Hand action generation method and device based on mechanical gloves

Abstract

The invention discloses a hand motion generation method and device based on a mechanical glove, and the method comprises the steps: obtaining a motion data set of the mechanical glove at a current moment, the motion set comprises the joint motion data of a hand, the pressure data of the hand, and the force data of the mechanical glove, according to the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, the motion score of the mechanical glove at the next moment of the current moment is estimated, and the mechanical glove is obtained based on the motion score of the mechanical glove at the next moment of the current moment; an action set of the mechanical glove at the next moment of the current moment is determined, the action set of the mechanical glove at the next moment of the current moment is executed, and the action set of the mechanical glove at the next moment of the current moment comprises joint actions and mechanical glove strength actions. By taking the motion score as a reference basis, the motion set of the mechanical glove at the next moment of the current moment is dynamically adjusted, so that the mechanical glove can treat the patient without adopting the preset motion.

Description

Technical Field

[0001] The present application relates to the technical field of mechanical gloves, and in particular to a method and device for generating hand movements based on mechanical gloves. Background Art

[0002] In the traditional rehabilitation model, the recovery of hand function relies on one-on-one manual operation and guidance from physical therapists. Although this approach is effective to a certain extent, it has many limitations, such as high treatment costs, uneven resource allocation, therapist fatigue, and restrictions on the frequency of patient visits. With the development of science and technology, robotic technology has been introduced into the field of rehabilitation, especially in hand function rehabilitation training. As a new type of assistive device, robotic gloves assist or enhance patients' hand function by imitating hand movements. However, most existing robotic gloves rely on preset movement patterns, which limits their ability to adapt to the specific needs of individual patients. Summary of the invention

[0003] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a method and device for generating hand movements based on a mechanical glove, which solves the problem that the mechanical gloves in the prior art rely on a preset movement mode.

[0004] According to one aspect of the present application, a method for generating hand motions based on a mechanical glove is provided, comprising:

[0005] Acquire a set of motion data of the mechanical glove at the current moment; wherein the motion set includes hand joint movement data, hand pressure data, and mechanical glove force data;

[0006] Estimate the motion score of the mechanical glove at the next moment after the current moment according to the joint motion data of the hand, the pressure data of the hand and the strength data of the mechanical glove;

[0007] Based on the motion score of the mechanical glove at the next moment after the current moment, determine the action set of the mechanical glove at the next moment after the current moment and execute the action set at the next moment after the current moment; wherein the action set at the next moment after the current moment includes joint actions and mechanical glove strength actions.

[0008] In one embodiment, estimating the motion score of the mechanical glove at the next moment of the current moment according to the joint motion data of the hand, the pressure data of the hand, and the strength data of the mechanical glove includes:

[0009] Get time difference error;

[0010] The motion score of the mechanical glove at the next moment after the current moment is calculated according to the time difference error, the joint movement data of the hand, the pressure data of the hand and the strength data of the mechanical glove.

[0011] In one embodiment, the calculating the motion score of the mechanical glove at the next moment after the current moment according to the time difference error, the joint movement data of the hand, the pressure data of the hand, and the strength data of the mechanical glove includes:

[0012] If the time difference error is at least two decimal places, the digit before the decimal point is 0, and the first decimal place after the decimal point is 0, then the motion score of the mechanical glove at the current moment is calculated according to the hand joint movement data, the hand pressure data, the mechanical glove strength data and the judgment function;

[0013] The motion score of the mechanical glove at the next moment after the current moment is calculated according to the motion score and the time difference error.

[0014] In one embodiment, after performing the action at the next moment of the current moment, the hand action generation method based on the mechanical glove further includes:

[0015] Calculating the friction force of the hand according to the pressure data of the hand and the strength data of the mechanical glove;

[0016] Obtaining a hand dexterity score according to the friction force of the hand and the joint movement data of the hand;

[0017] Based on the manual dexterity score, a treatment status outcome is determined.

[0018] In one embodiment, after obtaining the motion data set of the mechanical glove at the current moment, the hand motion generation method based on the mechanical glove further includes:

[0019] Obtaining a motion deviation according to a motion data set and a preset motion of the mechanical glove at a current moment;

[0020] Calculating a comprehensive score of the mechanical glove executing the action data set at the current moment based on the score corresponding to the action deviation;

[0021] Wherein, determining the action of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment and executing the action set at the next moment after the current moment includes:

[0022] Based on the comprehensive score and the motion score of the mechanical glove at the next moment after the current moment, the action set of the mechanical glove at the next moment after the current moment is adjusted.

