A Human-Computer Interaction Method for Intelligent Simulated Baby Robots Based on User Behavior Pattern Recognition

Abstract

This invention relates to the field of data processing, and more particularly to a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition. The method includes: real-time acquisition and preprocessing of user behavior data; extraction of preliminary feature data from the preprocessed behavior data and dimensionality reduction processing to obtain dimensionality-reduced feature data; intelligent fusion processing of the dimensionality-reduced feature data using an adaptive spatiotemporal feedback embedded multidimensional data fusion algorithm to obtain comprehensive behavior data and identify the user's current needs; prediction of user behavior based on the comprehensive behavior data; merging the user's current needs with the predicted user behavior results to obtain a comprehensive demand result; and calculation of the optimal response delay time based on the comprehensive demand result to realize the interaction between the intelligent simulated baby robot and the user. This method solves the technical problem of inaccurate and incomplete processing of user behavior data by traditional robots, leading to poor adaptability and stability in interaction with the simulated baby robot.

Description

Technical Field

[0001] This invention relates to the field of data processing, and in particular to a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition. Background Technology

[0002] In the field of intelligent simulated baby robots, with the rapid development of artificial intelligence and robotics, more and more robots are designed to interact naturally with humans, especially in childcare scenarios. These robots assist parents by simulating the behavior and needs of infants. However, existing intelligent simulated baby robots typically face several technical bottlenecks, mainly focusing on how to accurately identify and understand user needs, how to ensure the robot's real-time response to user behavior in complex environments, and how to efficiently utilize limited computing resources for rapid and accurate behavioral responses.

[0003] Currently, most intelligent robots rely on a single type of sensor for behavior recognition, typically using only visual or voice data to infer user needs. However, this method is limited by the sensor's limitations and is easily affected by environmental interference (such as changes in lighting or background noise), resulting in low recognition accuracy. In addition, traditional robots often use simple preset rules for behavior response, which is inadequate when dealing with complex and multi-dimensional user needs and cannot provide a personalized interactive experience.

[0004] In summary, the aforementioned traditional robots suffer from inaccurate and incomplete processing of user behavior data, resulting in poor adaptability and stability during interaction with the simulated baby robot. Summary of the Invention

[0005] This invention provides a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition, in order to solve the technical problem that traditional robots do not process user behavior data accurately or comprehensively, resulting in poor adaptability and stability in interaction with the simulated baby robot.

[0006] The present invention provides a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition, specifically including the following technical solutions:

[0007] A human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition includes the following steps:

[0008] S1. Real-time collection of user behavior data and preprocessing to obtain preprocessed behavior data; extraction of preliminary feature data from the preprocessed behavior data and dimensionality reduction to obtain dimensionality-reduced feature data; intelligent fusion processing of the dimensionality-reduced feature data through an adaptive spatiotemporal feedback embedded multidimensional data fusion algorithm to obtain comprehensive behavior data; behavior recognition and analysis based on comprehensive behavior data to obtain the user's current needs;

[0009] S2. Based on comprehensive behavioral data, predict user behavior to obtain user behavior prediction results; merge the user's current needs with the user behavior prediction results to obtain comprehensive needs results; based on the comprehensive needs results, introduce an interaction delay optimization algorithm to obtain the optimal response delay time of the needs, and realize the interaction between the intelligent simulation baby robot and the user.

[0010] Preferably, S1 specifically includes:

[0011] The adaptive spatiotemporal feedback embedded multidimensional data fusion algorithm intelligently fuses the dimensionality-reduced feature data through spatiotemporal perception reconstruction, nonlinear mapping and weighted adjustment to obtain comprehensive behavioral data.

[0012] Preferably, S1 specifically includes:

[0013] In the implementation of the adaptive spatiotemporal feedback embedding multidimensional data fusion algorithm, an adaptive weight factor is introduced, and combined with the spatiotemporal context information matrix, the spatiotemporal perception reconstruction of the dimensionality-reduced feature data is performed to obtain the spatiotemporal perception feedback matrix.

