CN-122018705-A - Interactive behavior guidance control method and system based on execution feedback data

CN122018705ACN 122018705 ACN122018705 ACN 122018705ACN-122018705-A

Abstract

The invention relates to the technical field of intelligent interaction, in particular to an interactive behavior guiding control method and system based on execution feedback data, wherein a personalized baseline model comprising spatial pressure distribution, fluctuation characteristics and action time sequence characteristics is constructed by collecting pressure distribution data of a user contact interface; the method comprises the steps of calculating the space feature deviation degree and the time feature consistency of the current pressure distribution relative to a baseline model, judging the current state of a user and evaluating the violation degree of ergonomic constraint, searching a feasible target point closest to the current state in a feasible domain meeting constraint conditions, generating a corresponding guide vector, outputting a directional touch guide signal according to user individuation parameters, simultaneously modeling response characteristics of the guide signal on line and deducing the response state of the user based on the execution feedback of the user to the guide signal, and further adaptively adjusting the guide parameters, so that the interactive behavior guide control with accuracy, comfort and learning capability is realized.

Inventors

Guo Xianzhe
WANG QIAN
Pan Beijia
HUANG QIANNING

Assignees

杭州医学院

Dates

Publication Date: 20260512
Application Date: 20260415

Claims (8)

1. An interactive behavior guidance control method based on execution feedback data, comprising: calculating the space feature deviation degree and time feature consistency of the current pressure distribution and the personalized baseline model, and outputting a state classification label and a confidence score; Receiving a state classification label and a confidence score, evaluating constraint violation degree of current pressure distribution, searching a feasible target point closest to the current state of a user in a feasible domain meeting constraint conditions, and calculating a guide vector from a current pressure center to the feasible target point; The method comprises the steps of receiving the guiding vector, setting a time sequence interval and an intensity envelope according to pre-calibrated user personalized parameters, and generating a directional touch guiding signal; the intensity of the directional tactile guide signal is collected as input data and the displacement of the pressure center of the user is taken as output data, the second-order system parameters of the response characteristic of the user are obtained through a recursive least square method, and the current response state of the user is deduced.
2. The interactive behavior guidance control method based on execution feedback data according to claim 1, wherein the process of establishing a personalized baseline model comprises the steps of extracting pressure average values and fluctuation standard deviations of all sensing units from time sequence data in a preset initial stage, respectively constructing a spatial pressure distribution reference graph and a spatial fluctuation feature graph, identifying a continuous high-pressure area in the pressure distribution reference graph as a supporting area and recording a pressure proportion relation of the continuous high-pressure area; And storing the spatial pressure distribution reference graph, the spatial fluctuation feature graph, the pressure proportion relation and the adjustment action duration range as the personalized baseline model.
3. The interactive behavior guidance control method based on execution feedback data according to claim 1, wherein the process of calculating the spatial feature deviation degree and the temporal feature consistency includes calculating normalized deviation of real-time pressure of each sensing unit in the support area from a reference value based on a spatial pressure distribution reference map and a spatial fluctuation feature map, and counting a unit duty ratio of the normalized deviation exceeding a preset threshold value as the spatial feature deviation degree; Judging pressure change of each supporting area in a preset sliding time window to obtain pressure change trend of each area, calculating standard deviation of pressure change rate to obtain rate stability index, marking time characteristics as consistent gradual change when the pressure change trend of all the supporting areas is the same and the rate stability index is lower than a preset threshold value, and marking time characteristics as non-consistent change when the pressure change trend of the supporting areas is different and/or the rate stability index is higher than the preset threshold value.
4. The interactive behavior guidance control method based on execution feedback data according to claim 1, wherein the process of obtaining the feasible target point comprises calculating a normalized violation of a pressure concentration, a pressure gradient distribution, a pressure sum ratio between symmetric regions and an effective contact area ratio based on a preset ergonomic constraint, and obtaining a comprehensive constraint violation by weighted summation; And (3) projecting the pressure data into a low-dimensional feature space, carrying out iterative search along the opposite direction of the violation gradient until a candidate point meeting the constraint and/or having the minimum violation degree is found, and if the low-dimensional Euclidean distance between the candidate point and the current point is smaller than the maximum adjustment amplitude of the history of the user, determining the candidate point as a feasible target point, otherwise, adjusting the step length.
5. The interactive behavior guidance control method based on execution feedback data according to claim 4, wherein the process of calculating the guidance vector includes directly extracting the pressure center position information from the low-dimensional feature space coordinates of the feasible target point as the target pressure center position, calculating the pressure weighted center coordinates of the current pressure distribution as the current pressure center position, calculating the space vector pointing from the current pressure center position to the target pressure center position to obtain the guidance direction, calculating the euclidean distance between the current pressure center position and the target pressure center position as the offset distance, mapping the offset distance to the guidance intensity reference value by the piecewise saturation function, judging whether the guidance direction points to the effective contact area edge and/or the recorded uncomfortable region, if so, correcting the direction to the nearest safe region, and outputting the guidance vector containing the corrected direction and the intensity reference value.
6. The interactive behavior guidance control method based on execution feedback data according to claim 1, wherein the process of generating the directional tactile guidance signal includes selecting an activation sequence along a guidance direction in an actuator array, setting an optimal time interval and an intensity envelope according to a parameter with highest user recognition accuracy, generating each actuator driving signal according to the activation sequence so that the intensity envelopes of adjacent actuators overlap in a preset proportion in time, monitoring a pressure center trajectory, increasing an interval and increasing intensity contrast when an actual moving direction exceeds a guidance direction angle, increasing intensity and shortening the interval when long-time displacement is insufficient, and/or switching to a single-point continuous vibration mode.
7. The interactive behavior guidance control method based on execution feedback data according to claim 1, wherein the process of deducing the current response state of the user comprises time-aligning the tactile guidance signal intensity input with the pressure center displacement output to construct a second-order differential equation, minimizing the prediction error by using the input-output data sequence, solving equation coefficients and extracting damping and gain parameters representing the dynamics of the system, classifying the user state as a fatigue state with slow response, a awake state with quick response or a resistance state according to the combination of the damping and gain parameters relative to the respective threshold values, and generating an adjustment command for the control parameters according to the result.
8. An interactive behavior-guided control system based on execution feedback data, comprising: the system comprises a user state identification module, a current pressure distribution module, a user state classification module, a user state analysis module and a user state analysis module, wherein the user state identification module is used for acquiring pressure distribution data of a user contact interface and establishing a personalized baseline model; the feasible target point calculation module is used for receiving the state classification labels and the confidence scores, evaluating the constraint violation degree of the current pressure distribution, searching the feasible target point closest to the current state of the user in a feasible domain meeting constraint conditions, and calculating a guide vector from the current pressure center to the feasible target point; The touch sense sequence coding module is used for receiving the guide vector, setting a time sequence interval and an intensity envelope according to pre-calibrated user individuation parameters and generating a directional touch sense guide signal, monitoring the moving direction of a user pressure center in real time, and adjusting the time sequence interval and the intensity envelope when detecting that the user moves in an error direction and/or does not respond for a long time; and the parameter collaborative optimization module is used for acquiring the intensity of the directional tactile guide signal as input data and the displacement of the pressure center of the user as output data, acquiring the second-order system parameters of the response characteristic of the user through a recursive least square method, and deducing the current response state of the user.