[0023] In one embodiment, the calculating, based on the score corresponding to the action deviation, a comprehensive score of the mechanical glove executing the action data set at the current moment comprises:

[0024] Obtaining the completion time corresponding to the execution of the action data set by the mechanical glove at the current moment;

[0025] Obtain a score corresponding to the completion time;

[0026] According to the score corresponding to the action deviation and the score corresponding to the completion time, a comprehensive score of the mechanical glove executing the action data set at the current moment is calculated.

[0027] In one embodiment, calculating the comprehensive score of the mechanical glove executing the action data set at the current moment according to the score corresponding to the action deviation and the score corresponding to the completion time includes:

[0028] Obtaining the smoothness of the mechanical glove executing the action data set at the current moment;

[0029] Obtaining a score corresponding to the degree of stability;

[0030] According to the score corresponding to the action deviation, the score corresponding to the completion time, and the score corresponding to the degree of smoothness, a comprehensive score of the mechanical glove executing the action data set at the current moment is calculated.

[0031] In one embodiment, determining the treatment status result based on the hand dexterity score includes:

[0032] Obtain recovery time and recovery goals;

[0033] Based on the LSTM neural network, the treatment status result is obtained according to the rehabilitation time, rehabilitation goal and the hand flexibility score.

[0034] In one embodiment, determining the action set of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment comprises:

[0035] Based on the motion score of the mechanical glove at the next moment after the current moment, obtaining a plurality of motion data sets of the mechanical glove at the next moment after the current moment;

[0036] Determine an optimal action data set among multiple action data sets at a next moment of the current moment; wherein the optimal action data set includes joint activity data and mechanical glove force data;

[0037] Based on the optimal motion data set and the motion action function, a motion set of the mechanical glove at the next moment after the current moment is generated.

[0038] According to another aspect of the present application, a hand motion generation device based on a mechanical glove is provided, comprising:

[0039] An acquisition module is used to acquire a set of motion data of the mechanical glove at the current moment; wherein the motion set includes hand joint movement data, hand pressure data, and mechanical glove force data;

[0040] an estimation module, configured to estimate a motion score of the mechanical glove at a next moment after the current moment according to the joint motion data of the hand, the pressure data of the hand, and the strength data of the mechanical glove;

[0041] A determination module is used to determine the action set of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment and execute the action set at the next moment after the current moment; wherein the action set at the next moment after the current moment includes joint actions and mechanical glove strength actions.

[0042] The hand motion generation method and device based on the mechanical glove provided by the present application include: obtaining a motion data set of the mechanical glove at the current moment, wherein the motion set includes the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, estimating the motion score of the mechanical glove at the next moment of the current moment according to the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, determining the motion set of the mechanical glove at the next moment of the current moment based on the motion score of the mechanical glove at the next moment of the current moment and executing the motion set of the next moment of the current moment, wherein the motion set of the next moment of the current moment includes the joint motion and the strength motion of the mechanical glove. By performing motion scoring on the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, the motion set of the mechanical glove at the next moment of the current moment is dynamically adjusted based on the motion score and the motion set of the mechanical glove at the next moment of the current moment is executed, so as to realize personalized and accurate rehabilitation support, avoid dependence on preset motions, reduce the burden on therapists, improve rehabilitation efficiency, and improve patient rehabilitation experience and confidence through data feedback, which helps to complete rehabilitation training more efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0044] Figure 1 It is a schematic diagram of the structure of a mechanical glove provided by an exemplary embodiment of the present application.

[0045] Figure 2 It is a flowchart of a method for generating hand movements based on a mechanical glove provided by an exemplary embodiment of the present application.

[0046] Figure 3 It is a flow chart of the feedback mechanism of the mechanical glove provided by an exemplary embodiment of the present application.

[0047] Figure 4 It is a flowchart of hand motion generation based on a mechanical glove provided by another exemplary embodiment of the present application.

[0048] Figure 5 It is a structural schematic diagram of a hand motion generation device based on a mechanical glove provided by an exemplary embodiment of the present application.

[0049] Figure 6 It is a structural diagram of an electronic device provided by an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0050] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0051] Figure 1 It is a schematic diagram of the structure of a mechanical glove provided by an exemplary embodiment of the present application. Figure 2 FIG. 1 is a flow chart of a method for generating hand movements based on a mechanical glove provided by an exemplary embodiment of the present application. Figure 1-Figure 2 As shown, the hand motion generation method based on the mechanical glove includes:

[0052] Step 110: Obtain a set of motion data of the mechanical glove at the current moment, wherein the motion set includes hand joint movement data, hand pressure data, and mechanical glove force data.

