Robot touch interaction control method

Abstract

The invention relates to the technical field of teaching equipment interaction, in particular to a robot touch interaction control method. The method comprises the following steps: acquiring a touch point and a touch position area on the preschool education equipment; time periods formed by continuous sampling moments with touch points and time periods formed by continuous sampling moments without touch points are obtained respectively and serve as a touch time period and a blank time period; touch point changes in two touch time periods adjacent to the blank time period are analyzed, and the touch time periods are updated; constructing a plurality of connected graphs based on the updated touch time period; in combination with the structure distribution characteristics of the connected graph, performing weighted adjustment on the node similarity, and determining a cluster based on a connected graph split clustering algorithm; and sampling moments in the clusters are extracted, interpolation supplement of the touch position area is carried out, and an interaction track is obtained. According to the method, the accuracy of man-machine interaction of the preschool education equipment is improved by extracting the change features of the touch action positions and analyzing the structure distribution relevance between the nodes.

Description

Technical Field

[0001] The present invention relates to the technical field of teaching equipment interaction, and particularly relates to a robot touch interaction control method. Background Art

[0002] Robot touch interaction refers to an interaction method in which a finger or palm is used to input and operate on a touch screen. For example, the human-computer interaction of an early childhood education device based on infrared touch belongs to a kind of robot touch interaction. The early childhood education device based on infrared touch can provide rich and diverse learning contents, such as characters, graphics, sounds, videos, etc. By touching the screen, children can learn basic knowledge such as numbers, letters, shapes, colors, etc., and can also carry out various learning interaction activities such as enlightenment education, interesting games, painting, etc. The user interaction experience of its human-computer interaction is closely related to the action recognition accuracy of the infrared touch panel. When there is water stain or oil stain on the surface of the touch panel or the user's finger, etc., problems such as disconnection are likely to occur, affecting the real-time performance and accuracy of the human-computer interaction. Therefore, it is necessary to improve the action recognition accuracy for the change trajectory of the user's touch action on the touch panel to enhance the human-computer interaction experience.

[0003] Currently, a common method for recognizing the human-computer interaction of early childhood education devices is to use the connected graph splitting clustering algorithm to cluster the touch action position data, which can effectively analyze the correlation characteristics of the trajectory changes of the user's touch action on the touch panel in terms of time sequence, so as to achieve accurate recognition of the touch action and thereby enhance the human-computer interaction experience of the early childhood education device; however, during the process of using the connected graph splitting clustering algorithm to process the touch action position data, multiple touch operations are similar to the performance during disconnection, and when multiple touch operations and disconnection occur, it will be considered as no touch operation, resulting in inaccurate analysis results. Summary of the Invention

[0004] In order to solve the technical problem that during the process of using the connected graph splitting clustering algorithm to process the touch action position data, the situation of multiple touch operations and disconnection will lead to inaccurate analysis results of the human-computer interaction, the purpose of the present invention is to provide a robot touch interaction control method, and the specific technical solution adopted is as follows:

[0005] Obtain the touch points generated during touch interaction on the early childhood education device of the infrared touch panel, and form a touch position area from the touch points;

[0006] Obtain the time periods composed of consecutive sampling moments with touch points and the time periods composed of consecutive sampling moments without touch points on the early childhood education equipment respectively, as the touch time periods and blank time periods; analyze the action change correlation degree of the touch point changes in two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period; screen the blank time period according to the action change correlation degree of the blank time period to obtain the associated time period; update the touch time period based on the associated time period;

[0007] Based on the updated touch time period, construct multiple connected graphs; determine the node discrimination confidence of the nodes in the connected graph in combination with the structural distribution characteristics of the connected graph; use the node discrimination confidence as the weight to weight the node similarity, and determine the clustering clusters based on the connected graph splitting clustering algorithm;

[0008] Extract the sampling moments belonging to the associated time period within the clustering cluster as the interaction moments; perform interpolation and supplementation on the touch position areas of the interaction moments to obtain the interaction trajectory formed by the final touch position areas.