Description

Interactive behavior guidance control method and system based on execution feedback data Technical Field The invention relates to the technical field of intelligent interaction, in particular to an interactive behavior guidance control method and system based on execution feedback data. Background With the rapid development of man-machine interaction technology, wearable interaction equipment is widely applied to scenes such as gesture assistance and behavior guidance. Such systems typically collect user gesture data via sensors and output guidance signals (e.g., vibrations, sounds, visual cues, etc.) based on preset rules or models to guide the user through specific actions or correct bad gestures. The existing interactive guiding control method mainly adopts an open-loop or simple closed-loop strategy. The simple closed loop mode can detect whether the user action reaches the target state or not, but only carries out binary judgment of 'up to standard/not up to standard', and lacks of deep analysis of feedback characteristics to the user. These methods have a common problem in that the response sensitivity of the target object to the guidance signal and the execution force level cannot be quantitatively evaluated, resulting in lack of basis for setting the subsequent guidance intensity. In particular, there is a significant difference in response characteristics of different users to the same intensity pilot signal. Some users react sensitively, the executive force is strong, the weaker guiding signal can produce the good effect, and some users need stronger or more frequent guiding to finish the action adjustment. The prior art adopts a uniform guidance intensity configuration, and the former can cause excessive intervention to cause conflict, and the latter can cause correction failure due to insufficient guidance. In addition, even if the same user is in different states (e.g., tired, distracted, etc.), its response characteristics may change and the guiding strategy of curing is difficult to accommodate for this dynamics. Due to the lack of an adaptive adjustment mechanism based on the execution feedback characteristics, the prior art cannot achieve accurate matching of the guidance intensity and the actual response capability of the user, so that the human-computer interaction efficiency is low, and the interaction experience is poor. Therefore, a technical solution capable of dynamically adjusting the guidance control strategy according to the feedback data executed by the user is needed. For this purpose, an interactive behavior guidance control method and system based on execution feedback data are provided. Disclosure of Invention The invention aims to provide an interactive behavior guiding control method and system based on execution feedback data, which are used for estimating the space and time deviation characteristics of current pressure distribution by constructing a personalized pressure baseline model, determining target points in a feasible domain meeting ergonomic constraint and generating directional touch guiding signals, and realizing self-adaptive and accurate behavior guiding control based on user execution feedback dynamic modeling of response characteristics. In order to achieve the above purpose, the present invention provides the following technical solutions: An interactive behavior guidance control method based on execution feedback data, comprising: calculating the space feature deviation degree and time feature consistency of the current pressure distribution and the personalized baseline model, and outputting a state classification label and a confidence score; Receiving a state classification label and a confidence score, evaluating constraint violation degree of current pressure distribution, searching a feasible target point closest to the current state of a user in a feasible domain meeting constraint conditions, and calculating a guide vector from a current pressure center to the feasible target point; The method comprises the steps of receiving the guiding vector, setting a time sequence interval and an intensity envelope according to pre-calibrated user personalized parameters, and generating a directional touch guiding signal; the intensity of the directional tactile guide signal is collected as input data and the displacement of the pressure center of the user is taken as output data, the second-order system parameters of the response characteristic of the user are obtained through a recursive least square method, and the current response state of the user is deduced. Preferably, the process of establishing the personalized baseline model comprises the steps of extracting the average pressure value and the standard deviation of fluctuation of each sensing unit from time sequence data in a preset initial stage, respectively constructing a space pressure distribution reference diagram and a space fluctuation characteristic diagram, identifying a continuous high-pre