CN114028775B - Ankle joint movement intention identification method and system based on sole pressure - Google Patents
Ankle joint movement intention identification method and system based on sole pressure Download PDFInfo
- Publication number
- CN114028775B CN114028775B CN202111489473.9A CN202111489473A CN114028775B CN 114028775 B CN114028775 B CN 114028775B CN 202111489473 A CN202111489473 A CN 202111489473A CN 114028775 B CN114028775 B CN 114028775B
- Authority
- CN
- China
- Prior art keywords
- pressure
- data
- sole
- plantar
- upper computer
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B23/00—Exercising apparatus specially adapted for particular parts of the body
- A63B23/035—Exercising apparatus specially adapted for particular parts of the body for limbs, i.e. upper or lower limbs, e.g. simultaneously
- A63B23/04—Exercising apparatus specially adapted for particular parts of the body for limbs, i.e. upper or lower limbs, e.g. simultaneously for lower limbs
- A63B23/08—Exercising apparatus specially adapted for particular parts of the body for limbs, i.e. upper or lower limbs, e.g. simultaneously for lower limbs for ankle joints
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/1036—Measuring load distribution, e.g. podologic studies
- A61B5/1038—Measuring plantar pressure during gait
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
- A61B5/6802—Sensor mounted on worn items
- A61B5/6804—Garments; Clothes
- A61B5/6807—Footwear
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
- A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
- A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- A—HUMAN NECESSITIES
- A63—SPORTS; GAMES; AMUSEMENTS
- A63B—APPARATUS FOR PHYSICAL TRAINING, GYMNASTICS, SWIMMING, CLIMBING, OR FENCING; BALL GAMES; TRAINING EQUIPMENT
- A63B71/00—Games or sports accessories not covered in groups A63B1/00 - A63B69/00
- A63B71/06—Indicating or scoring devices for games or players, or for other sports activities
- A63B71/0619—Displays, user interfaces and indicating devices, specially adapted for sport equipment, e.g. display mounted on treadmills
- A63B2071/0658—Position or arrangement of display
- A63B2071/0661—Position or arrangement of display arranged on the user
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Orthopedic Medicine & Surgery (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Molecular Biology (AREA)
- Software Systems (AREA)
- Pathology (AREA)
- Theoretical Computer Science (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Biophysics (AREA)
- Physical Education & Sports Medicine (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Dentistry (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention relates to an ankle joint movement intention identification method and system based on sole pressure. The method comprises the steps of firstly, obtaining the pressure of each part of a sole by using a sole pressure sensor, then realizing the communication function of an upper computer and a lower computer through a USB (universal serial bus), compiling an upper computer pressure acquisition program to realize the data transceiving function of the lower computer, and after obtaining the sole pressure of a patient, summarizing a data processing process capable of generally obtaining a good recognition rate through experiment comparison. The ankle movement intention of the patient can be identified through the sole pressure, and a good identification rate can be generally obtained.
Description
Technical Field
The invention relates to an ankle joint movement intention identification method and system based on sole pressure.
Background
The human ankle joint is one of the most complex joints in the skeleton and plays an important role in keeping the body balance, but once damaged, the ankle joint greatly limits the mobility of people. The patient can express his own motor intention only through weak motor ability in the early stage of rehabilitation, but currently, a related motor intention identification method is rarely available. Therefore, the ankle joint movement intention identification method based on the sole pressure is provided, the sole pressure sensors are used for obtaining the pressure of each part of the sole, the upper computer and the lower computer are communicated through compiling the upper computer acquisition program to achieve the purpose of data receiving and sending functions, and after the sole pressure of a patient is obtained, a data processing process capable of generally obtaining a good identification rate is summarized through experimental comparison.
Disclosure of Invention
The invention aims to provide a method and a system for identifying ankle joint movement intention based on plantar pressure, which can identify the ankle movement intention of a patient through plantar pressure and can generally obtain a good identification rate.
In order to achieve the purpose, the technical scheme of the invention is as follows: the utility model provides an ankle joint movement intention recognition method based on plantar pressure, at first uses plantar pressure sensor to acquire the pressure size of each part of sole, and the communication function of host computer is realized to rethread USB to write host computer pressure acquisition program and realize the data transceiver function to the host computer so that the host computer acquires plantar pressure, and then the host computer handles plantar pressure data, realizes ankle joint movement intention discernment based on plantar pressure.