[0009] Preferably, the analyzing the action change correlation degree of the touch point changes in two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period includes:

[0010] Obtain the touch position change degree of each sampling moment in two touch time periods adjacent to the blank time period;

[0011] Determine the interval change correlation degree of the touch positions at different sampling moments according to the touch position change degree at different sampling moments within the touch time period and the length of the touch time period;

[0012] Combine the interval change correlation degrees of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period to determine the action change correlation degree of the blank time period.

[0013] Preferably, the obtaining the touch position change degree of each sampling moment in two touch time periods adjacent to the blank time period includes:

[0014] Take any sampling moment within the touch time period as the target moment, use the difference in the number of touch points between the target moment and the previous sampling moment as the numerator, and use the time difference between the target moment and the previous sampling moment as the denominator. The fraction composed of the numerator and the denominator is used as the touch position change degree of the target moment.

[0015] Preferably, the calculation formula of the interval change correlation degree is:

[0016]

[0017] Among them, is the interval change correlation degree of the touch position at the i-th sampling moment within the touch time period; is the normalization function; is the time length of the touch time period in which the i-th sampling moment is located; is the touch position change degree at the i-th sampling moment within the touch time period; is the exponential function with the natural constant e as the base; is the number of sampling moments in the touch time period in which the i-th sampling moment is located; is the average value of the touch position change degrees of all sampling moments within the touch time period.

[0018] Preferably, determining the action change correlation degree of the blank time period by combining the interval change correlation degrees of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period includes:

[0019] Calculating the difference between the interval change correlation degrees of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period as the interval change degree; taking the negative correlation normalization value of the interval change degree as the action change correlation degree of the blank time interval.

[0020] Preferably, constructing multiple connected graphs based on the updated touch time period includes:

[0021] Taking the centroid point of the touch position area corresponding to a single sampling moment within the updated touch time period as a node of the connected graph, taking the number of nodes in the action time period within the touch time period as the preset incremental clustering increment value, using the preset incremental clustering increment value as the number of nodes within the touch time period, and constructing multiple connected graphs according to the connected graph splitting clustering algorithm.

[0022] Preferably, the calculation formula for the node discrimination confidence degree is:

[0023] ;

[0024] ;

[0025] Among them, is the node discrimination degree of the i-th node in the connected graph constructed by the r-th incremental clustering; is the node discrimination confidence degree of the i-th node; is the interval change correlation degree of the touch position at the i-th sampling moment within the touch time period; is the number of node connection edges of the first part divided by the i-th node in the connected graph constructed by the incremental clustering; The number of nodes in the first part of the connected graph constructed by incremental clustering for the i-th node; The number of nodes in the second part of the connected graph constructed by incremental clustering for the i-th node; The number of node connection edges in the second part of the connected graph constructed by incremental clustering for the i-th node; R is the number of times of incremental clustering; norm is the normalization function; exp is the exponential function with the natural constant e as the base; Is the absolute value symbol.

[0026] Preferably, using the node discrimination confidence as the weight to weight the node similarity includes:

[0027] The calculation formula of the weighted node similarity is:

[0028]

[0029] Wherein, Is the weighted node similarity between the a-th node and the b-th node; Is the node discrimination confidence of the k-th node in the a-th node and the b-th node; Is the initial node similarity between the a-th node and the b-th node.

[0030] Preferably, based on the associated time period, updating the touch time period includes:

[0031] Merging and updating the associated time period and the adjacent touch time period to obtain the updated touch time period.

[0032] Preferably, screening the blank time period according to the action change correlation degree of the blank time period to obtain the associated time period includes:

[0033] Taking the blank time period with the interval change correlation greater than the preset correlation degree threshold as the associated time period.

[0034] The embodiments of the present invention have at least the following beneficial effects:

[0035] Obtain the touch points and touch position areas generated during touch interaction on early childhood education equipment; respectively obtain the time periods composed of consecutive sampling moments with touch points and without touch points, as the touch time periods and blank time periods. Dividing the sampling moments into touch time periods and blank time periods is to subsequently analyze from the blank time periods the abnormal situations where there are no touch points on the early childhood education equipment that may be caused by multiple touch operations or touch interruptions; analyze the changes in touch points within two adjacent touch time periods to the blank time period, update the touch time periods, and achieve screening out from the blank time periods the sampling moments of abnormal situations where there are no touch points on the early childhood education equipment that may be caused by multiple touch operations or touch interruptions by analyzing the area change characteristics of the touch point set in time sequence, as the associated time periods. Based on the updated touch time periods, construct multiple connected graphs; then, in combination with the structural distribution characteristics of the connected graphs, perform weighted adjustment on the node similarity, and determine the clustering clusters based on the connected graph splitting clustering algorithm, so that the accurate processing result of the final connected graph splitting clustering algorithm realizes the accurate judgment of the touch accuracy affected by environmental factors such as touch interruptions; extract the sampling moments within the clustering clusters, perform interpolation supplementation on the touch position areas, obtain the interaction trajectory, and the interaction trajectory constructed based on the linearly interpolated touch position areas is more accurate, and perform real-time human-computer interaction feedback according to the interaction trajectory of the finally obtained touch position areas, effectively avoiding the influence of environmental factors such as touch interruptions on the human-computer interaction experience. Description of the Drawings

[0036] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0037] Figure 1 It is a flowchart of a method for a robot touch interaction control method provided by an embodiment of the present invention;

[0038] Figure 2 It is a flowchart of a method for determining the action change correlation degree of a blank time period provided by an embodiment of the present invention. Detailed Embodiments

[0039] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following specifically describes, in conjunction with the accompanying drawings and preferred embodiments, a robot touch interaction control method proposed according to the present invention, including its specific implementation manner, structure, features, and effects. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0040] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0041] The embodiment of the present invention provides a specific implementation method of a robot touch interaction control method, which is applicable to the human-computer interaction scenario of an early childhood education device with infrared touch. In order to solve the technical problem that in the process of processing touch action position data using the connected graph splitting clustering algorithm, there are multiple touch operations and touch breaks, which may lead to inaccurate analysis results of human-computer interaction. The present invention extracts the change characteristics of touch action positions, conducts precise analysis of the structural distribution correlation between nodes, improves the accuracy of action recognition, so that the precise processing result of the final connected graph splitting clustering algorithm realizes the precise judgment of touch accuracy affected by environmental factors such as touch breaks; extracts the sampling moments within the clustering cluster, interpolates and supplements the touch position area, and obtains an interaction trajectory. The interaction trajectory constructed based on the touch position area after linear interpolation is more accurate, so as to enhance the human-computer interaction experience.

[0042] The following specifically describes the specific solution of a robot touch interaction control method provided by the present invention in conjunction with the accompanying drawings.

[0043] Please refer to Figure 1 , which shows a step flow chart of a robot touch interaction control method provided by an embodiment of the present invention. The method includes the following steps:

[0044] Step S100, obtain the touch points generated during touch interaction on the early childhood education device with an infrared touch panel, and form a touch position area from the touch points.

[0045] In the embodiment of the present invention, through the infrared sensor array under the early childhood education device with an infrared touch panel, touch points are collected during user touch interaction, and the position coordinates of the touch points are obtained. Among them, the sampling frequency is 150 Hz, and each sampling obtains a touch position area, that is, the position coordinates of multiple touch points are obtained. Thus, a set of touch position coordinates in multiple time series is obtained. That is, a touch position area is formed from the touch points. Specifically: extract the position coordinates of the touch points to obtain the corresponding touch position area.

[0046] When there is water stain or oil stain on the surface of the touchpad or the user's finger, etc., the sensor of the infrared touchpad is blocked, resulting in the loss or instability of the touch signal, and problems such as touch interruption occur, affecting the user's human-computer interaction experience. Moreover, the touch interruption signal will exhibit characteristics similar to those of multiple touch interactions, and there are differences in the area change and position change characteristics of the touch area between the touch interruption and the process of multiple inputs.

[0047] Step S200: Respectively obtain the time periods composed of consecutive sampling moments with touch points and the time periods composed of consecutive sampling moments without touch points on the early childhood education device as the touch time periods and blank time periods; perform an action change correlation analysis on the touch point changes within two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period; screen the blank time period according to the action change correlation degree of the blank time period to obtain the associated time period; update the touch time period based on the associated time period.

