A motion trajectory prediction method and system based on intelligent wearable devices

Abstract

The present invention provides a method and system for predicting a motion trajectory based on an intelligent wearable device, relating to the field of motion trajectory prediction. The method includes the following steps: S1: Detect the motion space where the person to be measured is located and the position change information of the person to be measured; S2: Extract features from the motion space; S3: Detect the skin activity information of multiple leg joints of the person to be measured; S4: Analyze the influence degree of environmental obstacle information on the motion trajectory of the person to be measured, and output initial trajectory prediction information; S5: Predict the local state trajectory of the person to be measured according to the skin activity information of the leg joints, and output local trajectory prediction information; S6: Comprehensively analyze the initial trajectory prediction information and the local trajectory prediction information, and output a motion trajectory prediction result. The present invention comprehensively predicts the motion trajectory by analyzing the influence degree of environmental obstacle information on the motion trajectory of the person to be measured and detecting the skin activity information of multiple leg joints of the person to be measured, thereby improving the accuracy of motion trajectory prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of motion trajectory prediction, and in particular, to a motion trajectory prediction method and system based on an intelligent wearable device. Background Art

[0002] With the popularization of intelligent wearable devices, people are paying more and more attention to their health conditions, and sports have become an indispensable part of many people's daily lives. However, for some sports enthusiasts and medical workers, they hope to better understand their own or their patients' motion states in order to better formulate exercise or rehabilitation plans. Motion trajectory prediction mainly collects users' motion data through intelligent wearable devices, and analyzes and processes the data to predict the users' future motion trajectories. Furthermore, it can help users better understand their own motion states, provide personalized exercise suggestions, and at the same time provide more accurate health data for users.

[0003] In the existing motion trajectory prediction technologies, the commonly adopted method is image acquisition. Although it solves certain motion trajectory prediction problems, there is still room for improvement in the prediction accuracy, and it is greatly affected by the environment. Summary of the Invention

[0004] The present invention provides a motion trajectory prediction method and system based on an intelligent wearable device to solve the defects in the existing technologies that the commonly adopted method is image acquisition, although it solves certain motion trajectory prediction problems, there is still room for improvement in the prediction accuracy, and it is greatly affected by the environment.

[0005] On the one hand, the present invention provides a motion trajectory prediction method based on an intelligent wearable device, and the specific steps include:

[0006] S1: Detect the motion space where the person to be measured is located and the position change information of the person to be measured; the position change information includes a position change state and a position unchanged state;

[0007] S2: Extract features from the motion space and perform environmental obstacle analysis, and output environmental obstacle information;

[0008] S3: Detect the skin activity information of multiple leg joints of the person to be measured, including skin stretching information, compression information, and torsion information; the leg joints include the hip joint, knee joint, and ankle joint;

[0009] S4: Analyze the influence degree of the environmental obstacle information on the motion trajectory of the person to be measured, and output initial trajectory prediction information;

[0010] S5: Predict the local state trajectory of the person to be measured according to the skin activity information of the leg joints, and output local trajectory prediction information;

[0011] S6: Perform comprehensive analysis on the initial trajectory prediction information and the local trajectory prediction information, and output the motion trajectory prediction result.

[0012] According to a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S2, the specific steps for feature extraction of the motion space include:

[0013] S21: Collect image information of the motion space;

[0014] S22: Preprocess the image information, including image scaling, normalization processing, and image enhancement;

[0015] S23: Establish an obstacle recognition model based on a convolutional neural network;

[0016] S24: Input the preprocessed image information into the trained obstacle recognition model for recognition, and output the obstacle information recognition result;

[0017] S25: Perform dynamic analysis on the obstacle information recognition result, and record the motion speed, acceleration, and movement pattern of the environmental obstacle information.

[0018] According to a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S23, the specific steps for establishing the obstacle recognition model include:

[0019] S231: Divide the data set in the motion database into a training set, a validation set, and a test set;

[0020] S232: Use cross-entropy loss as the loss function of the convolutional neural network;

[0021] S233: Use the love evolution algorithm to optimize the convolutional neural network;

[0022] S234: Use the training set to train the obstacle recognition model;

[0023] S235: Use the test set to evaluate the model, calculate the model accuracy, and stop training when the accuracy on the validation set no longer changes, to obtain the trained obstacle recognition model.

