Wearable motion data acquisition equipment for motion intention recognition

Abstract

The utility model discloses a wearable exercise data acquisition device for exercise intention recognition, and belongs to the field of exercise data acquisition. Comprising an external material layer, a granular pressure sensor array, a memory alloy pressure applying assembly, a flexible skin contact layer, an inertial material assembly and a power supply and signal acquisition assembly arranged on the external material layer. A granular pressure sensor array is arranged in the outer material layer, and after pressure sensor data are processed, motion intentions (including directions and sizes) can be judged through a motion intention recognition model. The wearable motion intention recognition equipment can effectively distinguish a static state from a micro motion state, a memory alloy pressure applying mode ensures full contact and initial pressure between the pressure sensor array and a measured limb, and the equipment is designed into a chain belt mode and can adapt to limb parts of different sizes; the initial state is easy to calibrate and identify, and the electrical noise influence is small; and a solution is provided for identification of a tiny motion intention.

Description

Technical Field

[0001] The utility model relates to the field of data collection, in particular to a wearable sports data collection device. Background Art

[0002] With the increasing number of patients suffering from hemiplegia caused by cerebrovascular disease, exoskeleton robots that can provide wearers with functions such as movement assistance and rehabilitation training have become a hot topic of research. However, human-robot interaction requires the robot to actively understand the user's movement intentions. Under the influence of the disease, patients have a very limited range of autonomous limb movement, and traditional human-robot interaction methods based on programmable control cannot meet these requirements.

[0003] Based on the type of signal, intention recognition methods can be categorized as those based on human bioelectrical signals and those based on human-computer interaction force information. Research based on human-computer interaction force information focuses on mechanical and acceleration signals, detecting effective information by installing sensors at various locations on the body. Traditional discrete pressure acquisition devices require separate pressure sensors installed on various parts of the body, making them cumbersome to use and prone to detachment. They also require a separate acquisition processor, making them difficult to port and subject to varying installation locations, hindering subsequent functional implementation. Currently available integrated pressure detection kits lack skin adhesion. Furthermore, because they rely solely on the inertial force generated by limb movement, they lack resolution for small movements and initial limb posture at rest. Furthermore, these pressure detection kits are non-detachable, requiring the entire device to be recovered if one pressure sensor fails. Furthermore, these pressure detection kits are only applicable to a single location, making them inflexible and impractical. Summary of the Invention

[0004] The utility model solves the problem that traditional discrete pressure acquisition equipment requires independent installation of pressure sensors on various parts of the body, which is not portable and complicated to install; and solves the problem that integrated pressure detection kits can only detect one part, cannot replace sensors, and cannot detect small movements and the posture of limbs when they are initially still.

[0005] The utility model provides a wearable motion data acquisition device for motion intention recognition, which includes: an external material layer, a granular pressure sensor array, a memory alloy pressure component, an inertial material component, a flexible skin contact layer, and a power supply and signal acquisition component provided on the external material layer;

[0006] The outer material layer is an overall strip-shaped structure that can be bent arbitrarily. A groove along the axis is provided on the inner side of the strip-shaped structure. An inertial material component, a memory alloy pressure component, and a granular pressure sensor array are stacked in sequence at the bottom of the groove. The inertial material component is strip-shaped and laid throughout the groove of the outer material layer; the memory alloy pressure component is sheet-shaped and arranged at the bottom of each granular pressure sensor; the granular pressure sensors are arranged in sequence along the axial direction in the groove of the outer material layer, and then a flexible contact layer is covered on the granular pressure sensor array; the data line of the granular pressure sensor array extends from the outside of the outer material layer, and a power supply and signal acquisition component is provided at the extended position; connecting components are provided at both ends of the outer material layer. When the wearable motion data acquisition device is bent inward and worn next to the joint position of the human body, the wearable and secure device is achieved through the combination of the connecting components.

[0007] The specific material of the memory alloy pressure component is TINI.

