is determined that the requirement corresponding to the standard action feature meets the action limit information;

[0107] An integration unit for integrating the requirements that meet the action limit information to obtain a target requirement set.

[0108] The beneficial effects of the above design are as follows: By comprehensively analyzing from multiple aspects including language, physical signs, postures, and the surrounding environment to determine user requirements, the accuracy of the obtained user requirements is ensured, providing a basis for subsequent accurate companionship.

[0109] Embodiment 6:

[0110] Based on Embodiment 1, an Internet of Things-based companion robot system is provided in an embodiment of the present invention. The remote monitoring module includes:

[0111] A label configuration unit for dividing user requirements based on a requirement unit of machine learning to obtain multiple minimum sub-requirements, and configuring a first specified label for the minimum sub-requirements, dividing the current user state based on a state unit of machine learning to obtain multiple minimum sub-states, and configuring a second specified label for the minimum sub-states;

[0112] A model training unit for performing model training based on machine learning using the previously obtained first label and second label to obtain a policy matching model;

[0113] A policy determination unit for inputting the first specified label into the policy matching model to obtain a first companionship policy, and inputting the second specified label into the policy matching model to obtain a second companionship policy;

[0114] A key determination unit is used to compare the first escort strategy and the second escort strategy, determine the first escort focus of the first escort strategy and the second escort focus of the second escort strategy, and fuse the first escort focus and the second escort focus to obtain the target escort focus;

[0115] A selection and integration unit is used to select escort strategy points from the first escort strategy and the second escort strategy respectively based on the target escort focus, and integrate the escort strategy points to obtain a robot escort strategy for the user.

[0116] In this embodiment, the minimum sub-requirements cannot be further divided, and the minimum sub-states cannot be further divided.

[0117] The beneficial effects of the above design scheme are as follows: By analyzing from two aspects of requirements and states respectively, obtaining the corresponding escort strategies and then integrating them, remote monitoring of the user and determination of the escort strategy are realized, and the escort strategy can be adjusted flexibly, ensuring the accuracy of the obtained robot escort strategy.

[0118] Embodiment 7:

[0119] Based on Embodiment 6, an embodiment of the present invention provides an Internet of Things-based escort robot system. The model training unit includes:

[0120] A training unit is used to obtain the escort strategy points corresponding to the first label and the second label, input the first label and its corresponding escort strategy points into an initial machine learning model for training to obtain a requirements matching model, and input the second label and its corresponding escort strategy points into the initial machine learning model for training to obtain a state matching model;

[0121] A fusion unit is used to fuse the requirements matching model and the state matching model based on a preset model fusion method to obtain a strategy matching model.

[0122] In this embodiment, the preset model fusion method is selected from existing methods according to the actual situation.

[0123] The beneficial effects of the above design scheme are as follows: By training from two aspects of requirements and states respectively and then fusing the models, the comprehensiveness and accuracy of the obtained strategy matching model are ensured, providing a model basis for strategy determination and flexible adjustment.

[0124] Embodiment 8:

[0125] Based on Embodiment 1, an embodiment of the present invention provides an Internet of Things-based escort robot system. The robot execution module includes:

[0126] An instruction determination unit for parsing the robot escort strategy, determining the escort process of the escort robot, and matching operation instructions related to the escort process from the operation instruction library;

[0127] An execution unit for obtaining the execution resources required for the operation instructions and executing the escort robot in combination with the execution resources.

[0128] The beneficial effect of the above design solution is: by determining the operation instructions of the escort robot according to the robot escort strategy and executing according to the operation instructions, the sense of security and satisfaction of the user are provided.

[0129] Embodiment 9:

[0130] Based on Embodiment 8, an embodiment of the present invention provides an escort robot system based on the Internet of Things, where the execution unit includes:

[0131] A monitoring unit for monitoring the execution process of the robot to obtain monitoring data;

[0132] An early warning unit for performing corresponding early warning reminders based on the relationship between the monitoring data and the monitoring threshold.

[0133] The beneficial effect of the above design solution is: by determining the operation instructions of the escort robot according to the robot escort strategy and executing according to the operation instructions, the sense of security and satisfaction of the user are provided.