[0048] Therefore, in order to solve the situation of abnormal human-computer interaction when there is touch interruption or multiple inputs, in the embodiments of the present invention, the continuous sampling time period is first divided into a touch time period and a blank time period. The touch time period is the time period when there are touch points on the early childhood education device, and the blank time period is the time period when there are no touch points on the early childhood education device. However, in the blank time period, in addition to including the moments when it is truly not touched, there are also moments when there is truly a touch but the touch point is not detected by the early childhood education device due to touch interruption or multiple touch interactions. Therefore, after dividing the touch time period and the blank time period, further, combined with the action change correlation analysis of the touch time period adjacent to the blank time period, the time period associated with the touch time period is screened out from the blank time period as the associated time period.

[0049] Since there are area change and position change characteristics in time sequence for the acquired touch position area, in the embodiments of the present invention, by analyzing the area change characteristics of the touch position area in time sequence, the change correlation degree is obtained, and by analyzing the structural change consistency degree of the constructed connectivity graph, the similarity measurement between the nodes of the connectivity graph is adjusted to obtain an accurate clustering result.

[0050] First, according to the collected touch position area data, analyze the time sequence change characteristics to obtain the action change correlation degree. Specifically: Respectively obtain the time periods composed of consecutive sampling moments with touch points and the time periods composed of consecutive sampling moments without touch points on the early childhood education device as the touch time periods and blank time periods; perform an action change correlation analysis on the touch point changes within two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period.

[0051] When collecting touch position area data, corresponding to multiple touch position area data during sampling according to the sampling frequency, and due to the change of the touch action trajectory, the area of the touch position area will change, that is, the number of touch points within the area will change. When there is a large bend in the touch action trajectory, there will be a consistency feature in the area change on the trajectories before and after the bend, that is, they both increase or decrease together. When there is a break in contact, there will be a sudden decrease or disappearance of the area, without the change consistency. Therefore, the action change correlation degree can be obtained according to the temporal change characteristics of the touch point set area.

[0052] According to the obtained touch position area data, record the number of touch points in the touch position area corresponding to the sampling moment; when there is no touch action corresponding to the number of touch points at the sampling moment, it is recorded as a blank moment, and obtain multiple time segments with the number of touch points, which are recorded as touch moments. And respectively obtain the time period composed of consecutive sampling moments with touch points on the early childhood education device and the time period composed of consecutive sampling moments without touch points, as the touch time period and the blank time period. It should be noted that the moments within the touch time period are touch moments, and the moments within the blank time period are blank moments.

[0053] Please refer to Figure 2 , perform action change correlation degree analysis on the touch point changes in two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period. Specifically:

[0054] Step S201, obtain the touch position change degree of each sampling moment in two touch time periods adjacent to the blank time period.

[0055] Take any sampling moment within the touch time period as the target moment, use the difference in the number of touch points between the target moment and the previous sampling moment as the numerator, and use the time difference between the target moment and the previous sampling moment as the denominator. The fraction composed of the numerator and the denominator is the touch position change degree of the target moment.

[0056] Taking the i-th sampling moment as the target moment, the calculation formula for the touch position change degree of the corresponding target moment is:

[0057] ; where is the touch position change degree of the i-th sampling moment; is the number of touch points at the i-th sampling moment; is the number of touch points at the (i - 1)-th sampling moment; is the i-th sampling moment; is the (i - 1)-th sampling moment; is the absolute value symbol.

[0058] In the embodiment of the present invention, the degree of change in the number of touch points at adjacent moments is used to reflect the degree of change in the touch position at different sampling moments.

[0059] Step S202: Determine the interval change correlation degree of the touch positions at different sampling moments according to the degree of change in the touch position at different sampling moments within the touch time period and the length of the touch time period.

[0060] The calculation formula for the interval change correlation degree is:

[0061]

[0062] Wherein, is the interval change correlation degree of the touch position at the i-th sampling moment within the touch time period; is the normalization function; is the time length of the touch time period in which the i-th sampling moment is located; is the degree of change in the touch position at the i-th sampling moment within the touch time period; is the exponential function with the natural constant e as the base; is the number of sampling moments in the touch time period in which the i-th sampling moment is located; is the average value of the degrees of change in the touch positions at all sampling moments within the touch time period.