[0024] According to a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S233, the specific steps for hyperparameter optimization include:

[0025] S2331: Generate an initial love population, and each individual in the population serves as a hyperparameter combination;

[0026] S2332: Determine the fitness function of the individual; the formula is expressed as:

[0027] ;

[0028] Wherein, F(x) is the fitness function, N is the size of the love group, i is the index of the individual in the love group, and y i is the true label of the i-th data point, is the prediction result;

[0029] S2333: Selection stage, combine the fitness of each individual and use the roulette wheel method to select individuals;

[0030] S2334: Love stage, pair the individuals selected in the selection stage; high-quality individuals fall in love to generate new individuals; The formula is expressed as:

[0031] ;

[0032] Wherein, C j is the new individual, j is the position of the gene locus, P1[j] and P2[j] are two parent individuals, and α is the crossover coefficient;

[0033] S2335: Mutation operation, perform mutation on the newly generated individuals to increase diversity; The formula of the mutation operation is expressed as:

[0034] ;

[0035] Wherein, M j is the mutated individual, and β is the random perturbation;

[0036] S2336: Re-evaluate the fitness of the new generation of individuals;

[0037] S2337: Retain individuals with high fitness and replace individuals with low fitness to form a new generation of population;

[0038] S2338: Repeat steps S2333 - S2337 until the performance of the validation set no longer improves, then stop the iteration, and output the individual with the highest fitness found during the iteration as the optimal hyperparameter.

[0039] According to a method for predicting a motion trajectory based on an intelligent wearable device provided by the present invention, in step S3, the specific steps for detecting the skin activity information of multiple leg joints include:

[0040] S31: Establish a dot matrix for the skin areas at multiple leg joints of the person to be measured;

[0041] S32: Select multiple key points from the dot matrix as standard points;

[0042] S33: Calculate the Euclidean distance between the standard points and the dimensional change of the standard points; The detection method for the dimensional change includes:

[0043] S331: Select a standard point from each of the standard points of the joints as the initial origin;

[0044] S332: Record the relative coordinates of the standard points relative to the initial origin in the upright state of the person to be measured as the reference coordinates;

[0045] S333: Calculate the actual coordinates of the standard points when the legs of the person to be measured are in motion, and compare them with the reference coordinates to determine the dimensional changes of the standard points;

[0046] S34: Determine the stretching information and compression information according to the degree of change of the Euclidean distance; determine the torsion information according to the degree of change of the dimension.

[0047] According to a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S5, the specific steps of performing local state trajectory prediction include:

[0048] S51: Perform correlation matching between the skin activity information of multiple leg joints and the action information in the action database to determine the motion patterns corresponding to different skin activity information;

[0049] S52: Analyze the change of skin activity over time to identify repetitive motion patterns;

[0050] S53: Determine the change information of the motion amplitude of the local motion through the stretching information and compression information; determine the change information of the direction of the local motion through the torsion information;

[0051] S54: Integrate the motion amplitude change information and the direction change information and output local trajectory prediction information.

[0052] According to a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S54, the specific steps of performing information integration include:

[0053] S541: Determine the motion amplitude change information and direction change information corresponding to different leg joints;

[0054] S542: Extract the timestamps corresponding to the motion amplitude change information and the direction change information respectively, and determine the action time periods of the motion amplitude change information and the direction change information according to the timestamps;

[0055] S543: Analyze the consistency of the action time periods corresponding to different leg joints and output consistency information;

[0056] S544: Analyze the motion influence between the hip joint, knee joint and ankle joint according to the consistency information and output motion state information;

[0057] S545: Output local trajectory prediction information based on the motion state information.

[0058] In a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S544, the specific steps of performing motion impact analysis include:

[0059] S5441: Extract the change information of each joint and output joint state change information; the joint state change information includes: extracting the position change information of the knee joint relative to the hip joint and the position change information of the ankle joint relative to the hip joint according to the skin activity information of the hip joint; extracting the position change information of the ankle joint relative to the knee joint according to the skin activity information of the knee joint; extracting the state information of the foot according to the skin activity information of the ankle joint.

[0060] S5442: Output motion state information according to the joint state change information.

[0061] In a motion trajectory prediction method based on an intelligent wearable device provided by the present invention, in step S6, the specific steps of performing comprehensive analysis include:

[0062] S61: When the person to be measured is in a position change state, output initial trajectory prediction information and adjust the initial trajectory prediction information according to the local trajectory prediction information; when the person to be measured is in a position unchanged state, output local trajectory prediction information.