In an embodiment of the present invention, the plantar pressure sensor is an RX-ES42-16P piezoresistive flexible plantar pressure sensor, a sensing array of the plantar pressure sensor has 16 pressure sensitive points in total, an array design of 6 rows by 4 columns is adopted, and a plantar pressure value F suffered by an actual pressure sensitive point can be obtained through the following conversion i ：
Wherein, V i The measured voltage value of the ith pressure sensitive point under the condition of applying pressure; b i The measured voltage value of the ith pressure sensitive point under the condition of not applying pressure; g is the acceleration of gravity; k is a radical of i And obtaining the slope of the ith pressure sensitive point by linearly fitting the voltage value and the pressure value.
In one embodiment of the present invention, in order to solve the cross-coupling phenomenon occurring in the sensor array, a method of gating a row-column path is used, the cross-coupling phenomenon is solved by using the "virtual short" principle of the operational amplifier, and when the resistance value of a certain pressure sensitive point needs to be measured, a corresponding row-column gating switch is turned on.
In an embodiment of the present invention, the upper computer pressure acquisition program specifically includes: setting a format of 'start code + command code + end code' for the message sent by the upper computer, and setting a format of 'start code + command code + data + end code' for the message sent by the lower computer; the obtained plantar pressure data and the encoder data are sent in an ASCII code format, the length of single data is 8 bytes, a command code is 1 byte, uniqueness of a start code and an end code is guaranteed, and a receiving party can conveniently identify a complete data packet.
In an embodiment of the present invention, the specific steps of the upper computer sending the acquisition command to the lower computer and receiving the plantar pressure data are as follows:
after the lower computer is connected with the upper computer through a USB, data transmission is realized through a serial port communication mode, corresponding exercise intentions are selected, 5s single plantar pressure collection is carried out, and after all pressure data collection is finished, data needing to be output are selected and output in a form of a table; the pressure image on the right side of the interface in the acquisition process can display the pressure of each pressure sensitive point in real time, and the color gradient change can be adjusted according to the needs of a patient; the sole pressure sensor is arranged between the sole and the vamp in the collecting process.
In one embodiment of the invention, the voltage indication range for indicating the pressure is 0-3.3V, and the color scheme is determined by the pressure indication range of 0-3.3V according to the color gradient change sequence:
when the pressure index is in the range of 0-0.2V, the color matching is a gradual change from blue RGB (0, 255) to light blue RGB (0, 255, 255);
when the pressure index is in the range of 0.2-0.4V, the color matching is a gradual change from light blue RGB (0, 255, 255) to green RGB (0, 255, 0);
when the pressure index is in the range of 0.4-0.6V, the color matching is a gradual change from green RGB (0, 255, 0) to red RGB (255, 0);
when the pressure reading exceeds 0.6V, the color remains red RGB (255, 0);
the plantar pressure color image is used for judging the movement intention of the testee in a sitting posture state.
In one embodiment of the present invention, the upper computer processes sole pressure data, and the mode for realizing ankle joint movement intention recognition based on sole pressure is as follows:
after the plantar pressure data under different movement intentions are collected, the redundant samples are transversely compressed, and a standardization processing method is used for enabling a certain original value in a certain characteristic vector to be x i Normalized the data is y i Then the normalization formula is as follows:
wherein:
the feature number of the sample data is the number of pressure sensitive points of the plantar pressure sensor, namely the sample data has 16-dimensional features, so that the principal component analysis method is used for dimensionality reduction, then a support vector machine is used for learning and modeling, a radial basis kernel function is used, cross validation aiming at a test set and grid search algorithm optimization are carried out on parameters, and then ankle joint movement intention recognition based on plantar pressure is achieved.
The invention also provides an ankle joint movement intention recognition system based on the sole pressure, which comprises a sole pressure sensor unit for acquiring the pressure of each part of the sole, a power supply unit for supplying power to the whole system, a signal acquisition unit for acquiring and selecting the sole pressure of the corresponding pressure sensitive point of the sole pressure sensor unit and processing each sole pressure signal, and an upper computer unit for processing sole pressure data to realize the ankle joint movement intention based on the sole pressure.
Compared with the prior art, the invention has the following beneficial effects: the ankle movement intention of the patient can be recognized through the sole pressure, and a good recognition rate can be generally obtained; the invention fills the blank of the technology combining plantar pressure and intention recognition, and provides a technical basis for the development of the subsequent active rehabilitation function of the ankle joint rehabilitation robot.
Drawings
Fig. 1 is an overall structure of the present invention.
Fig. 2 is a schematic diagram of a plantar pressure sensor according to an embodiment of the present invention.
Fig. 3 is an equivalent circuit diagram of a plantar pressure sensor (right foot) according to an embodiment of the invention.
FIG. 4 is a schematic diagram of circuit gating according to an embodiment of the present invention.
FIG. 5 is a power circuit layout of an embodiment of the present invention.