[0063] The interval change correlation degree is calculated by combining the time length of the touch time period in which the sampling moment is located and the degree of change in the touch position. Among them, reflects the magnitude of the degree of change in the touch position at the touch time period in which the i-th sampling moment is located relative to the overall degree of change in the touch position; the interval change correlation degree of the sampling moment is constructed by the average value of the degrees of change in the touch positions at all sampling moments within the touch time period in which the sampling moment is located, which improves the fault tolerance when calculating the interval change correlation degree, and avoids the situation of large errors when constructing the interval change correlation degree corresponding to the sampling moment through a single degree of change in the touch position when an abnormal situation occurs at a certain sampling moment.

[0064] The interval change correlation degree of the i-th sampling moment The larger it is, the higher the accuracy of the user's touch position at this sampling moment, and the more stable the change of the multiple touch position areas in the corresponding touch time period.

[0065] Step S203: Combine the interval change correlation degrees of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period to determine the action change correlation degree of the blank time period.

[0066] Calculate the difference in the interval change correlation degree of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period, and use it as the interval change degree; use the negative correlation normalization value of the interval change degree as the action change correlation degree of the blank time interval. In the embodiments of the present invention, an exponential function with the natural constant as the base and the negative value of the interval change degree as the exponent is used to implement the negative correlation normalization operation of the interval change degree.

[0067] The calculation formula for the action change correlation degree is:

[0068] ; where is the action change correlation degree of the j-th blank time period; is an exponential function with the natural constant as the base; is the interval change correlation degree of the touch position corresponding to the sampling moment belonging to the touch time period at the left endpoint of the j-th blank time period; is the interval change correlation degree of the touch position corresponding to the sampling moment belonging to the touch time period at the right endpoint of the j-th blank time period; is the absolute value symbol; is the interval change degree.

[0069] According to the action change correlation degree of the blank time period, screen the blank time period to obtain the associated time period. Specifically: use the blank time period with the corresponding interval change correlation greater than the preset correlation degree threshold as the associated time period. In the embodiments of the present invention, the value of the preset correlation degree threshold is 0.68, and in other embodiments, the implementer can adjust this value according to the actual situation.

[0070] Obtain the action change correlation degrees of all blank time periods, and use the preset correlation degree threshold to screen the blank time periods. When the action change correlation degree of the blank time period is greater than the preset correlation degree threshold , it can be considered that the associated degree of the blank time period where it is located with the corresponding touch time periods on both sides is relatively high. The blank time period here should not have a differentiating degree. Thus, obtain all the remaining blank time periods to achieve the screening of the blank time periods.

[0071] After obtaining the associated time period, update the touch time period based on the associated time period. Specifically: merge and update the associated time period and the adjacent touch time periods to obtain the updated touch time period.

[0072] By analyzing the temporal change characteristics of the collected touch position area data, the correlation degree of action changes is obtained, and the blank time periods where the touch signal is lost or unstable, such as when there is water stain or oil stain on the touchpad surface or the user's finger, resulting in interruption of touch, are screened out and updated and fused with the touch time periods.

[0073] Step S300: Based on the updated touch time periods, construct multiple connected graphs; combine the structural distribution characteristics of the connected graphs to determine the node discrimination confidence of the nodes in the connected graphs; use the node discrimination confidence as the weight to weight the node similarity, and based on the connected graph splitting clustering algorithm, determine the clustering clusters.

[0074] After obtaining the correlation degree of action changes, based on the connected graph splitting clustering algorithm and combining the structural distribution characteristics of the connected graph, an accurate clustering result is obtained. Among them, the correlation degree of action changes obtained in step S200 is the analysis of the touch action correlation degree from the relationship of the change characteristics of the number of touch points in the touch position area caused by the touch action. However, in the actual use of early childhood education equipment, due to the diverse changes in the touch actions of the user group of young children, the obtained correlation degree of action changes often has difficulty coping with various complex changes. The main reason is that the touch actions of young children are less coherent, and the number fluctuation of the touch points in the touch position area on the change trajectory structure of the touch action is relatively complex. Therefore, the centroid points of the touch position area within the updated touch time periods can be used as the nodes of the connected graph to construct the connected graph, and further combined with the structural distribution characteristics of the connected graph to obtain an accurate clustering analysis result.