[0053] In the embodiments of the present application, Figure 1As shown, the fingers of the mechanical gloves are equipped with pressure sensors, which are used to collect pressure data of the patient's hands. The joints of the mechanical gloves are equipped with joint sensors, which can collect joint activity data of the hands.

[0054] The hand joint activity data is represented by T, T = (t1, t2, ..., t n ), where t n is the nth joint data. Each t also contains the joint angle (degrees) θ, movement speed (degrees / second) v and movement direction f, etc. Therefore, T = (t1, t2, ..., t n ) can be expressed as T = {(θ1,v1,f1),(θ2,v2,f2)......,(θ n ,v n ,f n )},θ n Indicates the angle of the nth joint activity data, v n Indicates the speed of the nth joint activity data, f n Indicates the movement direction of the nth joint activity data.

[0055] The hand pressure data is represented by J, which indicates the force applied by the hand when contacting the object. n represents the nth region of the hand, so J = (j1, j2, ..., j n ), where j includes the position identifier of the area (such as palm, fingertip) P, the applied pressure value (Newton) N and the pressure duration (seconds) s, etc. Therefore, the pressure data of the hand is expressed as:

[0056] J={(P1,N 1, s1),(P2,N2,s2),......,(P n ,N n ,s n )}, where pn represents the location identifier of the nth region, N n Indicates the pressure value applied to the nth area, S n Indicates the pressure duration of the nth zone.

[0057] The mechanical glove force data is expressed as N, where N = (n1, n2, ..., n n ), where n n Represents the nth force module in the robotic glove. The force module in the robotic glove is used to detect the force applied during the robotic glove assistance process.

[0058] Step 120: Estimate the motion score of the mechanical glove at the next moment from the current moment based on the hand joint movement data, the hand pressure data, and the mechanical glove strength data.

[0059] In an embodiment of the present application, the motion score of the mechanical glove at the next moment from the current moment is estimated through the hand joint movement data, the hand pressure data and the mechanical glove strength data, so as to track the hand movement in rehabilitation training in real time and provide the action set at the next moment to adjust and optimize the rehabilitation training task.

[0060] Step 130: Based on the motion score of the mechanical glove at the next moment after the current moment, determine the action set of the mechanical glove at the next moment after the current moment and execute the action set of the next moment after the current moment, wherein the action set of the next moment after the current moment includes the joint movement data of the hand, the pressure data of the hand and the strength data of the mechanical glove.

[0061] In an embodiment of the present application, based on the motion score of the mechanical glove at the current moment, the system can determine the set of actions it should take at the next moment and execute these actions. This action set includes hand joint activity data, hand pressure data, and mechanical glove strength data. Specifically, the system first evaluates and scores the current state, then predicts the possible actions and their effects at the next moment, and selects the optimal set of actions to achieve the target movement. Joint movements involve adjusting the angle and position of the mechanical glove, while strength movements involve controlling the force or reaction force on the environment. Through moment-by-moment evaluation and adjustment, the system can optimize the overall motion performance of the mechanical glove and improve the accuracy and efficiency of task execution.

[0062] The hand motion generation method based on the mechanical glove provided by the present application comprises: obtaining a motion data set of the mechanical glove at the current moment, wherein the motion set comprises the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, estimating the motion score of the mechanical glove at the next moment of the current moment according to the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, determining the motion set of the mechanical glove at the next moment of the current moment based on the motion score of the mechanical glove at the next moment of the current moment and executing the motion set of the next moment of the current moment, wherein the motion set of the next moment of the current moment comprises the joint motion and the strength motion of the mechanical glove. By performing motion scoring on the joint activity data, the pressure data and the strength data of the mechanical glove of the hand, the motion set of the mechanical glove at the next moment based on the current moment is dynamically adjusted and executed, and then the optimal motion data set is selected through the action function of the mechanical glove. Thus, personalized and accurate rehabilitation support is realized. The method avoids dependence on preset motions, reduces the burden on therapists and improves rehabilitation efficiency. At the same time, through data feedback, the patient's rehabilitation experience and confidence are enhanced, which helps to complete rehabilitation training more efficiently.

[0063] In one embodiment, step 120 can be specifically implemented as follows: obtaining a time differential error; and calculating a motion score of the mechanical glove at the next moment from the current moment based on the time differential error, hand joint movement data, hand pressure data, and mechanical glove strength data.