[0134] Embodiment 10:

[0135] Based on Embodiment 1, an embodiment of the present invention provides a remote monitoring method for an escort robot system based on the Internet of Things, as Figure 3 shown, including:

[0136] S1: Obtain the physical sign data and posture data of the user in the wearable device, obtain the living environment data of the user in the environmental detection device, and obtain the language data of the user;

[0137] S2: Comprehensively analyze the physical sign data, posture data, living environment data and language data of the user to determine the user's needs;

[0138] S3: Obtain the transmitted user needs and the current state of the user, and combine machine learning to determine the robot escort strategy for the user;

[0139] S4: According to the robot escort strategy, determine the operation instructions of the escort robot and execute according to the operation instructions.

[0140] In this embodiment, combining machine learning, the robot escort strategy for the user is obtained through analysis by a machine learning algorithm.

[0141] The beneficial effects of the above design are as follows: By obtaining the physical sign data and posture data of the user in the wearable device, the living environment data of the user and the language data of the user in the environmental detection device, multi-dimensional user information is obtained, providing a data basis for precise companionship. By comprehensively analyzing the physical sign data, posture data, living environment data and language data of the user, the user's needs are determined to ensure the accuracy of the obtained user needs. By obtaining and transmitting the user's needs and the user's current state, combined with machine learning, the robot companionship strategy for the user is determined, and the companionship strategy is flexibly adjusted to ensure the accuracy of the obtained robot companionship strategy. By determining the operation instructions of the companion robot according to the robot companionship strategy and executing according to the operation instructions, the user's sense of security and satisfaction are provided.

[0142] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of this application document and its equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. An escort robot system based on the Internet of Things, characterized in that, Including: A data interaction module, configured to obtain the physiological sign data and posture data of the user in the wearable device, obtain the living environment data of the user in the environmental detection device, and obtain the language data of the user; A data analysis module, configured to comprehensively analyze the physiological sign data, posture data, living environment data and language data of the user to determine the user's needs, including: A need determination unit, configured to obtain keywords from the language data, match the keywords with a need database, and obtain the number of needs with a matching degree greater than a preset matching degree; Determine whether the number of needs is greater than 1; If so, obtain the corresponding need set; Otherwise, obtain the corresponding initial need; An action analysis unit, configured to establish a physiological sign and posture structure of the user based on the physiological sign data and posture data, and perform dynamic action restrictions on the physiological sign and posture structure based on the physiological sign data and posture data to obtain action restriction information; A need screening unit, configured to obtain the standard action features corresponding to the needs in the need set, judge the standard action features based on the action restriction information, and select a target need set that meets the action restriction information; A reliability determination unit, configured to determine the current user state based on the physiological sign data, posture data and language data, and assign a first reliability to the physiological sign data and posture data and a second reliability to the language data based on the state difference between the current user state and the historical user average state; An evaluation unit, configured to evaluate the target need set based on the first reliability and the second reliability, and select the need with the highest evaluation value as the initial need; An environment analysis unit, configured to obtain the standard environment data required for the initial need, compare the standard environment data with the living environment data to obtain an environment difference, and judge whether the environment difference affects the execution of the initial need; If so, perform fine-tuning on the initial need based on the environment difference to obtain the user's needs; Otherwise, use the initial need as the user's needs; A remote monitoring module, configured to obtain the transmitted user needs and the current user state, and determine a robot escort strategy for the user in combination with machine learning; A robot execution module, configured to determine an operation instruction for the escort robot according to the robot escort strategy and execute according to the operation instruction.

2. The companion robot system based on the Internet of Things according to claim 1, wherein, It further includes: A transmission module, configured to transmit the user needs and the current user state to the remote monitoring module based on the Internet of Things; The transmission module includes: A transmission packet acquisition unit, configured to pack the user needs and the current user state according to a preset data packing method to obtain a data transmission packet; A transmission unit, configured to determine a transmission method based on the Internet of Things and transmit the data transmission packet to the remote monitoring module through the Internet of Things according to the transmission method.