[0075] Take the centroid point of the touch position area corresponding to a single sampling moment within the updated touch time period as a node of the connected graph, take the number of nodes in the action time period within the touch time period as the preset incremental clustering increment value, use the preset incremental clustering increment value as the number of nodes within the touch time period, and construct multiple connected graphs according to the CABDDCG algorithm combined with the density and hierarchical division method. The CABDDCG algorithm combined with the density and hierarchical division method is also the connected graph splitting clustering algorithm. Specifically, according to the obtained correlation degree of changes in the touch time period and the blank time period, re-segment the time to obtain multiple updated touch time periods, and take the centroid point of the touch position area corresponding to a single sampling moment within the updated touch time period as a node of the connected graph, thereby obtaining multiple connected graph nodes; perform incremental clustering on the multiple connected graph nodes. The specific method is: construct a two-dimensional sample space that is the same size or scaled proportionally to the infrared touchpad in the early childhood education equipment, and the coordinates of its nodes correspond to the coordinates of the centroid points of the touch position area, and obtain the number of nodes in the touch time period , the number of nodes in the action time period within the touch time period is used as the preset incremental clustering increment value. With the preset incremental clustering increment value being the number of nodes within the touch time period, multiple connected graphs are constructed. For example, if there are 3 updated touch time periods, then incremental clustering is performed three times to obtain 3 connected graphs, corresponding to time periods 1, 12, and 123 respectively. It should be noted that according to the literature by Deng Jianshuang, Zheng Qilun, Peng Hong, and Deng Weiwei, "Clustering Algorithm Based on Dynamic Splitting of Connected Graphs", Journal of South China University of Technology (Natural Science Edition), January 2007, Vol. 35, No. 1, 1000 - 565X(2007)01 - 0118 - 05, the method of establishing a connected graph according to the CABDDCG algorithm combined with the density and hierarchical partitioning method is used to implement the construction of multiple connected graphs with the preset incremental clustering increment value as the number of nodes within the touch time period in the embodiments of the present invention. That is, the embodiments of the present invention are improvements to the literature "Clustering Algorithm Based on Dynamic Splitting of Connected Graphs".

[0076] For each connected graph, there are structural changes. Its nodes can divide the connected graph into two parts. The higher the structural similarity of the node set of the two - part connected graph, the higher the credibility of the splitting threshold corresponding to this node.

[0077] Combined with the structural distribution characteristics of the connected graph, the node discrimination confidence of the nodes in the connected graph is determined.

[0078] The calculation formula for the node discrimination confidence is as follows:

[0079] ;

[0080] ;

[0081] Among them, is the node discrimination degree of the i - th node in the connected graph constructed by the r - th incremental clustering; is the node discrimination confidence of the i - th node; is the interval change correlation degree of the touch position at the i - th sampling moment within the touch time period; is the number of node connection edges of the first part divided by the i - th node of the connected graph constructed by incremental clustering; is the number of nodes in the first part divided by the i - th node of the connected graph constructed by incremental clustering; is the number of nodes in the second part divided by the i - th node of the connected graph constructed by incremental clustering; is the number of node connection edges of the second part divided by the i - th node of the connected graph constructed by incremental clustering; R is the number of times of incremental clustering; norm is the normalization function; exp is the exponential function with the natural constant e as the base; is the absolute value symbol.

[0082] By judging the structural similarity features of two parts divided by different nodes in a connected graph, the node discrimination confidence of each node is obtained. And the node discrimination degrees obtained by a node in different incremental clustering are summed up, and the sum values of the node discrimination degrees obtained multiple times are used as the node discrimination confidence of the node, so as to avoid the situation that there is a large deviation when using the node discrimination degree obtained by a single incremental clustering as the node discrimination confidence of the node.

[0083] The node discrimination confidence of each node is obtained. Each node has one or more node connection relationships on the largest connected graph. Specifically, the node similarity between nodes can be adjusted through the node discrimination confidence. Specifically: taking the node discrimination confidence of each node as the weight, the node similarity is weighted, and based on the connected graph splitting clustering algorithm, the clustering clusters are determined.

[0084] The calculation formula of the weighted node similarity is:

[0085]

[0086] Wherein, is the weighted node similarity between the a-th node and the b-th node; is the node discrimination confidence of the k-th node in the a-th node and the b-th node; is the initial node similarity between the a-th node and the b-th node.