FIG. 6 is a circuit diagram of an overall signal acquisition circuit according to an embodiment of the present invention.
Fig. 7 is a schematic diagram of signal acquisition of a single sensing unit according to an embodiment of the present invention.
FIG. 8 is a schematic diagram of the communication between the upper and lower computers according to an embodiment of the present invention.
FIG. 9 is a diagram of a host computer acquisition interface in accordance with an embodiment of the present invention.
FIG. 10 is a flow chart of an exemplary acquisition process according to the present invention.
Fig. 11 is a block diagram of a plantar pressure data collection module according to an embodiment of the present invention.
FIG. 12 is an image of pressure distribution for different exercise intentions according to an embodiment of the present invention.
Detailed Description
The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.
The invention relates to an ankle joint movement intention identification method based on plantar pressure.
The invention also provides an ankle joint movement intention recognition system based on the sole pressure, which comprises a sole pressure sensor unit for acquiring the pressure of each part of the sole, a power supply unit for supplying power to the whole system, a signal acquisition unit for acquiring and selecting the sole pressure of the corresponding pressure sensitive point of the sole pressure sensor unit and processing each sole pressure signal, and an upper computer unit for processing sole pressure data to realize the ankle joint movement intention based on the sole pressure.
The following is a specific implementation process of the present invention.
The invention mainly comprises four parts: the system comprises a hardware design part, an acquisition interface design part, a plantar pressure acquisition part and a data processing part, wherein the overall structure diagram is shown in figure 1, the hardware design part mainly explains the aspects of hardware acquisition and circuit design, the acquisition interface design part mainly explains the functions of an acquisition interface and a software control flow, the plantar pressure acquisition part mainly describes the overall acquisition flow and the acquisition effect aiming at the acquisition interface, and the data processing part is developed around the part which processes pressure data after plantar pressure acquisition. The hardware design and the acquisition interface design are the basis for realizing the method, and the data processing is the key for reflecting which method can achieve better recognition rate.
Regarding the aspect of the pressure acquisition sensor, the RX-ES42-16P piezoresistive flexible plantar pressure sensor is used herein, the physical diagram is shown in fig. 2, the sensing array has 16 pressure sensitive points, the array design of 6 rows by 4 columns is adopted, and the equivalent circuit diagram of the right plantar pressure sensor is shown in fig. 3. The pressure value F suffered by the actual sensitive unit can be obtained by the following formula conversion i ：
Wherein, V i The voltage value measured under the condition of applying pressure for the ith pressure sensitive point; b i The measured voltage value of the ith pressure sensitive point under the condition of not applying pressure; g is the acceleration of gravity; k is a radical of i Linearly fitting the voltage value and the pressure for the ith pressure sensitive pointThe resulting slope is evaluated.
In order to solve the cross coupling phenomenon in the acquisition array, the method for gating the row-column path is adopted, the cross coupling phenomenon can be effectively solved by utilizing the 'virtual short' principle of the operational amplifier, when the resistance value of a certain pressure sensing unit needs to be measured, only a corresponding row-column gating switch needs to be opened, and the principle is shown in fig. 4.
Power supply circuits are an important component of hardware circuits. For the power supply requirement of the data acquisition module, a power circuit design diagram is shown in fig. 5, and 5V input voltage is converted into 3.3V through a linear voltage stabilization chip (LM 1117-3.3). In order to reduce the output voltage ripple, capacitors (C1, C3 and C2, C4) with 10uF and 0.1uF are respectively connected in parallel at two ends of the voltage stabilizing chip. After the 3.3V voltage is obtained, two paths of voltage output ends which are not interfered with each other are separated by utilizing the characteristic that the 1R resistor has weak inductance. Similarly, bypass capacitors (C5, C6 and C7, C8) with 1uF and 0.1uF are connected in parallel at each voltage output end for filtering power supply noise. Then, one of the two outputs is used for supplying power to a signal conditioning circuit and the like; the other path is converted into-3.3V by a voltage inverting chip (TPS 60403) to be used by the negative input end of the operational amplifier, and the chip and 3 1uF capacitors (C9, C10 and C11 in the figure) form a DC-DC charge pump inverter to eliminate magnetic field and electromagnetic interference.
The signal acquisition circuit comprises a row and column scanning circuit, a signal inverting amplification circuit and a low-pass filter circuit, and gates the sensitive unit access circuit of a certain row and a certain column of the sensor according to the control signal to acquire signals. The schematic diagram of the signal acquisition circuit is shown in fig. 6. P1-P6 are row line signals of the sensor, P7-P10 are column line signals of the sensor, taking the collection of sensitive units of a first row and a second row as an example, a multi-channel analog switch U1 is communicated with an Out/Int pin and an Int/Out0 under the control of a control signal, a multi-channel analog switch U2 is communicated with an Int/Out1 pin under the control of a control signal, the signals are reversely amplified by an operational amplifier U3B, pass through the multi-channel analog switch U2 and are reversely amplified by a U4, then are accessed into a low-pass filter circuit consisting of R13 and C12, and finally are accessed into an analog-to-digital conversion channel to finish data collection.