[0064] In an embodiment of the present application, in a hand motion control system, first, the hand joint activity data, the hand pressure data, and the force data applied by the mechanical glove are obtained in real time through sensors, and the time differential error between the current moment and the previous moment is calculated. This error reflects the degree of change in the hand motion state. Based on this time differential error, a scoring function is established in combination with the flexibility of the hand joints (such as changes in joint angles and speeds), applied pressure (such as the strength when grasping objects), and the auxiliary effect of the mechanical glove on hand movements (such as output power and support). This function evaluates the motion performance of the mechanical glove at the current moment by integrating and analyzing these multidimensional data, and predicts its motion score at the next moment. The motion score not only quantifies the fluency and accuracy of the hand movements, but also provides data support for subsequent motion adjustments and optimizations, thereby improving the accuracy and comfort of hand movements.

[0065] In one embodiment, step 120 can be specifically implemented as follows: if the time differential error is at least two decimal places, the number of digits before the decimal point is 0, and the first decimal place after the decimal point is 0, then the motion score of the mechanical glove at the current moment is calculated based on the hand joint movement data, the hand pressure data, the mechanical glove strength data and the judgment function; based on the motion score and the time differential error, the motion score of the mechanical glove at the next moment after the current moment is calculated.

[0066] In an embodiment of the present application, the motion score of the mechanical glove at the next moment after the current moment is updated so that the time difference error is close to 0, and the motion score of the mechanical glove at the next moment after the current moment when the time difference error is close to 0 is obtained.

[0067] Specifically, the time difference (TD) error δ(t), as shown in the formula: δ(t) = R t+1 +γV(S t+1 )1-V(S t )1, where δ(t) represents the time difference error, r t represents the immediate reward after performing the action at time t, which may include the completion of the rehabilitation task, accuracy, fluency, etc. γ is the discount factor used to adjust the weight of future rewards. V(S t+1 )1 is the S at the next moment t+1 The motor score of the state, V(S t )1 is the current moment S t The motion score of the state, S = {T, J, N}, S represents the action set or state, T represents the joint activity data of the hand, J represents the pressure data of the hand, and N represents the force data of the mechanical glove. The process of minimizing the error: In rehabilitation training, suppose the robot hand needs to complete a grasping task, and the goal is to complete the grasping as accurately, quickly and smoothly as possible. The state of the mechanical glove at the current moment is S t , the robotic glove selects an action A at the current state t , after completing the action, enter the next moment state S t+1 , and receive instant rewards t , for example, based on the completion and time of the task. The time difference error δ(t) represents the current strategy for the state S t and S t+1 The difference between the estimated value of and the actual reward. By minimizing this error, we can ensure that the current strategy is getting closer and closer to the optimal strategy, so that the system selects action A each time. t Both can maximize the future cumulative rewards. Among them, the movement score represents the movement value of the mechanical glove.

[0068] Among them, the evaluation function is V(S)1, Among them, represents the motion score of the S state at time t, the motion score represents the motion value of the mechanical glove, E represents the expected value, which represents the expected value of the future cumulative reward based on the current strategy, γ represents the discount factor, which takes a value between 0<γ<1 and is used to adjust the weight of future rewards, and r t+ k +1 represents the reward at the future time step t+k+1, k>0 represents a certain moment in the future, r t+k+1 It may be related to the accuracy, time consumption, and fluency of the mechanical glove. t represents the time. In this application, the evaluation function needs to be updated, and V(S)1 is updated by V1(s)←V1(s)+ηδ(t), where V1(s) represents the learning function, η is the learning rate, and δ(t) represents the time difference error.

[0069] In one embodiment, after step 130, the hand motion generation method based on the mechanical glove can be specifically implemented as follows: calculating the friction of the hand according to the pressure data of the hand and the strength data of the mechanical glove; obtaining the hand flexibility score according to the friction of the hand and the hand joint movement data; and determining the treatment status result based on the hand flexibility score.

[0070] In the embodiment of the present application, the hand strength N is calculated by 手 . N 手 =ρ(JN), where ρ is a hyperparameter used to compensate for the loss of friction. Then, through N 手 And hand joint movement data to assess hand dexterity.