[0063] S62: When the person to be measured leaves the motion space, output the motion trajectory prediction result.

[0064] On the other hand, the present invention also provides a motion trajectory prediction system based on an intelligent wearable device, including:

[0065] A space information collection module for collecting environmental information.

[0066] A space information processing module for extracting environmental obstacle information from the environmental information.

[0067] An initial trajectory prediction module for analyzing the correlation between the environmental obstacle information and the person to be measured and outputting initial trajectory prediction information.

[0068] A skin activity information collection module for collecting the skin activity information of multiple leg joints of the person to be measured and outputting joint skin activity information.

[0069] A local trajectory prediction module for predicting the local state trajectory of the person to be measured according to the skin activity information of multiple leg joints and outputting local trajectory prediction information.

[0070] A motion trajectory prediction module for comprehensively analyzing the initial trajectory prediction information and the local trajectory prediction information and outputting the motion trajectory prediction result.

[0071] A method and system for predicting a motion trajectory based on an intelligent wearable device provided by the present invention analyze the influence degree of environmental obstacle information on the motion trajectory of a person to be measured and output initial trajectory prediction information; then detect the skin activity information of multiple leg joints of the person to be measured, and predict the local state trajectory of the person to be measured according to the skin activity information of the leg joints, and output local trajectory prediction information; the combination of the two prediction methods improves the accuracy of motion trajectory prediction; and reduces the influence of the environment on the prediction result.

[0072] A method and system for predicting a motion trajectory based on an intelligent wearable device provided by the present invention establish an obstacle recognition model based on a convolutional neural network, and use an evolutionary algorithm of love to optimize the convolutional neural network, which improves the recognition accuracy of the model for obstacle information, and further improves the robustness of motion trajectory prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0074] Figure 1 is a schematic flowchart of a method for predicting a motion trajectory based on an intelligent wearable device provided in Embodiment 1 of the present invention;

[0075] Figure 2 is a schematic structural diagram of a system for predicting a motion trajectory based on an intelligent wearable device provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0076] To make the objectives, technical solutions and advantages of the present invention clearer, the following will clearly and completely describe the technical solutions in the present invention with reference to the drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0077] The following will be combined with Figure 1 - Figure 2 Describe a method and system for predicting a motion trajectory based on an intelligent wearable device of the present invention.

[0078] Figure 1 is a schematic flowchart of a method for predicting a motion trajectory based on an intelligent wearable device provided in an embodiment of the present invention.

[0079] AsFigure 1 As shown in Figure 1 , a motion trajectory prediction method based on a smart wearable device provided by an embodiment of the present invention mainly includes the following steps:

[0080] S1: Detect the motion space where the person to be measured is located and the position change information of the person to be measured. The position change information includes a position change state and a position unchanged state. When detecting the motion space where the person to be measured is located, it is necessary to first determine the space range where the person to be measured is located, usually through devices such as motion sensors, cameras, and GPS for detection. After determining the space range, the position change information of the person to be measured is further determined.

[0081] S2: Extract features from the motion space and perform environmental obstacle analysis, and output environmental obstacle information. Features refer to some objects and the states of the objects contained in the motion space, which can be obtained through technical means such as image processing and point cloud processing. The environmental obstacle information refers to the features that have a certain obstructive effect on the movement of the person to be measured. Usually, it includes some obstacles and some movable people or objects. The specific steps for extracting features from the motion space include:

[0082] S21: Collect image information of the motion space. By using image acquisition devices such as cameras and drones, image data in the motion space is obtained. These image data contain information such as the position, shape, and color of the objects.

[0083] S22: Preprocess the image information, including image scaling, normalization processing, and image enhancement. The preprocessing is used to improve the quality of the image data, making it more suitable for feature extraction and recognition. Image scaling is to adjust the image to a unified size. Normalization processing is used to eliminate the intensity differences caused by factors such as illumination and shooting angle in the image, and improve the contrast of the image. Image enhancement is used to highlight important features such as edges and corners in the image, and improve the accuracy of feature extraction.