In the signal acquisition circuit, the acquisition principle of the signal of a single sensitive unit is shown in fig. 7.
Obtaining output voltage V after two-stage reverse proportional amplification out ，V out The data are collected by an analog-to-digital conversion channel of the main control chip after passing through the filter circuit. Known reference voltage V ref And a resistor R 1 、R 2 、R 3 Resistance value of, sensitive unit resistance R ij And V out The relationship of (1) is:
in view of the fact that the flexible sensor used in the invention is only 16 sensitive units, the data volume is small, and the requirement on real-time performance is not high, an RC low-pass filter circuit is introduced to eliminate partial noise. The RC filter circuit is a commonly used and effective hardware filter circuit, and the cutoff frequency is:
f＝1/(2πRC)
the output voltage filtered by the RC circuit can approach the actual voltage after 3-5 time constants, so that the acquisition interval time of two adjacent sensitive units can meet the following requirements in order not to influence the acquisition precision of data:
t＞＝(3～5)τ
wherein:
τ＝RC
in order to ensure that the upper computer and the lower computer can correctly identify the data or the command transmitted from the other side, the format of the message sent by the upper computer is set as 'start code + command code + end code', and the format of the message sent by the lower computer is set as 'start code + command code + data + end code'. The obtained plantar pressure data and the encoder data are sent in an ASCII code format, and the length of single data is 8 bytes. The command code is 1 byte, and may be a hexadecimal number of one byte except 0x3C, 0x3E, 0x2E, 0x 30-0 x 39. The uniqueness of the start code and the end code is ensured, and a complete data packet is conveniently identified by a receiving party. The communication schematic diagram of the upper computer and the lower computer is shown in figure 8.
In order to send acquisition commands to the lower computer and receive relevant plantar pressure data, an acquisition interface of the upper computer is designed as shown in fig. 9. The specific operation steps are as follows: after the acquisition module is connected with an upper computer through a USB, data transmission can be achieved through a serial port communication mode, corresponding exercise intentions are selected, 5s single plantar pressure acquisition is carried out, and after all pressure data acquisition is completed, data needing to be output in the selection can be output in a form of a table. The pressure image on the right side of the interface in the acquisition process can display the pressure of each sensitive unit in real time, and the color gradient change can be adjusted according to the needs of a patient, and the whole acquisition process is shown in fig. 10.
After the USB data line is used for connecting the data acquisition module, the program is opened and the data acquisition module is connected, so that the plantar pressure data under different exercise intentions can be acquired. Accordingly, the experimental device for collecting plantar pressure data is shown in fig. 11. Before data acquisition, a sandal with a binding band is worn on the right foot of a subject, the sole pressure acquisition array is placed between the sole and the vamp, and the binding band is tightened to ensure that relative sliding among the sandal, the sensor and the ankle does not occur. The subject is in a sitting posture, and pressure acquisition is carried out according to the direction of an arrow on the interface. Data acquisition was intended for each exercise for a duration of 5s after stabilization of plantar pressure. In 5s of single acquisition, the data volume is about 100, and due to the low frequency of human motion, a large number of repeated or similar samples exist in sample data, so that the redundant samples are transversely compressed. And averaging every continuous 5 sample data in the single experimental data to serve as a compressed new sample. In a group of sole data acquisition processes, the testee is required to complete data acquisition of 9 motion directions.
In order to visually display the magnitude of the plantar pressure value, the voltage value of the pressure is represented by color. Referring to the color gradient change commonly used for temperature field profiles, the value of zero is represented in blue and the value of peak is represented in red, and the color gradient change order is blue-light blue-green-red. The voltage index range for expressing the pressure is 0-3.3V, and the color scheme is determined according to the color gradient change sequence by the pressure index range of 0-3.3V:
when the pressure readings are in the range of 0-0.2V, the color match is a gradual change from blue RGB (0, 255) to light blue RGB (0, 255, 255).
When the pressure indication is in the range of 0.2-0.4V, the color scheme is a gradual transition from light blue RGB (0, 255, 255) to green RGB (0, 255, 0).
When the pressure index is in the range of 0.4-0.6V, the color match is a gradual change from green RGB (0, 255, 0) to red RGB (255, 0).