[0071] Specifically, set S = {N 手 ,T}, S represents the current state of the mechanical glove or the action data set at the current moment. The hand flexibility is evaluated by V(S)2, and the hand flexibility score is obtained, where Among them, V(S)2 represents the hand dexterity score, S * is the maximum index of the i-th specified task or the optimal state of the i-th specified task or the best action data set at the current moment, representing the ideal or target hand rehabilitation state, N is the number of times the index is reached, and w i Weight coefficient, used to adjust the relative importance of the error in the total loss, n is the number of sensors, S i Represents the state of the i-th specified task or the current state of the i-th specified task or the action data set at the current moment.

[0072] The treatment status results are then determined by scoring the hand dexterity level, helping therapists monitor the rehabilitation progress and adjust the training plan in a timely manner.

[0073] In one embodiment, after step 110, the hand motion generation method based on the mechanical glove can be specifically implemented as follows: according to the motion data set of the mechanical glove at the current moment and the preset motion, the motion deviation is obtained; based on the score corresponding to the motion deviation, the comprehensive score of the motion data set executed by the mechanical glove at the current moment is calculated; step 120 can be specifically implemented as follows: based on the comprehensive score and the motion score of the mechanical glove at the next moment after the current moment, the motion set of the mechanical glove at the next moment after the current moment is adjusted.

[0074] In the embodiment of the present application, assuming that the target object is grasped, the angle of the fingers, the rotation of the wrist, etc. can be detected by the sensor and compared with the corresponding preset action. Taking the mechanical glove grasping the target object as an example, if the deviation between the target angle between the mechanical glove and the target object and the preset angle is less than 5 degrees, the score is 90 points. If the deviation between the target angle and the preset angle is between 5 and 15 degrees, the score is 75 points. If the deviation between the target angle and the preset angle exceeds 15 degrees, the score is 50 points.

[0075] Then, the weight corresponding to the action deviation is used to calculate the comprehensive score of the action data set executed by the mechanical glove at the current moment, and then the action set at the next moment is adjusted by the comprehensive score and the motion score of the mechanical glove at the next moment of the current moment, so as to ensure that the action set at the next moment of the current moment can be more accurate. Among them, the action deviation includes angle deviation, rotation angle deviation, etc.

[0076] In one embodiment, after step 110, the hand motion generation method based on the mechanical glove can be specifically implemented as follows: obtaining the completion time corresponding to the action data set executed by the mechanical glove at the current moment; obtaining the score corresponding to the completion time; and calculating the comprehensive score of the action data set executed by the mechanical glove at the current moment based on the score corresponding to the action deviation and the score corresponding to the completion time.

[0077] In the embodiment of the present application, the completion time can affect the comprehensive score. For example, the mechanical glove should complete the action data set within 2 seconds at the current moment. If it is completed within 2 seconds, the time penalty is 0 points. If it exceeds 2 seconds, 5 points will be deducted for every additional 0.5 seconds, that is, -5.

[0078] In one embodiment, after step 110, the hand motion generation method based on the mechanical glove can be specifically implemented as follows: obtaining the smoothness corresponding to the action data set executed by the mechanical glove at the current moment; obtaining the score corresponding to the smoothness; and calculating the comprehensive score of the action data set executed by the mechanical glove at the current moment according to the score corresponding to the action deviation, the score corresponding to the completion time, and the score corresponding to the smoothness.

[0079] In the embodiment of the present application, the sensor can detect the smoothness of the manipulator's movements. Abrupt or unstable movements will reduce the fluency score, for example, the movement is smooth and uninterrupted: 100 points. The movement is slightly interrupted or unstable: 80 points. The movement has obvious jitter or interruption: 60 points.

[0080] Finally, the comprehensive score is calculated as:

[0081] Score=α1Accuracy+α2Time_Penalty+α3Smoothness

[0082] Among them, Accuracy: accuracy score, which indicates the score corresponding to the movement deviation, Time_Penalty indicates the score corresponding to the completion time, that is, time penalty, and points will be deducted as the movement time increases, Smoothness indicates the score corresponding to the smoothness, which indicates the smoothness of the movement, α1, α2, α3, the weights of each scoring factor, can be adjusted according to different rehabilitation tasks.

[0083] For another example, the robot's finger angle deviation is 6 degrees, so the accuracy score is 75 points. The robot took 2.5 seconds to complete the task, so the time penalty is 5×1=5 points. The robot's movement is slightly shaky, and the score is 80 points, so the comprehensive score is equal to Score=0.5×75+0.3×(100-5)+0.2×80=82 points.