[0008] Furthermore, the connecting components arranged at both ends of the external material layer are Velcro, one end of the external material layer extends a section of Velcro outward, and Velcro is arranged on the back of the other end.

[0009] Furthermore, 30 granular pressure sensors are fixed to the bottom of the groove on the inner side of the outer material layer by screws or elastic locks, and the granular pressure sensors are distributed in a linear array along the outer material layer.

[0010] Furthermore, the power supply and signal acquisition component is also provided with a data storage device and a wired or wireless transmission transpose.

[0011] This application has the following beneficial effects:

[0012] The discrete pressure sensors are integrated into a whole, which improves the portability of the device. Based on the integrated sensors, the inner layer material is made of flexible material to fit the skin. At the same time, the sensor is made detachable, which is more flexible and practical. The device is designed as a chain belt mode, which can be worn on different limbs and has a wide range of applications. The pressure sensor particles are initially pressed using memory alloy to ensure that each pressure sensor in this identification device has sufficient contact and initial pressure with the limb skin, which is conducive to filtering out random noise interference. The device is weighted so that the device itself has a certain gravity, which can detect small movement intentions based on the inertial force during movement, breaking the limitation that pressure signals cannot be used to detect small movements. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 It is a physical schematic diagram of the equipment provided by the utility model;

[0014] Figure 2 This is a schematic diagram of the interior of the outer material of the device provided by the utility model;

[0015] Figure 3 This is a schematic diagram of the internal structure of the pressure measuring unit of the utility model;

[0016] Figure 4 This is a schematic diagram of wearing the device provided by the present invention;

[0017] 1. External material layer; 2. Granular pressure sensor array; 3. Flexible skin contact layer; 4. Power supply and signal acquisition and processing components; 5. Velcro; 6. Inertial material component; 7. Memory alloy pressure component; 100. Wearable motion intention recognition device; 200. Zero position; DETAILED DESCRIPTION

[0018] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following description is given in conjunction with the embodiments and the accompanying drawings.

[0019] See also Figure 1-4 The utility model provides a wearable motion data acquisition device for motion intention recognition, including an external material layer 1, a granular pressure sensor array 2, a power supply and signal acquisition processing component 4, a flexible skin contact layer 3, a Velcro 5, an inertial material component 6, and a memory alloy pressure component 7.

[0020] As a further description of the above technical solution: The wristband body comprises an outer material layer 1 and a flexible contact layer 3. The two layers are 60 cm long, making them easy to wear on the thigh, upper arm, and other areas. The flexible contact layer 3 is made of cotton fabric, which enhances the wearer's experience and allows the granular pressure sensor 2 located between the outer material layer 1 and the flexible contact layer 3 to more directly sense the force during exercise, further preventing information collection failure. The two layers are then fixed together with Velcro 5, connecting them to the other end. The outer corners of the outer material layer 1 are rounded, effectively improving wearing comfort.

[0021] As a further description of the above technical solution: the inner side of the outer material layer 1 has a circle of grooves, and 30 granular pressure sensors 2 are fixed to the upper part of the grooves by screws or elastic locks, so that a pressure sensor is distributed every 2 cm on the chain strap. The granular pressure sensors 2 are distributed in a linear array along the outer material layer (60 cm), so that when worn, they can detect forces acting on the limbs in all directions.

[0022] As a further description of the above technical solution: the granular pressure sensor 2 installed on the groove is detachable, which is convenient for replacement after the sensor is damaged, greatly enhancing practicality;

[0023] As a further description of the above technical solution: the bottom of each pressure sensor 2 inside the outer material layer 1 contacts the skin of the measured part through the flexible skin contact layer 3. The back of the pressure sensor 2 is configured with a memory alloy pressure component 7 and an inertial material component 6, so that the pressure sensor maintains a certain initial pressure on the measured limb in the initial state. The device itself has a certain weight, which facilitates the use of inertial force to detect small movement trends;

[0024] As a further description of the above technical solution: the outer side of the outer material layer 1 is connected to the power supply and signal acquisition component 4 and the Velcro 5, and a wire outlet and a wire inlet are provided at the center of the connection portion between the concave outer ring and the power supply and signal acquisition component 4, and the outer end of the wire outlet of the outer material layer 1 coincides with the outer end of the wire inlet of the power supply and signal acquisition component 4;

[0025] As a further description of the above technical solution: the length of the Velcro 5 can be changed manually, and can be tightened or loosened. After wearing, the Velcro 5 can be used to adjust the chain to the best position to prevent it from falling.