When the pressure reading exceeds 0.6V, the color remains red RGB (255, 0).
The plantar pressure color image is mainly used for judging the athletic intention of a subject in a sitting posture, and fig. 12 shows a pressure distribution image of a certain subject under the athletic intention aiming at dorsiflexion and inward and outward turning of toes.
After the relevant plantar pressure data are collected, in order to eliminate the dimensional influence among the data features and accelerate the convergence speed of the learning algorithm, the feature scaling method is often used for preprocessing sample data i Normalized to y i Then the normalization formula is as follows:
wherein:
in addition, data with too high dimensionality not only increases the amount of computation, but may also affect the screening of useful information. In the invention, the characteristic number of the sample data is the number of the sensitive units of the plantar pressure sensor, namely the sample data has 16-dimensional characteristics. Therefore, the method uses the principal component analysis method to reduce the dimension, and the comparison of the early test shows that the sample data has higher identification rate after being reduced to 6 dimensions.
The reasonable algorithm is the core of the whole identification system, the invention adopts a machine learning algorithm of a support vector machine, and compares and optimizes the selection of the kernel function and the related parameters, in the aspect of kernel function parameter selection, the parameter search ranges of c and g are both [0,100], and 0.2 is the grid search step length. Finally, the conclusion is drawn: the parameters obtained by selecting the radial basis kernel function and performing the parameter optimization algorithm aiming at the test set have the highest identification accuracy.
In the experiment, the single person collects a plurality of groups of data, so that the time is consumed, and the collection group number is reduced on the premise of ensuring certain identification accuracy, and the test subject can be helped to carry out active rehabilitation exercise training as soon as possible. Therefore, the invention can draw the conclusion through experimental comparison: when the number of the training set groups is 6, the method generally has high recognition accuracy, and can be considered as follows: after 6 times of data acquisition and training, the ankle joint movement intention can be recognized through sole pressure.
From the above experiments it can be concluded that: when at least 6 groups of data of a subject are collected and standardized preprocessing method is carried out on the data, the data are reduced to 6 dimensions according to a principal component analysis method, a radial basis kernel function is adopted after learning modeling is carried out through a support vector machine, cross validation aiming at a test set and optimization of a grid search algorithm are carried out on parameters, and then higher identification accuracy can be generally obtained.
According to the invention, the ankle joint movement intention of the patient is recognized by acquiring the plantar pressure of the patient, the acquisition module is small and portable, the acquired data is high in accuracy, the pressure of each region can be seen in real time through the upper computer acquisition program, the processing method of the conclusion obtained by adopting the subsequent experiment generally has good recognition accuracy, the whole work from modeling to recognition can be completed in a short time, the time is saved for the patient to carry out rehabilitation training, and the rehabilitation training effect is improved. The ankle joint movement intention recognition method is a novel ankle joint movement intention recognition method, and provides a technical basis for realizing the active rehabilitation function of the ankle rehabilitation robot subsequently.
The use process of the invention is as follows: after completing the wiring of the plantar pressure data acquisition module shown in fig. 12, connecting a USB cable with an upper computer, opening an acquisition interface of the upper computer, selecting a real pressure image after completing serial port connection at the upper left corner to monitor plantar pressure change in real time, clicking a single acquisition 5s button after selecting a corresponding movement intention, clicking the next movement intention after completing acquisition, clicking the single acquisition 5s again, sequentially circulating, completing data to be derived in all intention acquisition post-selection, clicking a lead-out to excel button to derive relevant pressure data, then opening matlab software, processing and modeling at least 6 groups of data by the experimental summary method, and inputting pressure data to be subjected to intention identification.
The above are preferred embodiments of the present invention, and all changes made according to the technical solutions of the present invention that produce functional effects do not exceed the scope of the technical solutions of the present invention belong to the protection scope of the present invention.