[0014] Preferably, S1 specifically includes:

[0015] In the implementation of the adaptive spatiotemporal feedback embedding multidimensional data fusion algorithm, the dimensionality-reduced feature data is nonlinearly mapped to obtain the nonlinear mapping result; based on the nonlinear mapping result, combined with the spatiotemporal perception feedback matrix, and with the introduction of a Gaussian decay factor, the comprehensive behavioral data is calculated.

[0016] Preferably, S2 specifically includes:

[0017] By combining historical user behavior data, a user behavior prediction model is constructed and trained; the comprehensive behavior data is then processed using the user behavior prediction model to obtain the user behavior prediction results.

[0018] Preferably, S2 specifically includes:

[0019] In the implementation of the interaction delay optimization algorithm, the priority of the requirements is defined based on the comprehensive requirement results; and the optimal response delay time of the requirements is calculated based on the priority of the requirements.

[0020] Preferably, S2 specifically includes:

[0021] The optimal response delay time for the demand is calculated based on the demand priority, introducing a priority adjustment coefficient, and combining the demand resource consumption and the remaining available computing resources of the intelligent simulated baby robot.

[0022] Preferably, S2 specifically includes:

[0023] The intelligent simulated baby robot dynamically schedules resources based on the optimal response delay time and priority of the demand, and performs behavioral feedback to enable interaction between the intelligent simulated baby robot and the user.

[0024] The beneficial effects of the technical solution of the present invention are:

[0025] 1. By embedding a multi-dimensional data fusion algorithm with adaptive spatiotemporal feedback, data from visual, auditory, motion, and tactile sensors can be effectively fused. Utilizing spatiotemporal perception reconstruction, nonlinear mapping, and weighted adjustment, it can process and fuse multi-sensor data in real time and accurately, even under changing environmental conditions and sensor states. Compared to traditional single-sensor data processing methods, this multimodal fusion provides more comprehensive and accurate user behavior analysis, enabling robots to better understand and respond to complex user needs.

[0026] 2. The interaction delay optimization algorithm ensures that the intelligent simulated baby robot can obtain the optimal response delay time for each demand based on its priority and remaining available computing resources through real-time calculation and dynamic scheduling. For high-priority demands, the intelligent simulated baby robot will allocate computing resources first to ensure a fast response. For low-priority demands, the intelligent simulated baby robot can appropriately delay its response, thereby effectively managing limited computing resources. This not only improves the efficiency of the intelligent simulated baby robot's response but also ensures the efficient use of resources and avoids resource waste. Attached Figure Description

[0027] Figure 1 This is a flowchart of a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition, as described in this invention. Detailed Implementation

[0028] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0030] The following description, in conjunction with the accompanying drawings, details a specific scheme for a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition provided by the present invention.

[0031] See attached document Figure 1 The diagram illustrates a human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition, according to an embodiment of the present invention. The method includes the following steps:

[0032] S1. Real-time collection of user behavior data and preprocessing to obtain preprocessed behavior data; extraction of preliminary feature data from the preprocessed behavior data and dimensionality reduction to obtain dimensionality-reduced feature data; intelligent fusion processing of the dimensionality-reduced feature data through an adaptive spatiotemporal feedback embedded multidimensional data fusion algorithm to obtain comprehensive behavior data; behavior recognition and analysis based on comprehensive behavior data to obtain the user's current needs;

[0033] The intelligent simulated baby robot collects user behavior data in real time through sensors, including visual information (such as facial expressions, gestures, body posture, etc.), sound information (such as voice, tone, etc.), motion information (such as acceleration, angle changes, amplitude of movement, etc.), and tactile information (such as the user's touch, pressure, contact, etc.). The user behavior data undergoes preprocessing such as noise reduction, data synchronization and time alignment, format conversion, standardization, and normalization to obtain preprocessed behavior data. The sensors include visual sensors (cameras), sound sensors (microphones), motion sensors (accelerometers, gyroscopes), and tactile sensors. The preprocessing process employs techniques well-known to those skilled in the art and will not be elaborated upon here.