[0084] S23: Establish an obstacle recognition model based on a convolutional neural network. A convolutional neural network (CNNs) is basically composed of a convolutional layer, an activation layer, a pooling layer, a fully connected layer, a batch normalization layer, and a dropout layer. The convolutional layer is the core of the CNN, which extracts features of the input data through convolutional operations. Each convolutional layer consists of multiple convolutional kernels that slide over the input data to generate feature maps. The activation layer usually uses ReLU (Rectified Linear Unit) as the activation function to introduce non-linearity and help the network learn more complex features. The pooling layer is used to reduce the dimension of the feature maps, reduce the computational amount, and at the same time retain important features. Common pooling operations include max pooling and average pooling. The fully connected layer is located at the end of the network and is used to flatten all feature maps and connect them to the fully connected layer for classification or regression tasks. The batch normalization and dropout layers are used to accelerate the training process and reduce overfitting. The specific steps to establish the obstacle recognition model are as follows:

[0085] S231: Divide the dataset in the motion database into a training set, a validation set, and a test set. The training set is used to train the model, the validation set is used to adjust the parameters of the model during training, and the test set is used to evaluate the performance of the model. The proportion of each parameter set is usually 70% for the training set, 15% for the validation set, and 15% for the test set.

[0086] S232: Use cross-entropy loss as the loss function of the convolutional neural network. Cross-entropy loss is used to measure the difference between the output of the model and the true labels. By minimizing the cross-entropy loss, we can train an accurate obstacle recognition model. The formula of the function is expressed as:

[0087] ;

[0088] In the formula, M is the total number of samples, representing the degree of difference between the model prediction result and the true label. K is the number of classes, a is the sample label, b is the class label, x ab is the actual label. When the sample a belongs to the class b, x ab is 1, otherwise it is 0. is the predicted probability.

[0089] S233: Use the love evolution algorithm to optimize the convolutional neural network. The specific steps for hyperparameter optimization are as follows:

[0090] S2331: Generate an initial love population, and each individual in the population is used as a combination of hyperparameters.

[0091] S2332: Determine the fitness function of the individual. The formula is expressed as:

[0092] ;

[0093] In the formula, F is the fitness function, N is the size of the love group, i is the index of the individual in the love group, and y i is the true label of the i-th data point, and

[0094] S2333: Selection stage, that is, select individuals for reproduction according to the fitness of the individuals. Combining the fitness of each individual, the roulette wheel method is used for individual selection. The formula of the roulette wheel is expressed as:

[0095] ;

[0096] In the formula, P i is the probability that the i-th individual is selected, F i is the fitness of the i-th individual, and

[0097] is the sum of the fitnesses of all individuals in the love group. Then generate a random number r ∈ [0, 1]. Select individuals until the cumulative probability reaches the random number r.

[0098] ;

[0099] In the formula, C is the new individual, P1 and P2 are two parent individuals, α is the crossover coefficient, and α is a random number between [0, 1].

[0100] S2335: Mutation operation, mutate the newly generated individuals to increase diversity. The mutation stage simulates gene mutation. Randomly change each gene position of the individual. The formula of the mutation operation is expressed as:

[0101] ;

[0102] In the formula, M j is the mutated individual. β is the random perturbation, which follows the normal distribution. The formula of the normal distribution is expressed as:

[0103] ;

[0104] In the formula, x is the random variable, the mean μ is set to 0, and the standard deviation σ is an index to measure the width of the data distribution. Its value directly affects the size of the perturbation. The larger the standard deviation, the greater the perturbation and the larger the search space; the smaller the standard deviation, the smaller the perturbation and the smaller the search space.

[0105] S2336: Re-evaluate the fitness of the new generation of individuals.

[0106] S2337: Retain the individuals with high fitness and replace those with low fitness to form a new generation of population.

[0107] S2338: Repeat steps S2333 - S2337 until the performance on the validation set no longer improves, then stop the iteration and output the individual with the highest fitness found during the iteration as the optimal hyperparameter.

[0108] S234: Use the training set to train the obstacle recognition model. Update the weights in each epoch. During one training, the model traverses the entire training dataset once, which is called an epoch. The way to update the weights is expressed as:

[0109] ;

[0110] where θ is the parameter, η is the learning rate, is the gradient of the loss function with respect to the parameter.

[0111] S235: Use the test set to evaluate the model, calculate the model accuracy. When the accuracy on the validation set no longer changes, the highest accuracy of the model training can be obtained at this time. Then stop the training to get the trained obstacle recognition model.