3. The accompanying robot system based on the Internet of Things according to claim 1, characterized in that, The data interaction module includes: A wearable device interaction unit, configured to obtain the collected sensing data of the user in the wearable device, perform abnormal data removal and data type division processing on the collected sensing data to obtain the physiological sign data and posture data of the user; An environmental device interaction unit, configured to obtain environmental sensing data of a user from an environmental detection device, remove abnormal data from the environmental sensing data, and classify and process the data according to data types, so as to obtain the living environmental data of the user; A voice interaction unit, configured to obtain voice sensing data of a user from a voice collection device, and parse the voice sensing data to obtain language data of the user.

4. The accompanying robot system based on the Internet of Things according to claim 1, characterized in that, The requirement screening unit includes: A judgment unit, configured to judge whether a standard action feature contains an action feature in action restriction information; If so, determine that the requirement corresponding to the standard action feature does not meet the action restriction information; Otherwise, determine that the requirement corresponding to the standard action feature meets the action restriction information; An integration unit, configured to integrate the requirements that meet the action restriction information to obtain a target requirement set.

5. The accompanying robot system based on the Internet of Things according to claim 1, wherein, The remote monitoring module includes: A label configuration unit, configured to divide user requirements based on a requirement unit of machine learning to obtain a plurality of minimum sub-requirements, and configure a first specified label for the minimum sub-requirements, divide the current state of the user based on a state unit of machine learning to obtain a plurality of minimum sub-states, and configure a second specified label for the minimum sub-states; A model training unit, configured to perform model training based on machine learning based on the previously obtained first label and second label to obtain a policy matching model; A policy determination unit, configured to input the first specified label into the policy matching model to obtain a first escort policy, and input the second specified label into the policy matching model to obtain a second escort policy; A key point determination unit, configured to compare the first escort policy and the second escort policy, determine a first escort key point of the first escort policy and a second escort key point of the second escort policy, and fuse the first escort key point and the second escort key point to obtain a target escort key point; A selection and integration unit, configured to select escort policy points from the first escort policy and the second escort policy respectively based on the target escort key point, and integrate the escort policy points to obtain a robot escort policy for the user.

6. The accompanying robot system based on the Internet of Things according to claim 5, characterized in that, The model training unit includes: A training unit, configured to obtain escort policy points corresponding to the first label and the second label, input the first label and its corresponding escort policy points into an initial machine learning model for training to obtain a requirement matching model, and input the second label and its corresponding escort policy points into the initial machine learning model for training to obtain a state matching model; A fusion unit, configured to fuse the requirement matching model and the state matching model based on a preset model fusion method to obtain a policy matching model.

7. The companion robot system based on the Internet of Things according to claim 1, characterized in that, The robot execution module includes: An instruction determination unit, configured to parse the robot escort policy, determine the escort process of the escort robot, and match operation instructions related to the escort process from an operation instruction library; An execution unit, configured to obtain execution resources required for the operation instructions, and execute the escort robot in combination with the execution resources.

8. The companion robot system based on the Internet of Things according to claim 7, characterized in that, The execution unit includes: A monitoring unit, configured to monitor the execution process of the robot to obtain monitoring data; An early warning unit, configured to perform corresponding early warning reminders based on the relationship between the monitoring data and a monitoring threshold.

9. The remote monitoring method of an escort robot system based on the Internet of Things according to claim 1, characterized in that, Includes: S1: Obtain the physical sign data and posture data of the user in the wearable device, obtain the living environment data of the user in the environmental detection device, and obtain the language data of the user; S2: Conduct a comprehensive analysis of the user's physical sign data, posture data, living environment data and language data to determine the user's needs; S3: Obtain the transmitted user needs and the user's current state, and combine machine learning to determine the robot escort strategy for the user; S4: According to the robot escort strategy, determine the operation instructions of the escort robot and execute them according to the operation instructions.

Citation Information

Patent Citations

  • Elder accompanying robot subsystem based on emotion recognition

    CN111079440A

  • Sensor integrated nursing system of Internet of Things

    CN117197998A