[0087] In the embodiment of the present invention, the initial node similarity between two nodes is the Euclidean distance between the two nodes. It should be noted that the node similarity between two nodes is calculated here, so the maximum value of k is 2.

[0088] After obtaining the weighted node similarity, further complete the subsequent connected graph splitting clustering algorithm to obtain an accurate clustering result, that is, determine the clustering clusters. Each clustering cluster contains multiple nodes, and the change trajectory of the touch position area corresponding to the nodes is the touch position area adjusted according to requirements.

[0089] Step S400, extract the sampling moments belonging to the associated time period in the clustering cluster as interaction moments; perform interpolation and supplementation on the touch position area of the interaction moments to obtain the interaction trajectory formed by the final touch position area.

[0090] According to the obtained clustering results, extract the blank area in the corresponding touch position area of the clustering cluster. This blank area is the defect caused by touch interruption. Then, perform linear interpolation processing to obtain the final touch position area trajectory. Among them, the time period corresponding to the blank area in the corresponding touch position area of the clustering cluster is also the selected associated time period. Therefore, according to the obtained clustering results, extracting the blank area in the corresponding touch position area of the clustering cluster is also extracting the sampling moments within the associated time period in the clustering cluster. Take the extracted sampling moments as interaction moments, and perform linear interpolation processing on the interaction moments to achieve interpolation and supplementation of the touch position area at the interaction moments. Furthermore, obtain the final interpolated and supplemented touch position area, and the touch position areas at different moments form the interaction trajectory of the early childhood education device.

[0091] And perform real-time human-computer interaction feedback according to the obtained interaction trajectory of the final touch position area, such as vibration feedback and sound feedback of the touchpad. It is also possible to estimate the pressing force based on the change in the touch area size in the time sequence of the final touch position area trajectory to construct a more user-friendly somatosensory feedback, thereby realizing the intelligent human-computer interaction of the early childhood education device.

[0092] In summary, the embodiments of the present invention relate to the technical field of teaching equipment interaction. The method first obtains the touch points during touch interaction on the early childhood education device of the infrared touchpad, and forms a touch position area from the touch points. Respectively obtain the time periods composed of consecutive sampling moments with touch points and the time periods composed of consecutive sampling moments without touch points on the early childhood education device as the touch time periods and blank time periods. Analyze the action change correlation degree of the change in touch points within two touch time periods adjacent to the blank time period to determine the action change correlation degree of the blank time period. Screen the blank time period according to the action change correlation degree of the blank time period to obtain the associated time period. Update the touch time period based on the associated time period. Construct multiple connected graphs based on the updated touch time period. Combine the structural distribution characteristics of the connected graphs to determine the node discrimination confidence of the nodes in the connected graphs. Use the node discrimination confidence as the weight to weight the node similarity, and based on the connected graph splitting clustering algorithm, determine the clustering clusters. Extract the sampling moments within the associated time period in the clustering clusters as the interaction moments. Perform interpolation and supplementation of the touch position area for the interaction moments to obtain the interaction trajectory formed by the final touch position area. By extracting the change characteristics of the touch action positions, accurately analyze the structural distribution correlation between nodes to eliminate the problem of inaccurate analysis results caused by multiple touch operations and touch interruption.

[0093] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the particular order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0094] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key point of each embodiment is to illustrate the differences from other embodiments.

Claims

1. A robot touch interaction control method, characterized in that, The method includes the following steps: Obtain the touch points generated during touch interaction on the early childhood education device with an infrared touchpad, and form a touch position area from the touch points; Respectively obtain the time periods composed of consecutive sampling moments with touch points on the early childhood education device and the time periods composed of consecutive sampling moments without touch points, as the touch time periods and blank time periods; analyze the correlation of action changes in the touch point changes within two touch time periods adjacent to the blank time period to determine the correlation of action changes in the blank time period; screen the blank time period according to the correlation of action changes in the blank time period to obtain the associated time period; update the touch time period based on the associated time period; Based on the updated touch time period, construct multiple connected graphs; combine the structural distribution characteristics of the connected graphs to determine the node discrimination confidence of the nodes in the connected graphs; use the node discrimination confidence as a weight to weight the node similarity, and determine the clustering clusters based on the connected graph splitting clustering algorithm; Extract the sampling moments within the associated time period in the clustering clusters as the interaction moments; perform interpolation and supplementation of the touch position area for the interaction moments to obtain the interaction trajectory formed by the final touch position area.