[0026] As a further description of the above technical solution: the data processed by the power supply and signal acquisition processing component 4 can be stored or transmitted to an external device via wired / wireless means; the current limb state is sensed and recorded by the returned pressure data, and the posture of the tested limb when it is initially at rest can be determined;

[0027] As a further description of the above technical solution: when wearing this device, the first pressure sensor in the chain belt needs to be placed in the direction of rotation of the limb joint, and this position is set to zero position 200 to facilitate subsequent data processing;

[0028] The specific working steps of the device are as follows:

[0029] Step 1: The experimenter wears the motion intention recognition device 100 in the specified way. The specified way is that the first pressure sensor is located in the direction of the current limb joint rotation. Set the position to zero 200, that is, the contact position of the first pressure sensor with the limb is zero 200 as shown in the figure Figure 3 As shown;

[0030] Step 2: Use the motion intention recognition device 100 to collect and process the mechanical signals of human motion to obtain the original data of the mechanical signals;

[0031] Step 3: Use a filter to process the interference in the original data and extract the pressure change data from the original signal; use autocorrelation analysis to extract the effective period of the pressure change data, thereby removing duplicate data and obtaining a valid pressure change data set;

[0032] Step 4: Extract features from the dataset and divide the preprocessed dataset into training samples and test samples according to a preset ratio;

[0033] Step 5: Input the obtained data features into the neural network of the adaptive enhancement algorithm to train a motion intention recognition model;

[0034] Step 6: Use the trained motion intention recognition model to identify the newly collected pressure sensor data.

Claims

1. A wearable motion data acquisition device for motion intention recognition, characterized by: The device includes: an external material layer, a granular pressure sensor array, a memory alloy pressure component, an inertial material component, a flexible skin contact layer, and a power supply and signal acquisition component provided on the external material layer; The outer material layer is an overall strip-shaped structure that can be bent arbitrarily, and a groove along the axis is provided on the inner side of the strip-shaped structure. The bottom of the groove is stacked with an inertial material component, a memory alloy pressure component, and a granular pressure sensor array in sequence. The inertial material component is strip-shaped and is laid in the groove of the outer material layer; the memory alloy pressure component is sheet-shaped and is arranged at the bottom of each granular pressure sensor; the granular pressure sensors are arranged in the groove of the outer material layer in sequence along the axial direction, and then a flexible contact layer is covered on the granular pressure sensor array; the data line of the granular pressure sensor array extends from the outside of the outer material layer, and a power supply and signal acquisition component is provided at the extended position; connecting components are provided at both ends of the outer material layer. When the wearable motion data acquisition device is bent inward and worn next to the joint position of the human body, the wearing and tightening are achieved through the combination of the connecting components. The specific material of the memory alloy pressure component is TINI.

2. A wearable motion data acquisition device for motion intention recognition according to claim 1, characterized in that: The connecting components arranged at both ends of the external material layer are Velcro. One end of the external material layer extends a section of Velcro outwards, and the back of the other end is provided with Velcro.

3. A wearable motion data acquisition device for motion intention recognition according to claim 1, characterized in that: 30 granular pressure sensors are fixed at the bottom of the groove on the inner side of the outer material layer by screws or elastic locks, and the granular pressure sensors are distributed in a linear array along the outer material layer.

4. A wearable motion data acquisition device for motion intention recognition according to claim 1, characterized in that: The power supply and signal acquisition components are also provided with data storage devices and wired or wireless transmission transposition.

Images