[0112] S24: Input the pre - processed image information into the trained obstacle recognition model for recognition, and output the recognition result of the obstacle information. The recognition result usually includes information such as the type, location, and size of the obstacle. Then perform processing such as result screening, fusion, and optimization on the recognition result to improve the accuracy and practicality of the recognition result.

[0113] S25: Perform dynamic analysis on the recognition result of the obstacle information, and record the movement speed, acceleration, and movement pattern of the environmental obstacle information. After the model recognizes the obstacle information, perform dynamic analysis on it by identifying the feature changes of the obstacle information. The analysis results include the speed, acceleration, movement pattern of the obstacle information, and the distance from the measured person.

[0114] S3: Detect the skin activity information of multiple leg joints of the measured person, including skin stretching information, compression information, and torsion information. Usually, flexible sensors are used to detect the skin activity information. The leg joints include the hip joint, knee joint, and ankle joint. The specific steps to detect the skin activity information of multiple leg joints are as follows:

[0115] S31: Establish a dot matrix for the skin areas at multiple leg joints of the measured person. Mark some key points at the leg joints, and these key points can represent the center of the joint or important positions of the bones. Then, establish a dot matrix around these key points, and the size and density of the dot matrix can be determined according to actual needs.

[0116] S32: Select multiple key points from the dot matrix as standard points. When selecting standard points, the range of motion and direction of movement of the joints need to be considered to ensure that the standard points can comprehensively reflect the movement of the joints.

[0117] S33: Calculate the Euclidean distance between the standard points and the dimensional changes of the standard points. The detection methods for dimensional changes include:

[0118] S331: Select one standard point from the standard points of each joint as the initial origin.

[0119] S332: Record the relative coordinates of the standard points relative to the initial origin in the upright state of the person being measured as the reference coordinates.

[0120] S333: Calculate the actual coordinates of the standard points when the person being measured's leg is in motion and compare them with the reference coordinates to determine the dimensional changes of the standard points.

[0121] S34: Determine the stretching information and compression information based on the degree of change in the Euclidean distance. Determine the torsional information based on the degree of change in the dimension. The calculation method of the Euclidean distance is expressed as:

[0122] ;

[0123] In the formula, d(A,B) is the Euclidean distance between points A and B, and (c1,d1,e1) and (c2,d2,e2) are the relative coordinates of A and B respectively. The degree of change in the dimension is judged based on the change in the relative coordinates between points A and B. If the coordinates in the skin normal direction change, it means that the dimension of the standard point has changed, that is, the skin has twisted.

[0124] S4: Analyze the degree of influence of the environmental obstacle information on the movement trajectory of the person being measured and output the initial trajectory prediction information. The formula for analyzing the degree of influence is expressed as:

[0125] ;

[0126] In the formula, g(E) is the movement direction prediction function, D0 is the influence value of the obstacle in the environmental obstacle information on the movement trajectory of the person being measured; D1 is the degree of influence of the ground condition, which is divided into having an influence and having no influence, and is represented by a numerical value as D1 ∈ {0,1}; h0 is the intercept, h1 and h2 are the weight coefficients of the obstacle influence factor and the ground condition influence factor respectively, and ε is the error term.

[0127] S5: Predict the local state trajectory of the person being measured based on the skin activity information of the leg joints and output the local trajectory prediction information. The specific steps for predicting the local state trajectory include:

[0128] S51: Correlate the multiple leg joint skin activity information with the action information in the action database to determine the motion patterns corresponding to different skin activity information. Motion patterns typically include walking, running, jumping, etc.

[0129] S52: Analyze the change of skin activity over time to identify repetitive motion patterns. Establish a set of the same motion patterns for the repetitive motion patterns, which contains different skin activity information under the same motion pattern.

[0130] S53: Determine the change information of the motion amplitude of local motion through the stretching information and compression information. Analyze the stretching information and compression information of the skin to determine the size of the motion amplitude of each joint. The greater the degree of stretching or compression, the greater the motion amplitude of the joint. Determine the change information of the direction of local motion through the torsion information. The skin will twist as the joint moves, and this torsion can accurately judge the motion direction of each joint.

[0131] S54: Integrate the motion amplitude change information and the direction change information, and output the local trajectory prediction information. The specific steps for information integration include:

[0132] S541: Determine the motion amplitude change information and the direction change information corresponding to different leg joints.