Claims (4)
1. The ankle joint movement intention recognition method based on the sole pressure is characterized in that a sole pressure sensor is used for obtaining the pressure of each part of a sole, the communication function of an upper computer and a lower computer is realized through a USB (universal serial bus), an upper computer pressure acquisition program is compiled to realize the data transceiving function of the lower computer so that the upper computer obtains the sole pressure, and then the upper computer processes sole pressure data to realize the ankle joint movement intention recognition based on the sole pressure;
the plantar pressure sensor is an RX-ES42-16P piezoresistive flexible plantar pressure sensor, a sensing array of the plantar pressure sensor has 16 pressure sensitive points in total, an array design of 6 rows by 4 columns is adopted, and the plantar pressure value F of an actual pressure sensitive point is obtained through the following conversion i ：
Wherein, V i The voltage value measured under the condition of applying pressure for the ith pressure sensitive point; b i Is the ith pressure sensitivityThe voltage value measured under the condition of not applying pressure is counted; g is the acceleration of gravity; k is a radical of i Obtaining a slope for the ith pressure sensitive point by linearly fitting the voltage value and the pressure value;
the upper computer pressure acquisition program specifically comprises the following steps: setting a format of 'start code + command code + end code' for the message sent by the upper computer, and setting a format of 'start code + command code + data + end code' for the message sent by the lower computer; the obtained plantar pressure data and the encoder data are sent in an ASCII code format by respectively taking 0x3C and 0x3E as a starting zone bit and an ending zone bit of a data frame, the length of single data is 8 bytes, a command code is 1 byte, the uniqueness of a starting code and an ending code is ensured, and a receiving party can conveniently identify a complete data packet;
the specific steps of the upper computer sending an acquisition command to the lower computer and receiving plantar pressure data are as follows:
after the lower computer is connected with the upper computer through a USB, data transmission is realized through a serial port communication mode, corresponding exercise intentions are selected, 5s single plantar pressure collection is carried out, and after all pressure data collection is finished, data needing to be output are selected and output in a form of a table; the pressure image on the right side of the interface displays the pressure of each pressure sensitive point in real time in the acquisition process, or the color gradient change is adjusted according to the requirement of a patient; the sole pressure sensor is arranged between the sole and the vamp in the collecting process;
the host computer handles plantar pressure data, realizes the mode of ankle joint motion intention discernment based on plantar pressure and does:
after the plantar pressure data under different movement intentions are collected, the redundant samples are transversely compressed, and a standardization processing method is used for enabling a certain original value in a certain characteristic vector to be x i Normalized the data is y i Then the normalization formula is as follows:
wherein:
the feature number of the sample data is the number of pressure sensitive points of the plantar pressure sensor, namely the sample data has 16-dimensional features, so that the principal component analysis method is used for dimensionality reduction, then a support vector machine is used for learning and modeling, a radial basis kernel function is used, cross validation aiming at a test set and grid search algorithm optimization are carried out on parameters, and then ankle joint movement intention recognition based on plantar pressure is achieved.
2. The ankle joint movement intention identification method based on plantar pressure as claimed in claim 1, characterized in that, in order to solve the cross coupling phenomenon occurring in the sensing array, a method of gating the row-column path is used, the cross coupling phenomenon is solved by using the 'virtual short' principle of the operational amplifier, and when the resistance value of a certain pressure sensitive point needs to be measured, the corresponding row-column gating switch is turned on.
3. The ankle joint movement intention recognition method based on plantar pressure according to claim 1, wherein the voltage indications representing the pressure range from 0 to 3.3V, and the color scheme is determined by the pressure indications ranging from 0 to 3.3V in the order of color gradient change:
when the pressure index is in the range of 0-0.2V, the color matching is gradual change from blue RGB (0, 255) to light blue RGB (0, 255, 255);
when the pressure index is in the range of 0.2-0.4V, the color matching is a gradual change from light blue RGB (0, 255, 255) to green RGB (0, 255, 0);
when the pressure indication is in the range of 0.4-0.6V, the color matching is a gradual change from green RGB (0, 255, 0) to red RGB (255, 0);
when the pressure reading exceeds 0.6V, the color remains red RGB (255, 0);
the plantar pressure color image is used for judging the movement intention of the testee in a sitting posture state.