[0034] Preliminary feature extraction is performed on the preprocessed behavioral data using existing feature engineering techniques to obtain preliminary feature data. express Time of the first Preliminary feature data from each sensor, including motion speed, acceleration, speech frequency and amplitude, and facial expression feature points, are collected. This preliminary feature data is then processed using existing feature reduction techniques (such as principal component analysis) to obtain the dimensionality-reduced feature data. The dimensionality-reduced feature data is intelligently fused using an adaptive spatiotemporal feedback embedding multidimensional data fusion algorithm to obtain comprehensive behavioral data. This algorithm, through spatiotemporal perception reconstruction, nonlinear mapping, and weighted adjustment, enables real-time and accurate data fusion based on environmental changes, sensor states, and dynamic changes in user behavior data to obtain comprehensive behavioral data. The specific implementation process is as follows:

[0035] In the first stage, based on the adaptive weighting algorithm and spatiotemporal awareness technology, the dimensionality-reduced feature data is reconstructed using spatiotemporal awareness to obtain the spatiotemporal awareness feedback matrix. The specific calculation formula is as follows:

[0036] in, It is the first One sensor in The spatiotemporal perception feedback matrix at time t represents the first time. The sensor's perception results under the current environmental and spatiotemporal conditions; It is the first One sensor in The adaptive weighting factor at each time step is obtained through an existing adaptive control algorithm that adjusts weights based on environmental changes and sensor states. The reference value range is [range to be filled in]. ; It is the first One sensor in The spatiotemporal context information matrix at any given moment is a standardized matrix generated using environmental perception technology based on the sensor's operating state and the external environment (such as noise, illumination, etc.). This is a well-known technique in the field and will not be elaborated upon here. Is the first sensor in The dimensionality-reduced feature data at each time step; It is the Hadamard product, which refers to the element-by-element multiplication of corresponding elements in the matrix. Through the above process, the spatiotemporal context information is fused with the feature data after dimensionality reduction from the sensor, providing dynamic and flexible spatiotemporal feedback for subsequent fusion, so that the subsequent fusion process can more accurately reflect changes in the environment.

[0037] In the second stage, firstly, the dimensionality-reduced feature data of all sensors are nonlinearly mapped using a nonlinear mapping function to obtain the nonlinear mapping results for each sensor. The nonlinear mapping function is used to handle the hidden nonlinear features in the dimensionality-reduced feature data of each sensor. For example, image data from a vision sensor and acceleration data from a motion sensor may have complex nonlinear relationships; these hidden nonlinear features can be extracted using the nonlinear mapping function. Next, the spatiotemporal perception feedback matrix and the sensor nonlinear mapping results are weighted and fused to obtain the comprehensive data for each time step. Finally, a Gaussian decay factor is introduced. By combining the deviation between the dimensionality-reduced feature data of each sensor and the average dimensionality-reduced feature data of all sensors, the fusion effect is further optimized to obtain comprehensive behavioral data; the specific formula is: in, yes The fused behavioral data at each moment represents the final comprehensive behavioral data obtained after multi-sensor data processing, spatiotemporal perception feedback, nonlinear mapping and weighted adjustment, and Gaussian decay processing. It refers to the number of sensors; yes Time of the first The result of nonlinear mapping of the dimension-reduced feature data of each sensor. It is a nonlinear mapping function used to enhance the expressive power of the feature data after dimensionality reduction of the sensor, so as to capture the complex patterns in the feature data after dimensionality reduction. The specific application scenario determines the function, such as ReLU (Rectified Linear Unit), sigmoid or tanh. yes Time of the first The nonlinear adjustment factor of each sensor is used to control the influence of the spatiotemporal sensing feedback matrix of the sensor on the nonlinear weighting process. It is determined according to the expert experience method and can be set to 1. yes Time of the first The Gaussian attenuation factor for each sensor is used to attenuate the deviation between the dimensionality-reduced feature data of the sensor and the overall average value. The value is determined based on the specific scenario, with a reference range of [value missing]. ; yes The average value of the dimension-reduced feature data of all sensors at any given time is used as a reference standard; It is a weighted adjustment of the spatiotemporal perception feedback matrix. Through this weighted adjustment, the intelligent simulated baby robot can dynamically adjust the contribution of each sensor according to the data quality of the sensor and the influence of the environment. It is a Gaussian attenuation term, which can effectively suppress noise and outliers in sensor data, making the integrated behavioral data obtained by fusion more stable and reliable, and able to accurately reflect the user's real needs;

[0038] After obtaining comprehensive behavioral data, the intelligent simulated baby robot uses existing deep learning-based behavior recognition algorithms (such as convolutional neural networks, long short-term memory networks, etc.) to perform behavior recognition analysis, identify the user's current needs, such as feeding, soothing or changing diapers, picking up the baby or rocking the baby.

[0039] S2. Based on comprehensive behavioral data, predict user behavior to obtain user behavior prediction results; merge the user's current needs with the user behavior prediction results to obtain comprehensive needs results; based on the comprehensive needs results, introduce an interaction delay optimization algorithm to obtain the optimal response delay time of the needs, and realize the interaction between the intelligent simulation baby robot and the user.

[0040] User historical behavior data is retrieved from the existing database, and a user behavior prediction model is constructed and trained by selecting existing machine learning algorithms (such as regression analysis, random forest, long short-term memory network, etc.) according to the specific application scenario. The comprehensive behavior data is then processed by the user behavior prediction model to obtain user behavior prediction results, including feeding needs prediction, soothing needs prediction, diaper changing needs prediction, sleep / rest needs prediction, and interaction / entertainment needs prediction.

[0041] Furthermore, the user's current needs and the predicted user behavior are combined to obtain a comprehensive demand result. Based on this comprehensive demand result, an interaction delay optimization algorithm is used to realize the interaction between the intelligent simulated baby robot and the user. This interaction delay optimization algorithm reduces the response delay between the intelligent simulated baby robot and the user through real-time optimization, while ensuring that the intelligent simulated baby robot's behavioral responses are personalized and accurate. The specific implementation process is as follows:

[0042] Based on the comprehensive requirements, and utilizing existing multimodal behavior recognition technologies and sentiment analysis models, the priority of each requirement is defined. The higher the priority value of a demand, the higher its priority. The calculation method for the demand priority is a well-known technique and will not be elaborated here. Furthermore, based on the demand priority, the optimal response latency time for each demand is calculated using a resource optimization scheduling algorithm based on deep learning. The specific calculation formula is as follows:

[0043] in, It is the first The optimal response latency for each requirement; This is a priority adjustment coefficient used to control the impact of demand priority on response latency. It is determined based on expert experience combined with the response capability and processing speed of the intelligent simulated baby robot, with a reference value range of [value missing]. ; This is a delay sensitivity coefficient used to control the impact of resource allocation on the response delay time of demand. It is determined through expert experience combined with the hardware configuration of the intelligent simulated baby robot, and the reference value range is [insert range here]. ; It is the first The amount of resources consumed by the demand can be monitored through resource management tools; This represents the remaining available computing resources of the intelligent simulated baby robot, reflecting the robot's ability to process the first... When a demand is made, the amount of remaining available computing resources can be obtained by monitoring through resource management tools; This is a factor affecting response latency due to resource load, determining the amplification effect of resource load on response latency. The value is determined based on specific requirements, with a reference range of values. ; It is the nonlinear effect of resource load on response latency. By amplifying or reducing the ratio of demanded resource consumption to remaining available computing resources, the increase in response latency is more significant when computing resources are scarce, thereby providing reasonable priority scheduling. It is a weighted adjustment of latency sensitivity to resource allocation. By adjusting the weight, it affects the relationship between resource allocation and response latency, ensuring that the robot can balance the relationship between resource allocation and response time when handling demands. It is the effect of the priority adjustment coefficient on response latency, which ensures that high-priority requests can receive a fast response, while low-priority requests can have their response appropriately delayed, thereby optimizing the system's resource allocation and response time.