[0133] S542: Extract the timestamps corresponding to the motion amplitude change information and the direction change information respectively, and the exact moment when each motion amplitude change information and direction change information occurs can be determined. And determine the action time period of the motion amplitude change information and the direction change information according to the timestamps. That is, the process time between the start and end times of the motion amplitude change information and the direction change information.

[0134] S543: Analyze the consistency of the action time periods corresponding to different leg joints, and output the consistency information. It is used to determine whether there is coordinated motion or mutual influence between joints.

[0135] S544: According to the consistency information, analyze the motion influence between the hip joint, knee joint and ankle joint, and output the motion state information. The specific steps for motion influence analysis include:

[0136] S5441: Extract the change information of each joint and output the joint state change information. The joint state change information includes: according to the skin movement information of the hip joint, extract the position change information of the knee joint relative to the hip joint and the position change information of the ankle joint relative to the hip joint. According to the skin movement information of the knee joint, extract the position change information of the ankle joint relative to the knee joint. According to the skin movement information of the ankle joint, extract the state information of the foot. The hip joint is a ball-and-socket joint that allows the thigh to move in multiple planes, including flexion, extension, adduction, abduction, and rotation. The knee joint mainly performs flexion and extension movements. During walking and running, the knee joint plays a key role in supporting body weight and absorbing shock. When the hip joint flexes, the knee joint also flexes accordingly; when the hip joint extends, the knee joint also extends. The ankle joint is mainly responsible for dorsiflexion and plantar flexion of the foot. During walking and running, plantar flexion of the ankle joint helps to propel the body forward.

[0137] S5442: Output the motion state information according to the joint state change information. That is, the motion state information of the leg corresponding to the movement of each joint. Different joint movements will produce different motion effects. For example, during walking, when the hip joint on one side flexes, the knee joint and ankle joint on the same side also flex to prepare for the sole of the foot to touch the ground. Subsequently, when the hip joint extends, the knee joint and ankle joint also extend to propel the body forward.

[0138] S545: Output the local trajectory prediction information according to the motion state information. By analyzing the motion state of each joint, predict the possible motion state changes of each joint in the next motion cycle, and then perform short-term motion trajectory prediction.

[0139] S6: Conduct comprehensive analysis on the initial trajectory prediction information and the local trajectory prediction information, and output the motion trajectory prediction result. The specific steps for comprehensive analysis include:

[0140] S61: When the person being measured is in a position change state, output the initial trajectory prediction information and adjust the initial trajectory prediction information according to the local trajectory prediction information. When the person being measured is in a position unchanged state, output the local trajectory prediction information.

[0141] S62: When the person being measured leaves the motion space, output the motion trajectory prediction result. This motion trajectory prediction result is a prediction of the motion trajectory of the person being measured in the next motion space, and the prediction result will be updated as the person being measured moves.

[0142] In summary, the present invention analyzes the degree of influence of environmental obstacle information on the movement trajectory of the person to be measured and outputs initial trajectory prediction information; then detects the skin activity information of multiple leg joints of the person to be measured, and predicts the local state trajectory of the person to be measured according to the skin activity information of the leg joints, and outputs local trajectory prediction information; the combination of the two prediction methods improves the accuracy of movement trajectory prediction; and reduces the influence of the environment on the prediction result. By establishing an obstacle recognition model based on a convolutional neural network and using an evolutionary algorithm of love to optimize the convolutional neural network, the recognition accuracy of the model for obstacle information is improved, and thus the robustness of movement trajectory prediction is improved.

[0143] Embodiment 2:

[0144] As Figure 2 shown, the present invention also provides a movement trajectory prediction system based on an intelligent wearable device, which can be implemented by using a movement trajectory prediction method based on an intelligent wearable device as in Embodiment 1. The prediction system includes:

[0145] A spatial information acquisition module for acquiring environmental information. This includes, but is not limited to, geographical location, terrain features, obstacle positions and types, etc. These information are usually obtained through sensors.

[0146] A spatial information processing module for extracting environmental obstacle information from the environmental information. The spatial information processing module extracts environmental obstacle information based on the recognition of features in the environment by a convolutional neural network.