4. An ankle joint movement intention recognition system based on sole pressure is characterized by comprising a sole pressure sensor unit, a power supply unit, a signal acquisition unit and an upper computer unit, wherein the sole pressure sensor unit is used for acquiring the pressure of each part of a sole, the power supply unit is used for supplying power to the whole system, the signal acquisition unit is used for acquiring and selecting the sole pressure of a corresponding pressure sensitive point of the sole pressure sensor unit and processing each sole pressure signal, and the upper computer unit is used for processing sole pressure data so as to realize ankle joint movement intention recognition based on sole pressure;
the sole pressure sensor unit adopts an RX-ES42-16P piezoresistive flexible sole pressure sensor, the sensing array of the sole pressure sensor has 16 pressure sensitive points, the array design of 6 rows by 4 columns is adopted, and the sole pressure value F of the actual pressure sensitive point is obtained by the following conversion i ：
Wherein, V i The voltage value measured under the condition of applying pressure for the ith pressure sensitive point; b i The voltage value measured under the condition that no pressure is applied to the ith pressure sensitive point; g is gravity acceleration; k is a radical of i Obtaining a slope for the ith pressure sensitive point by linearly fitting the voltage value and the pressure value;
the host computer unit is stored with a host computer pressure acquisition program, and the host computer pressure acquisition program specifically comprises: setting a format of 'start code + command code + end code' for the message sent by the upper computer unit, and setting a format of 'start code + command code + data + end code' for the message sent by the lower computer unit, namely the plantar pressure sensor unit + signal acquisition unit; the method comprises the steps that 0x3C and 0x3E are respectively used as a start zone bit and an end zone bit of a data frame, obtained plantar pressure data and encoder data are sent in an ASCII code format, the length of single data is 8 bytes, a command code is 1 byte, uniqueness of a start code and an end code is guaranteed, and a receiving party can conveniently identify a complete data packet;
the specific steps of sending an acquisition command to the lower computer unit by the upper computer unit and receiving plantar pressure data by the lower computer unit are as follows:
after the lower computer unit is connected with the upper computer unit through a USB, data transmission is realized through a serial port communication mode, corresponding exercise intentions are selected, 5s single plantar pressure collection is carried out, and after all pressure data collection is finished, data needing to be output are selected and output in a form of a table; the pressure image on the right side of the interface in the acquisition process displays the pressure of each pressure sensitive point in real time, or the color gradient change is adjusted according to the requirement of a patient; the sole pressure sensor is arranged between the sole and the vamp in the collecting process;
the upper computer unit processes sole pressure data, and the mode of realizing ankle joint movement intention recognition based on sole pressure is as follows:
after the sole pressure data under different movement intentions are collected, the redundant samples are transversely compressed, and a standardization processing method is used for enabling a certain original value in a certain characteristic vector to be x i Normalized to y i Then the normalization formula is as follows:
wherein:
as the characteristic number of the sample data is the number of the pressure sensitive points of the plantar pressure sensor, namely the sample data has 16-dimensional characteristics, the principal component analysis method is used for dimensionality reduction, then a support vector machine is used for learning and modeling, a radial basis kernel function is used, cross validation aiming at a test set and grid search algorithm optimization are carried out on parameters, and then the ankle joint movement intention identification based on the plantar pressure is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111489473.9A CN114028775B (en) | 2021-12-08 | 2021-12-08 | Ankle joint movement intention identification method and system based on sole pressure |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111489473.9A CN114028775B (en) | 2021-12-08 | 2021-12-08 | Ankle joint movement intention identification method and system based on sole pressure |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114028775A CN114028775A (en) | 2022-02-11 |
| CN114028775B true CN114028775B (en) | 2022-10-14 |
Family
ID=80146402
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111489473.9A Active CN114028775B (en) | 2021-12-08 | 2021-12-08 | Ankle joint movement intention identification method and system based on sole pressure |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114028775B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111557828A (en) * | 2020-04-29 | 2020-08-21 | 天津科技大学 | Active stroke lower limb rehabilitation robot control method based on healthy side coupling |
| CN113515967A (en) * | 2020-03-27 | 2021-10-19 | 深圳市联合视觉创新科技有限公司 | Movement intention recognition model generation method, device, equipment and storage medium |
Family Cites Families (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100635906B1 (en) * | 2005-03-24 | 2006-10-18 | 경북대학교 산학협력단 | Apparatus for Rehabilitation Training and Method of the Rehabilitation Training using the Weight Load and the Angle of Articulation for variable |
| KR20120100059A (en) * | 2011-03-02 | 2012-09-12 | 주식회사 엠엔에스에이 | A system for measuring the information of a pedestrian |