[0084] Figure 3 FIG. 1 is a flow chart of a feedback mechanism of a mechanical glove provided by an exemplary embodiment of the present application. Figure 3 As shown, after step 130, the hand motion generation method based on the mechanical glove can be specifically implemented as follows: obtaining the rehabilitation time and rehabilitation goals; based on the LSTM neural network, obtaining the treatment status results according to the rehabilitation time, rehabilitation goals and hand flexibility scores.

[0085] In an embodiment of the present application, the user interface receives feedback S(t) to calculate the accuracy and fluency of the action, and compares it with the preset rehabilitation goal G to generate a feedback score. S(t) is an LSTM-based neural network, where t = {V(S)2, G, T}. It is used to discriminate data on the flexibility of the comprehensive time, rehabilitation goal G and evaluation function V(S)2 during the rehabilitation process. Through real-time data analysis and feedback mechanism, the mechanical glove can not only provide patients with accurate rehabilitation training support, but also help therapists monitor the rehabilitation process and adjust the training plan in time.

[0086] Figure 4 FIG. 1 is a flow chart of hand motion generation based on a mechanical glove provided by another exemplary embodiment of the present application. Figure 4As shown, step 130 can be specifically implemented as follows: based on the motion score of the mechanical glove at the next moment after the current moment, obtain multiple action data sets of the mechanical glove at the next moment after the current moment; determine the optimal action data set among the multiple action data sets at the next moment after the current moment; wherein the optimal action data set includes joint activity data and mechanical glove strength data; based on the optimal action data set and the action action function, generate the action set of the mechanical glove at the next moment after the current moment.

[0087] In the embodiment of the present application, based on the action score of the mechanical glove at the current moment, the movement at the next moment is calculated, and at the same time, multiple action data sets of the mechanical glove at the current moment are obtained based on the movement at the next moment, and then the optimal action data set is selected through the mechanical glove action function. The calculation formula of the mechanical glove action function is: A t =π θ (S t ), where A t represents the action selected by the mechanical glove at time t, S t represents the state of the mechanical glove at time t (including the hand angle, pressure and other sensor feedback data and other action data sets), π θ (S t ) represents the policy function, which is based on the current state S t Choose an optimal action A t , θ is the parameter of the policy function.

[0088] By A t Get the optimal action data set T′, N′, where T′ represents joint activity data and N′ represents mechanical glove force data, and then use π θ (·|S) generates the action set of the robotic glove at the next moment at the current moment, that is, S = {T, J, N}.

[0089] Action Generator π θ (·|S) The policy gradient update is expressed as: Among them, α is the learning rate, the step size of the control policy update, and δ t is the time difference error, is the gradient of the logarithmic policy function, which reflects the sensitivity of the probability of choosing a specific action in a given state to the policy parameters θ.

[0090] Figure 5 FIG. 1 is a schematic diagram of a hand motion generation device based on a mechanical glove provided by an exemplary embodiment of the present application. Figure 5As shown, a hand motion generation device based on a mechanical glove may include: an acquisition module 201, used to acquire a set of motion data of the mechanical glove at a current moment; wherein the motion set includes hand joint activity data, hand pressure data and mechanical glove strength data; an estimation module 202, used to estimate a motion score of the mechanical glove at the next moment after the current moment based on the hand joint activity data, hand pressure data and mechanical glove strength data; a determination module 203, used to determine a motion set of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment and execute the motion set of the next moment after the current moment; wherein the motion set of the next moment after the current moment includes joint motions and mechanical glove strength motions.

[0091] The hand motion generating device based on the mechanical glove provided by the present application comprises: an acquisition module acquires a set of motion data of the mechanical glove at the current moment, wherein the motion set comprises the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove; an estimation module estimates the motion score of the mechanical glove at the next moment of the current moment according to the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove; a determination module determines the motion set of the mechanical glove at the next moment of the current moment based on the motion score of the mechanical glove at the next moment of the current moment and executes the motion set of the next moment of the current moment, wherein the motion set of the next moment of the current moment comprises the joint motion and the strength motion of the mechanical glove. By performing motion scoring on the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove, the motion set of the mechanical glove at the next moment of the current moment is determined based on the motion score and the motion set of the mechanical glove at the next moment of the current moment is executed, so that the mechanical glove does not need to use preset motions to treat the patient, but can match the patient's hand treatment to achieve a better treatment effect.

[0092] In one embodiment, the estimation module 202 may be specifically configured to: obtain a time difference error;

[0093] According to the time difference error, hand joint movement data, hand pressure data and mechanical glove strength data, the motion score of the mechanical glove at the next moment from the current moment is calculated.

[0094] In one embodiment, the estimation module 202 can be specifically configured as follows: if the time differential error is at least two decimal places, the number of digits before the decimal point is 0, and the first decimal place after the decimal point is 0, then the motion score of the mechanical glove at the current moment is calculated based on the hand joint movement data, the hand pressure data, the mechanical glove strength data and the judgment function; based on the motion score and the time differential error, the motion score of the mechanical glove at the next moment after the current moment is calculated.

[0095] In one embodiment, after determining module 203, the hand motion generation device based on the mechanical glove can be specifically configured as follows: calculating the friction of the hand according to the pressure data of the hand and the strength data of the mechanical glove; obtaining the hand flexibility score according to the friction of the hand and the joint movement data of the hand; and determining the treatment status result based on the hand flexibility score.

[0096] In one embodiment, after the acquisition module 201, the hand motion generation device based on the mechanical glove can be specifically configured as follows: according to the motion data set of the mechanical glove at the current moment and the preset motion, the motion deviation is acquired; based on the score corresponding to the motion deviation, the comprehensive score of the motion data set executed by the mechanical glove at the current moment is calculated; wherein, the determination module 203 can be specifically configured as follows: based on the comprehensive score and the motion score of the mechanical glove at the next moment after the current moment, the motion set of the mechanical glove at the next moment after the current moment is adjusted.

[0097] In one embodiment, after obtaining module 201, the hand motion generation device based on the mechanical glove can be specifically configured as follows: obtaining the completion time corresponding to the action data set executed by the mechanical glove at the current moment; obtaining the score corresponding to the completion time; and calculating the comprehensive score of the action data set executed by the mechanical glove at the current moment based on the score corresponding to the action deviation and the score corresponding to the completion time.

[0098] In one embodiment, after obtaining module 201, the hand motion generation device based on the mechanical glove can be specifically configured as follows: obtaining the degree of smoothness corresponding to the action data set executed by the mechanical glove at the current moment; obtaining the score corresponding to the degree of smoothness; and calculating the comprehensive score of the action data set executed by the mechanical glove at the current moment according to the score corresponding to the action deviation, the score corresponding to the completion time, and the score corresponding to the degree of smoothness.

[0099] In one embodiment, after determining module 203, the hand motion generation device based on the mechanical glove can be specifically configured to: obtain rehabilitation time and rehabilitation goals; based on the LSTM neural network, obtain the treatment status results according to the rehabilitation time, rehabilitation goals and hand flexibility scores.

[0100] In one embodiment, the determination module 203 can be specifically configured as follows: based on the motion score of the mechanical glove at the next moment after the current moment, obtain multiple action data sets of the mechanical glove at the next moment after the current moment; determine the optimal action data set among the multiple action data sets at the next moment after the current moment; wherein the optimal action data set includes joint movement data and mechanical glove strength data; based on the optimal action data set and the action action function, generate the action set of the mechanical glove at the next moment after the current moment.

[0101] like Figure 6As shown, the electronic device 10 includes one or more processors 11 and a memory 12 .

[0102] The processor 11 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0103] The memory 12 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may run the program instructions to implement the above-mentioned hand motion generation method based on the mechanical glove and / or other desired functions of the various embodiments of the present application. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0104] In one example, the electronic device 10 may further include: an input device 13 and an output device 14, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0105] When the electronic device 10 is a stand-alone device, the input device 13 may be a communication network connector for receiving collected input signals from the first device and the second device.

[0106] In addition, the input device 13 may also include, for example, a keyboard, a mouse, etc.

[0107] The output device 14 can output various information to the outside, including the determined distance information, direction information, etc. The output device 14 can include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0108] Of course, to simplify, Figure 6 Only some of the components related to the present application in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0109] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0110] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0111] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A hand motion generation method based on a mechanical glove, characterized in that: include: Acquire a set of motion data of the mechanical glove at the current moment; wherein the motion set includes hand joint movement data, hand pressure data, and mechanical glove force data; Estimate the motion score of the mechanical glove at the next moment after the current moment according to the joint motion data of the hand, the pressure data of the hand and the strength data of the mechanical glove; Based on the motion score of the mechanical glove at the next moment after the current moment, determine the action set of the mechanical glove at the next moment after the current moment and execute the action set at the next moment after the current moment; wherein the action set at the next moment after the current moment includes hand joint movement data, hand pressure data and mechanical glove strength data.