[0147] An initial trajectory prediction module for analyzing the correlation between the environmental obstacle information and the person to be measured and outputting initial trajectory prediction information. The initial trajectory prediction module uses the environmental obstacle information obtained from the spatial information processing module to predict its possible movement trajectory. This module analyzes the correlation between the environmental obstacle information and the object to be measured based on a convolutional neural network and relative distance calculation, and generates a preliminary trajectory prediction.

[0148] A skin activity information acquisition module for acquiring the skin activity information of multiple leg joints of the person to be measured and outputting joint skin activity information. This module monitors and records the skin activity at the leg joints of the human body through sensors attached to the human skin.

[0149] A local trajectory prediction module for predicting the local state trajectory of the person to be measured according to the skin activity information of multiple leg joints and outputting local trajectory prediction information. This module infers the state of human movement, such as walking, running, jumping, etc., by analyzing the skin activity information of the leg joints.

[0150] The motion trajectory prediction module is used to comprehensively analyze the initial trajectory prediction information and the local trajectory prediction information and output the motion trajectory prediction result. When the person to be measured is in a state of position change, it outputs the initial trajectory prediction information and adjusts the initial trajectory prediction information according to the local trajectory prediction information. When the person to be measured is in a state of unchanged position, it outputs the local trajectory prediction information. When the person to be measured leaves the motion space, it outputs the motion trajectory prediction result. This motion trajectory prediction result is a prediction of the motion trajectory of the person to be measured in the next motion space and will update the prediction result as the person to be measured moves.

[0151] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0152] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0153] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features. These modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for predicting a motion trajectory based on an intelligent wearable device, characterized in that, The specific steps include: S1: Detect the movement space where the person to be tested is located and the position change information of the person to be tested; the position change information includes a position change state and a position unchanged state; S2: Extract features from the movement space and perform environmental obstacle analysis, and output environmental obstacle information; S3: Detect the skin activity information of multiple leg joints of the person to be tested, including skin stretching information, compression information, and torsion information; the leg joints include the hip joint, knee joint, and ankle joint; S4: Analyze the degree of influence of the environmental obstacle information on the movement trajectory of the person to be tested, and output initial trajectory prediction information; S5: Predict the local state trajectory of the person to be tested based on the skin activity information of the leg joints, and output local trajectory prediction information; S6: Perform comprehensive analysis on the initial trajectory prediction information and the local trajectory prediction information, and output a movement trajectory prediction result; the specific steps for comprehensive analysis include: S61: When the person to be tested is in the position change state, output the initial trajectory prediction information, and adjust the initial trajectory prediction information according to the local trajectory prediction information; when the person to be tested is in the position unchanged state, output the local trajectory prediction information; S62: When the person to be tested leaves the movement space, output the movement trajectory prediction result.

2. The method for predicting a motion trajectory based on an intelligent wearable device according to claim 1, wherein, In step S2, the specific steps for extracting features from the movement space include: S21: Collect the image information of the movement space; S22: Preprocess the image information, including image scaling, normalization processing, and image enhancement; S23: Establish an obstacle recognition model based on a convolutional neural network; S24: Input the preprocessed image information into the trained obstacle recognition model for recognition, and output an obstacle information recognition result; S25: Perform dynamic analysis on the obstacle information recognition result, and record the movement speed, acceleration, and movement pattern of the environmental obstacle information.

3. The motion trajectory prediction method based on a smart wearable device according to claim 2, wherein In step S23, the specific steps for establishing the obstacle recognition model include: S231: Divide the data set in the movement database into a training set, a validation set, and a test set; S232: Use cross-entropy loss as the loss function of the convolutional neural network; S233: Use the love evolution algorithm to optimize the convolutional neural network; S234: Use the training set to train the obstacle recognition model; S235: Use the test set to evaluate the model, calculate the model accuracy, and stop training when the accuracy on the validation set no longer changes, and obtain the trained obstacle recognition model.