[0044] Furthermore, based on the optimal response delay time and priority of the demand, dynamic scheduling is performed using existing scheduling algorithms (such as preemptive priority scheduling) to achieve resource allocation: for high-priority demands, the intelligent simulated baby robot will be allocated computing resources first to ensure a fast response; for low-priority demands, the response can be appropriately delayed to ensure the efficient use of the intelligent simulated baby robot's resources.

[0045] Furthermore, once the intelligent simulated baby robot obtains the optimal response delay time based on demand priority and resource allocation, it begins to execute corresponding behavioral feedback: for example, if the intelligent simulated baby robot predicts that the user will perform a soothing behavior, it will simulate a baby's crying or make corresponding movements; if the user shows signs of feeding behavior, the intelligent simulated baby robot will mimic the baby's hunger needs and send relevant signals, thus realizing the interaction between the intelligent simulated baby robot and the user.

[0046] In summary, a human-computer interaction method for intelligent simulated baby robots based on user behavior pattern recognition has been completed.

[0047] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0048] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0049] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition, characterized in that, Includes the following steps: S1. Collect user behavior data in real time and preprocess it to obtain preprocessed behavior data; Preliminary feature data is extracted from the preprocessed behavioral data and then dimensionality reduction is performed to obtain the dimensionality-reduced feature data. Introduction An adaptive spatiotemporal feedback embedded multidimensional data fusion algorithm intelligently fuses the dimensionality-reduced feature data through spatiotemporal perception reconstruction, nonlinear mapping and weighted adjustment to obtain comprehensive behavioral data. The specific process of the adaptive spatiotemporal feedback embedding multidimensional data fusion algorithm is as follows: An adaptive weighting factor is introduced, and combined with the spatiotemporal context information matrix, spatiotemporal perception reconstruction is performed on the dimensionality-reduced feature data to obtain a spatiotemporal perception feedback matrix. Simultaneously, a nonlinear mapping is performed on the dimensionality-reduced feature data to obtain the nonlinear mapping result. The spatiotemporal perception feedback matrix and the nonlinear mapping result are then weighted and fused to obtain comprehensive data for each time step. By introducing a Gaussian decay factor and combining the deviation between the dimensionality-reduced feature data and its average value, comprehensive behavioral data is calculated. Based on the comprehensive behavioral data, behavioral recognition analysis is performed to obtain the user's current needs. S2. Based on comprehensive behavioral data, predict user behavior to obtain user behavior prediction results; The user's current needs are combined with the user behavior prediction results to obtain a comprehensive needs result; based on the comprehensive needs result, an interaction delay optimization algorithm is introduced to define the priority of the needs; Based on the priority of the demand, the optimal response delay time is calculated to realize the interaction between the intelligent simulated baby robot and the user.

2. The human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition according to claim 1, characterized in that, S2 specifically includes: By combining historical user behavior data, a user behavior prediction model is constructed and trained; the comprehensive behavior data is then processed using the user behavior prediction model to obtain the user behavior prediction results.

3. The human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition according to claim 1, characterized in that, S2 specifically includes: The optimal response delay time for the demand is calculated based on the demand priority, introducing a priority adjustment coefficient, and combining the demand resource consumption and the remaining available computing resources of the intelligent simulated baby robot.

4. The human-computer interaction method for an intelligent simulated baby robot based on user behavior pattern recognition according to claim 3, characterized in that, S2 specifically includes: The intelligent simulated baby robot dynamically schedules resources based on the optimal response delay time and priority of the demand, and performs behavioral feedback to enable interaction between the intelligent simulated baby robot and the user.

Images