2. The hand motion generation method based on the mechanical glove according to claim 1 is characterized in that: The estimating the motion score of the mechanical glove at the next moment after the current moment according to the joint motion data of the hand, the pressure data of the hand and the strength data of the mechanical glove comprises: Get time difference error; The motion score of the mechanical glove at the next moment after the current moment is calculated according to the time difference error, the joint movement data of the hand, the pressure data of the hand and the strength data of the mechanical glove.

3. The hand motion generation method based on the mechanical glove according to claim 2 is characterized in that: The calculating the motion score of the mechanical glove at the next moment after the current moment according to the time difference error, the joint activity data of the hand, the pressure data of the hand and the strength data of the mechanical glove comprises: If the time difference error is at least two decimal places, the digit before the decimal point is 0, and the first decimal place after the decimal point is 0, then the motion score of the mechanical glove at the current moment is calculated according to the hand joint movement data, the hand pressure data, the mechanical glove strength data and the judgment function; The motion score of the mechanical glove at the next moment after the current moment is calculated according to the motion score and the time difference error.

4. The hand motion generation method based on the mechanical glove according to claim 1 is characterized in that: After the action of executing the next moment of the current moment is performed, the method further includes: Calculating the friction force of the hand according to the pressure data of the hand and the strength data of the mechanical glove; Obtaining a hand dexterity score according to the friction force of the hand and the joint movement data of the hand; Based on the manual dexterity score, a treatment status outcome is determined.

5. The hand motion generation method based on the mechanical glove according to claim 4 is characterized in that: After obtaining the action data set of the mechanical glove at the current moment, the method further includes: Obtaining a motion deviation according to a motion data set and a preset motion of the mechanical glove at a current moment; Based on the score corresponding to the action deviation, calculating a comprehensive score of the mechanical glove executing the action data set at the current moment; Wherein, determining the action of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment and executing the action set at the next moment after the current moment includes: Based on the comprehensive score and the motion score of the mechanical glove at the next moment after the current moment, the action set of the mechanical glove at the next moment after the current moment is adjusted.

6. The method for generating hand motions based on a mechanical glove according to claim 5, characterized in that: The calculating, based on the score corresponding to the action deviation, of the comprehensive score of the mechanical glove executing the action data set at the current moment comprises: Obtaining the completion time corresponding to the execution of the action data set by the mechanical glove at the current moment; Obtain a score corresponding to the completion time; According to the score corresponding to the action deviation and the score corresponding to the completion time, a comprehensive score of the mechanical glove executing the action data set at the current moment is calculated.

7. The method for generating hand motions based on a mechanical glove according to claim 6, characterized in that: Calculating the comprehensive score of the mechanical glove executing the action data set at the current moment according to the score corresponding to the action deviation and the score corresponding to the completion time includes: Obtaining the smoothness of the mechanical glove executing the action data set at the current moment; Obtaining a score corresponding to the degree of stability; According to the score corresponding to the action deviation, the score corresponding to the completion time, and the score corresponding to the degree of smoothness, a comprehensive score of the mechanical glove executing the action data set at the current moment is calculated.

8. The method for generating hand motions based on a mechanical glove according to claim 4, characterized in that: Determining the treatment status result based on the hand dexterity score includes: Obtain recovery time and recovery goals; Based on the LSTM neural network, the treatment status result is obtained according to the rehabilitation time, rehabilitation goal and the hand flexibility score.

9. The hand motion generation method based on a mechanical glove according to claim 1, characterized in that: The determining of the action set of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment comprises: Based on the motion score of the mechanical glove at the next moment after the current moment, obtaining a plurality of motion data sets of the mechanical glove at the next moment after the current moment; Determine an optimal action data set among multiple action data sets at a next moment of the current moment; wherein the optimal action data set includes joint activity data and mechanical glove force data; Based on the optimal motion data set and the motion action function, a motion set of the mechanical glove at the next moment after the current moment is generated.

10. A hand motion generation device based on a mechanical glove, characterized in that: include: An acquisition module is used to acquire a set of motion data of the mechanical glove at the current moment; wherein the motion set includes hand joint movement data, hand pressure data, and mechanical glove force data; an estimation module, configured to estimate a motion score of the mechanical glove at a next moment after the current moment according to the joint motion data of the hand, the pressure data of the hand, and the strength data of the mechanical glove; A determination module is used to determine the action set of the mechanical glove at the next moment after the current moment based on the motion score of the mechanical glove at the next moment after the current moment and execute the action set at the next moment after the current moment; wherein the action set at the next moment after the current moment includes joint actions and mechanical glove strength actions.

Images