4. A method for predicting a motion trajectory based on an intelligent wearable device according to claim 3, characterized in that, In step S233, the specific steps for hyperparameter optimization include: S2331: Generate an initial love population, and each individual in the population is used as a hyperparameter combination; S2332: Determine the fitness function of the individual; the formula is expressed as: Wherein, F(x) is the fitness function, N is the size of the love group, i is the index of the individual in the love group, and y i is the true label of the i-th data point, is the prediction result; S2333: In the selection stage, combine the fitness of each individual and use the roulette wheel method to select individuals; S2334: In the courtship stage, pair the individuals selected in the selection stage; high-quality individuals court each other to generate new individuals; the formula is expressed as: C j = αP1[j] + (1 - α)P2[j] Where C j is the new individual, j is the position of the gene locus, P1[j] and P2[j] are two parental individuals, and α is the crossover coefficient; S2335: Mutation operation, which mutates the newly generated individuals to increase diversity; the formula representation of the mutation operation is: M j = C j + β where M j is the mutated individual and β is the random perturbation; S2336: Re-evaluate the fitness of the new generation of individuals; S2337: Retain the individuals with high fitness and replace the individuals with low fitness to form a new generation of population; S2338: Repeat steps S2333 - S2337 until the performance of the validation set no longer improves, then stop the iteration and output the individual with the highest fitness found during the iteration as the optimal hyperparameter.

5. A method for predicting a motion trajectory based on a smart wearable device according to claim 1, characterized in that, In step S3, the specific steps for detecting the skin activity information of multiple leg joints include: S31: Establish a dot matrix for the skin areas at multiple leg joints of the person to be measured; S32: Select multiple key points from the dot matrix as standard points; S33: Calculate the Euclidean distance between the standard points and the dimensional changes of the standard points; the detection method for the dimensional changes includes: S331: Select one standard point from the standard points of each joint as the initial origin; S332: Record the relative coordinates of the standard points with respect to the initial origin in the upright state of the person to be measured as the reference coordinates; S333: Calculate the actual coordinates of the standard points when the person's legs are in motion and compare them with the reference coordinates to determine the dimensional changes of the standard points; S34: Determine the stretching information and the compression information according to the degree of change of the Euclidean distance; determine the torsion information according to the degree of change of the dimension.

6. The motion trajectory prediction method based on a smart wearable device according to claim 5, characterized in that, In step S5, the specific steps for performing local state trajectory prediction include: S51: Perform correlation matching between the skin activity information of multiple leg joints and the action information in the action database to determine the motion patterns corresponding to different skin activity information; S52: Analyze the change of skin activity over time to identify repetitive motion patterns; S53: Determine the motion amplitude change information of local motion through the stretching information and the compression information; determine the direction change information of local motion through the torsion information; S54: Integrate the motion amplitude change information and the direction change information and output local trajectory prediction information.

7. The method for predicting a motion trajectory based on a smart wearable device according to claim 6, wherein In step S54, the specific steps for information integration include: S541: Determine the motion amplitude change information and the direction change information corresponding to different leg joints; S542: Extract the timestamps corresponding to the motion amplitude change information and the direction change information respectively, and determine the action time periods of the motion amplitude change information and the direction change information according to the timestamps; S543: Analyze the consistency of the action time periods corresponding to different leg joints and output consistency information; S544: Analyze the motion influence between the hip joint, the knee joint and the ankle joint according to the consistency information and output motion state information; S545: Output local trajectory prediction information according to the motion state information.

8. A method for predicting a motion trajectory based on an intelligent wearable device according to claim 6, characterized in that, In step S544, the specific steps for performing motion influence analysis include: S5441: Extract the change information of each joint and output the joint state change information; the joint state change information includes: according to the skin activity information of the hip joint, extract the position change information of the knee joint relative to the hip joint and the position change information of the ankle joint relative to the hip joint; according to the skin activity information of the knee joint, extract the position change information of the ankle joint relative to the knee joint; according to the skin activity information of the ankle joint, extract the state information of the foot. S5442: Output the motion state information according to the joint state change information.

9. A motion trajectory prediction system based on a smart wearable device, which adopts a motion trajectory prediction method based on a smart wearable device as described in any one of claims 1 to 8, characterized in that, The motion trajectory prediction system includes: A spatial information acquisition module for acquiring environmental information; A spatial information processing module for extracting the environmental obstacle information from the environmental information; An initial trajectory prediction module for analyzing the correlation between the environmental obstacle information and the person to be measured and outputting initial trajectory prediction information; A skin activity information acquisition module for acquiring the skin activity information of multiple leg joints of the person to be measured and outputting joint skin activity information; A local trajectory prediction module for predicting the local state trajectory of the person to be measured according to the skin activity information of multiple leg joints and outputting local trajectory prediction information; A motion trajectory prediction module for comprehensively analyzing the initial trajectory prediction information and the local trajectory prediction information and outputting a motion trajectory prediction result.

Images