2. The robot touch interaction control method according to claim 1, wherein The analysis of the correlation of action changes in the touch point changes within two touch time periods adjacent to the blank time period to determine the correlation of action changes in the blank time period includes: Obtain the touch position change degree of each sampling moment within two touch time periods adjacent to the blank time period; According to the touch position change degree of different sampling moments within the touch time period and the length of the touch time period, determine the interval change correlation of the touch positions at different sampling moments; Combine the interval change correlations of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period to determine the correlation of action changes in the blank time period.

3. The robot touch interaction control method according to claim 2, wherein, The obtaining of the touch position change degree of each sampling moment within two touch time periods adjacent to the blank time period includes: Take any sampling moment within the touch time period as the target moment, use the difference in the number of touch points between the target moment and the previous sampling moment as the numerator, and use the time difference between the target moment and the previous sampling moment as the denominator. The fraction composed of the numerator and the denominator is used as the touch position change degree of the target moment.

4. The robot touch interaction control method according to claim 2, wherein, The calculation formula for the interval change correlation is: ； Among them, is the interval change correlation degree of the touch position at the i-th sampling moment within the touch time period; is the normalization function; is the time length of the touch time period in which the i-th sampling moment is located; is the touch position change degree at the i-th sampling moment within the touch time period; is the exponential function with the natural constant e as the base; is the number of sampling moments in the touch time period in which the i-th sampling moment is located; is the mean value of the touch position change degrees of all sampling moments within the touch time period.

5. The robot touch interaction control method according to claim 2, characterized in that, The combination of the interval change correlations of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period to determine the correlation of action changes in the blank time period includes: Calculate the difference in the interval change correlations of the touch positions corresponding to the sampling moments belonging to the touch time period at the left and right endpoints of the blank time period as the interval change degree; use the negative correlation normalization value of the interval change degree as the correlation of action changes in the blank time interval.

6. The robot touch interaction control method according to claim 1, wherein, The construction of multiple connected graphs based on the updated touch time period includes: Use the centroid point of the touch position area corresponding to a single sampling moment within the updated touch time period as a node of the connectivity graph, and use the number of nodes in the action time period within the touch time period as the preset incremental clustering increment value. Take the preset incremental clustering increment value as the number of nodes within the touch time period, and construct multiple connectivity graphs according to the connectivity graph splitting clustering algorithm.

7. The robot touch interaction control method according to claim 1, wherein The calculation formula for the node discrimination confidence is: ； ； Among them, is the node discrimination degree of the connected graph constructed by the i-th node in the r-th incremental clustering; is the node discrimination confidence of the i-th node; is the interval change correlation degree of the touch position at the i-th sampling moment within the touch time period; is the number of node connection edges in the first part of the connected graph constructed by the i-th node for incremental clustering; is the number of nodes in the first part of the connected graph constructed by the i-th node for incremental clustering; is the number of nodes in the second part of the connected graph constructed by the i-th node for incremental clustering; is the number of node connection edges in the second part of the connected graph constructed by the i-th node for incremental clustering; R is the number of incremental clustering times; norm is the normalization function; exp is the exponential function with the natural constant e as the base; is the absolute value symbol.

8. The robot touch interaction control method according to claim 1, wherein, Using the node discrimination confidence as the weight to weight the node similarity includes: The calculation formula for the weighted node similarity is: ； Among them, is the node similarity after weighting the a-th node and the b-th node; is the node discrimination confidence of the k-th node in the a-th node and the b-th node; is the initial node similarity of the a-th node and the b-th node.

9. The robot touch interaction control method according to claim 1, characterized in that, Based on the associated time period, updating the touch time period includes: Merge and update the associated time period and the adjacent touch time period to obtain the updated touch time period.

10. The robot touch interaction control method according to claim 1, wherein, According to the action change correlation degree of the blank time period, screening the blank time period to obtain the associated time period includes: Use the blank time period with the corresponding interval change correlation greater than the preset correlation degree threshold as the associated time period.

Images