| CN102631276B (en) * | 2011-12-31 | 2015-04-15 | 中国科学院深圳先进技术研究院 | Interaction system for rehabilitation training robot |
| CN103330565A (en) * | 2013-04-16 | 2013-10-02 | 北京航空航天大学 | Human body gait recognition device based on sole pressure distribution change |
| CN104434128B (en) * | 2014-12-25 | 2016-08-24 | 中国科学院合肥物质科学研究院 | A kind of left and right based on plantar nervous arch information foot dynamic identifying method |
| CN105125216B (en) * | 2015-08-29 | 2019-01-01 | 深圳市老年医学研究所 | A kind of gait detecting system based on plantar pressure |
| KR101680740B1 (en) * | 2015-08-31 | 2016-11-30 | 한국과학기술연구원 | Recognition method of human walking speed intention from surface electromyogram signals of plantar flexor and walking speed control method of a lower-limb exoskeleton robot |
| CN105078463A (en) * | 2015-09-08 | 2015-11-25 | 常州先进制造技术研究所 | Distributed multi-channel signal acquisition system for human lower-limb motion intention recognition |
| JP6692018B2 (en) * | 2015-12-18 | 2020-05-13 | Cyberdyne株式会社 | Walking training system and walking training device |
| CN107997929B (en) * | 2017-11-29 | 2020-09-01 | 深圳睿瀚医疗科技有限公司 | Portable ankle joint rehabilitation robot based on active intention control |
| CN108992778A (en) * | 2018-08-01 | 2018-12-14 | 龚映清 | A kind of functional muscle electric stimulation system and method based on sensor intelligent insole |
| CN109940584A (en) * | 2019-03-25 | 2019-06-28 | 杭州程天科技发展有限公司 | The detection method that a kind of exoskeleton robot and its detection human motion are intended to |
| CN110141239B (en) * | 2019-05-30 | 2020-08-04 | 东北大学 | Movement intention recognition and device method for lower limb exoskeleton |
| CN110653817B (en) * | 2019-08-20 | 2022-04-15 | 南京航空航天大学 | Exoskeleton robot power-assisted control system and method based on neural network |
| CN111582108B (en) * | 2020-04-28 | 2022-09-20 | 河北工业大学 | Gait recognition and intention perception method |
| CN112603764A (en) * | 2020-12-30 | 2021-04-06 | 中国医学科学院生物医学工程研究所 | Lower limb bedside movement rehabilitation device based on coupling driving mode |
| CN112766142B (en) * | 2021-01-15 | 2022-12-20 | 天津大学 | Sole pressure image processing method, sole pressure image identification method and gait analysis system |
| CN113229801B (en) * | 2021-03-04 | 2022-07-01 | 南京邮电大学 | Insole type sole pressure measuring system and method |
| CN113273999B (en) * | 2021-05-25 | 2023-08-01 | 南开大学 | Wearable multidimensional gait analysis system and method |
| CN113627500A (en) * | 2021-07-28 | 2021-11-09 | 中国长江三峡集团有限公司 | Gait information acquisition system based on multi-sensor fusion and recognition algorithm |
| CN113730879B (en) * | 2021-10-15 | 2022-05-10 | 福州大学 | Ankle rehabilitation method based on plantar force feedback |
-
2021
- 2021-12-08 CN CN202111489473.9A patent/CN114028775B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113515967A (en) * | 2020-03-27 | 2021-10-19 | 深圳市联合视觉创新科技有限公司 | Movement intention recognition model generation method, device, equipment and storage medium |
| CN111557828A (en) * | 2020-04-29 | 2020-08-21 | 天津科技大学 | Active stroke lower limb rehabilitation robot control method based on healthy side coupling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114028775A (en) | 2022-02-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105125216B (en) | A kind of gait detecting system based on plantar pressure | |
| CN202489940U (en) | Pulse condition information collecting system | |
| CN107440694A (en) | A kind of individualized intelligent diagnosis by feeling the pulse instrument system and analysis method based on proportion measurement method | |
| CN101785659B (en) | Portable general biological signal acquisition and analysis system and method thereof | |
| CN104062421A (en) | Measuring system for comfort level of clothes | |
| CN108444436A (en) | A kind of hand gestures measuring system and method based on flexible large deformation sensor | |
| CN109765996A (en) | Insensitive gesture detection system and method are deviated to wearing position based on FMG armband | |
| CN110074779A (en) | A kind of EEG signal identification method and device | |
| CN115054243B (en) | Closed-loop design method and system for upper limb rehabilitation training system | |
| CN114028775B (en) | Ankle joint movement intention identification method and system based on sole pressure | |
| CN116909390A (en) | Multi-mode data acquisition system based on glove | |
| CN110532898A (en) | A kind of physical activity recognition methods based on smart phone Multi-sensor Fusion | |
| CN106267731A (en) | A kind of training aids | |
| CN101644952A (en) | Portable wireless man-machine interactive system based on electro-oculogram signal | |
| CN105054486A (en) | Health monitoring shoes | |
| CN116089797A (en) | Pulse condition identification method and system based on convolutional neural network | |
| CN107220584A (en) | Infrared gesture recognition system | |
| CN111481197A (en) | A living-machine multimode information acquisition fuses device for man-machine natural interaction | |
| CN204636340U (en) | A kind of pulse test meter comparing shaping circuit based on self adaptation | |
| CN209070491U (en) | A kind of pliable pressure sensing hand language recognition device | |
| CN108255303A (en) | A kind of gesture identification method based on self-control data glove | |
| CN212186496U (en) | Simulation system for real-time compression sampling reconstruction of electromyographic signals | |
| CN206147273U (en) | Multifunctional intelligent watch | |
| Chen et al. | Posenet based acupoint recognition of blind massage robot | |
| CN205862343U (en) | A kind of personal recognition micro-shape information input equipment merged for computer key Mus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |