CN106821391B - Human body gait acquisition and analysis system and method based on inertial sensor information fusion - Google Patents

Human body gait acquisition and analysis system and method based on inertial sensor information fusion Download PDF

Info

Publication number
CN106821391B
CN106821391B CN201710178883.9A CN201710178883A CN106821391B CN 106821391 B CN106821391 B CN 106821391B CN 201710178883 A CN201710178883 A CN 201710178883A CN 106821391 B CN106821391 B CN 106821391B
Authority
CN
China
Prior art keywords
information
gait
acceleration
human body
leg
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
Application number
CN201710178883.9A
Other languages
Chinese (zh)
Other versions
CN106821391A (en
Inventor
郭雅静
朱晓荣
王福德
郑继贵
黄玉平
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Research Institute of Precise Mechatronic Controls
Original Assignee
Beijing Research Institute of Precise Mechatronic Controls
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Research Institute of Precise Mechatronic Controls filed Critical Beijing Research Institute of Precise Mechatronic Controls
Priority to CN201710178883.9A priority Critical patent/CN106821391B/en
Publication of CN106821391A publication Critical patent/CN106821391A/en
Application granted granted Critical
Publication of CN106821391B publication Critical patent/CN106821391B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • 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
    • A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
    • A61B5/112—Gait analysis
    • 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
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6823—Trunk, e.g., chest, back, abdomen, hip
    • 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
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6828—Leg
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7271—Specific aspects of physiological measurement analysis
    • A61B5/7275—Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G06V40/23—Recognition of whole body movements, e.g. for sport training
    • G06V40/25—Recognition of walking or running movements, e.g. gait recognition

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Animal Behavior & Ethology (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Surgery (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physiology (AREA)
  • Artificial Intelligence (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Fuzzy Systems (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Evolutionary Computation (AREA)
  • Dentistry (AREA)
  • Mathematical Physics (AREA)
  • Social Psychology (AREA)
  • Human Computer Interaction (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

A human body gait acquisition and analysis system and method based on inertial sensor information fusion comprises at least 2 IMU sensors, 2 triaxial accelerometer sensors, a controller and an upper computer; the two same IMU sensors are arranged on the thigh and the shank on at least one side of a human body, are coaxially and homodromous arranged at the positions outside the thigh and the shank and are used for transmitting the information of the acceleration and the angular velocity of the limb sections of the thigh and the shank which are sensitive to the sensors to the controller; the two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped steel plate, the direction of a sensitive axis x of each sensor is parallel to the long edge of the steel plate, the direction of a sensitive axis y of each sensor is parallel to the short edge of the steel plate, the strip-shaped steel plate is arranged at the center of the back of the waist, the long edge of the steel plate is parallel to the spine direction of the human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration information in the spine direction of the waist of the human body is sensed and transmitted to; the controller and the upper computer finish classifying the asynchronous state and predicting the gait.

Description

Human body gait acquisition and analysis system and method based on inertial sensor information fusion
Technical Field
The invention relates to a human body gait acquisition and analysis system and method based on inertial sensor information fusion.
Background
The rehabilitation and disability-assisting lower limb exoskeleton and the military load-bearing exoskeleton need to help a wearer to walk and stand or assist the wearer to bear load, and the rehabilitation and disability-assisting lower limb exoskeleton and the military load-bearing exoskeleton should have coordinated gait laws. In order to realize good gait tracking and control of the lower limb exoskeleton and improve the dynamic stability of the lower limb exoskeleton, gait information acquisition, classification, tracking, prediction and other analysis of gait data of a wearer are required.
(1) With the development of sensing technology and digital technology, more and more methods are available for detecting human body movement gait information, and the commonly used methods mainly include image sequence analysis, electromyographic signal detection, angle/angular velocity detection, accelerometer detection and the like.
(2) Analyzing the image sequence, namely detecting a motion and a moving object from the image sequence by utilizing a computer vision technology and carrying out motion analysis, tracking or identification on the motion and the moving object; electromyographic signal detection: the surface electromyogram signal is a complex comprehensive result of the electrical activity of the sub-epidermal muscle on the surface of the skin in time and space, and the electromyogram signal of the surface of the human body is detected by an electromechanical sensor and is subjected to movement gait detection and analysis; angle/angular velocity detection: integrating data output by a gyroscope constant angular velocity sensor to calculate a human body motion angle, and further performing human body gait tracking, predicting and analyzing and the like; the accelerometer detection method comprises the following steps: and detecting the gravity vector component by using an acceleration sensor, and calculating the motion angle of the limbs of the human body so as to perform gait tracking, analysis and operation and the like.
Analyzing an image sequence: the real-time detection of gait data has high cost, large data processing capacity and serious time delay
Electromyographic signal detection: the used sensor and the matched device have high price, the difficulty of signal identification and classification is high, the electromyographic signal detection is influenced by the detection position, sweat, temperature and the like, the electromyographic signal detection is easily interfered by signals, and the repeatability is not high
Angle/angular velocity detection: due to the adoption of an integral algorithm, errors are continuously accumulated, so that the error of the required joint angle is gradually increased
The accelerometer detection method comprises the following steps: the method is only suitable for static calculation and is not popularized to dynamic joint angle calculation.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the defects of the prior art are overcome, and a human body gait acquisition and analysis system and a human body gait acquisition and analysis method based on inertial sensor information fusion are provided.
The technical solution of the invention is as follows: the human gait acquisition and analysis system based on IMU information fusion comprises at least 2 IMU sensors, 2 triaxial accelerometer sensors, a controller and an upper computer;
the two same IMU sensors are arranged on the thigh and the shank on at least one side of a human body, are coaxially and homodromous arranged at the positions outside the thigh and the shank and are used for transmitting the information of the acceleration and the angular velocity of the limb sections of the thigh and the shank which are sensitive to the sensors to the controller;
the two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped steel plate, the direction of a sensitive axis x of each sensor is parallel to the long edge of the steel plate, the direction of a sensitive axis z of each sensor is parallel to the short edge of the steel plate, the strip-shaped steel plate is arranged at the center of the back of the waist, the long edge of the steel plate is parallel to the spine direction of the human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration sensors are used for sensing the acceleration information in the spine direction of the waist of the;
the controller obtains knee joint angle information according to the received acceleration and angular velocity information of the limbs of the large leg and the small leg and transmits the knee joint angle information to the upper computer, and differential operation is carried out according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist and transmits the posture information to the upper computer;
the upper computer obtains the step frequency and the step length information of the wearer according to the received knee joint angle information and the leg length of the wearer, obtains the step frequency and the step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information, and determines and stores the more reliable gait information of the wearer finally according to the two step frequency and step length information; after the gait information is stored, classifying the dyssynchrony according to the gait information at each moment; and finally, establishing a human body gait prediction model according to the gait data at the current moment and the gait information at the previous moment to predict the gait.
Further, the controller obtains posture information of the human waist and spine direction according to the received acceleration information of the human waist and spine direction of the 2 acceleration sensors, and the specific implementation mode is as follows: firstly, defining a coordinate system o-xyz of the back of a human body, wherein the upward direction of the back of the human body along the spine is the positive direction of an x axis, the left side direction of the intersection line of the plane of the back of the human body and a horizontal plane is the positive direction of a z axis, and the forward direction vertical to the x axis and the z axis is the positive direction of a y axis;
then, according to the triaxial acceleration information of the 2 accelerometer sensors, obtaining triaxial acceleration information a after filtering out dynamic errors caused by movement by adopting a differential calculation method, wherein the acceleration information is a component a under a coordinate system of the back of the human bodyx、ay、az，ax、azIs the gravity vector component under the posture of the back of the human body;
finally, the human back posture information can be calculated according to the gravity vector component informationThe size of the message: pitch angle
Figure BDA0001253118590000031
And roll angle
Figure BDA0001253118590000032
g is the acceleration of gravity; the pitch angle is the angle of rotation about the z-axis and the roll angle is the angle of rotation about the y-axis.
Further, the difference calculating method specifically includes: the acceleration collected by 2 accelerometer sensors coaxially and equidirectionally arranged on a steel plate is a1、a2With acceleration of gravity g, mounting vector R of two accelerometer sensors1、R2And a mounting position r1、r2Angular velocity ω of rotation1、ω2In the context of a correlation, the correlation,
Figure BDA0001253118590000033
then the value after filtering out the motion disturbance acceleration by adopting a differential calculation method is as follows:
Figure BDA0001253118590000034
further, the controller obtains knee joint angle information according to the received acceleration and angular velocity information of the upper leg limb segment and the lower leg limb segment, and the specific implementation mode is as follows:
firstly, calculating acceleration posture information of the large leg limb segment and the small leg limb segment by adopting a gravity vector projection method according to the acceleration information;
then, obtaining the angular velocity posture information of the large leg limb segment and the small leg limb segment by adopting an integral algorithm according to the output angular velocity information of the gyroscope;
then, carrying out information fusion according to the angular velocity posture information and the acceleration posture information of the large and small leg limb segments to obtain the processed large and small leg limb segment posture information;
and finally, obtaining knee joint angle information according to the difference of the posture information of the fused thigh and leg limb sections.
Further, the information fusion step is as follows:
(1) establishing a state equation aiming at the calculation of the angular velocity attitude information
Figure BDA0001253118590000041
Wherein X1For angular velocity attitude information calculated from the gyroscope output, X2Is the gyro output error, ukOutputting angular velocity information for the gyroscope, and T is a calculation time interval;
(2) establishing a measurement equation for accelerometer calculation attitude information
Figure BDA0001253118590000042
Wherein Y iskFor outputting calculated angular velocity attitude information from accelerometers, RkOutputting an error for the accelerometer;
(3) and performing recursion calculation according to the output values of the gyroscope and the accelerometer at each k moment, and calculating the posture information of the limbs of the big leg and the small leg after processing.
Furthermore, the upper computer establishes a gait classification model by adopting a support vector machine algorithm according to the gait information at each moment, and classifies the different gait by utilizing the gait classification model.
Further, the previous gait information is at least data of four time points before the current time.
Further, 4-7 time point data are typically selected.
Further, the human gait prediction model is established by adopting a mode of carrying out a neural network model, wherein the input quantity of the neural network is as follows:
Figure BDA0001253118590000043
wherein ImIn order to input the quantity of the input,
Figure BDA0001253118590000044
gait information of the previous N moments of the current moment;
the output quantity is as follows:
Figure BDA0001253118590000045
wherein O ismIn order to be an output quantity,
Figure BDA0001253118590000046
the gait information at the current moment.
A human body gait acquisition and analysis method based on inertial sensor information fusion comprises the following steps:
(1) two identical IMU sensors are arranged on the upper leg and the lower leg on at least one side of the human body, are ensured to be coaxially and homodromous arranged at the positions outside the upper leg and the lower leg and are used for sensing the acceleration and angular velocity information of the limb sections of the upper leg and the lower leg;
the method comprises the steps that two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped steel plate, the x-axis direction of a sensitive axis of each sensor is parallel to the long edge of the steel plate, the y-axis direction of the sensitive axis of each sensor is parallel to the short edge of the steel plate, the strip-shaped steel plate is arranged in the center of the back of the waist, the long edge of the steel plate is parallel to the spine direction of a human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration information in the spine;
(2) processing the acceleration and angular velocity information of the large leg limb segment and the small leg limb segment to obtain knee joint angle information; carrying out differential operation according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist;
(3) obtaining the step frequency and step length information of the wearer according to the knee joint angle information and the leg length of the wearer; obtaining the step frequency and step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information;
(4) determining and storing more reliable gait information of the final wearer according to the two step frequencies and the step length information in the step (3);
(5) after the gait information in the preset time period is stored, a gait classification model is established by adopting a support vector machine algorithm according to the gait information at each moment, and different gait is classified;
(6) and finally, establishing a human body gait prediction model by adopting a neural network according to the gait data at the current moment and the gait information at the previous moment so as to predict the gait.
Compared with the prior art, the invention has the beneficial effects that:
the system and the method of the invention are matched with the exoskeleton robot, and at present, in the aspect of controlling the exoskeleton robot for civil rehabilitation and disabled-assisting walking training or in the aspect of controlling the exoskeleton robot for military load, gait information acquisition, classification, tracking, prediction and other analysis of gait data of a human body are required to be carried out on normal gait data of the human body so as to realize the gait tracking and control of the exoskeleton of lower limbs and improve the dynamic stability of the exoskeleton.
The invention provides a human body gait acquisition and analysis system and method based on inertial sensor information fusion. The system consists of 2 accelerometers, an inertial sensor network of 2 IMU sensors, a binding band, a controller, a wireless transmission module and an upper computer. The system can realize real-time non-contact measurement of human gait rules, establish an accurate mathematical model of the human gait rules through information fusion processing of different output data of the sensor, and realize classification, tracking and prediction of the human gait rules by fusing algorithms such as a support vector machine, a neural network model and the like.
Drawings
FIG. 1 is a schematic diagram of the system of the present invention;
FIG. 2 is a flow chart of the present invention.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings.
The system proposed herein is shown in fig. 1 and consists of 2 accelerometers, an inertial sensor network of 2 MEMS type IMU sensors, a strap, a controller, and an upper computer, as shown in fig. 1. The system can realize real-time non-contact measurement of human gait rules and perform information fusion processing of different output data of the sensor. The concrete description is as follows
Sensor arrangement
The two same IMU sensors are arranged on the thigh and the shank on at least one side of a human body, are coaxially and homodromous arranged at the positions outside the thigh and the shank and are used for transmitting the information of the acceleration and the angular velocity of the limb sections of the thigh and the shank which are sensitive to the IMU sensors to the controller;
the two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped steel plate, the direction of a sensitive axis x of each sensor is parallel to the long edge of the steel plate, the direction of a sensitive axis y of each sensor is parallel to the short edge of the steel plate, the strip-shaped steel plate is arranged at the center of the back of the waist, the long edge of the steel plate is parallel to the spine direction of the human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration information in the spine direction of the waist of the human body is sensed and transmitted to;
(II) controller
The controller comprises a core processing unit, a power supply module, a wireless communication module (WIFI or Bluetooth module), an inertial navigation network sensor control and communication module and the like, and works such as sensor acquisition control, transmission, storage and the like. The controller transmits the data to the upper computer through the wireless communication module for data processing and analysis. The power module supplies power for the sensor network and the controller. The inertial navigation network sensor control and communication module has the main function of sending a control instruction to acquire a sensor signal and transmitting the sensor signal back to the controller for the core processing unit to process the signal.
The core processing unit in the controller mainly completes the following two tasks:
1. the knee joint angle information is obtained according to the received acceleration and angular velocity information of the limbs of the large and the small legs and is transmitted to an upper computer, and the specific implementation mode is as follows:
firstly, calculating acceleration posture information of the large leg limb segment and the small leg limb segment by adopting a gravity vector projection method according to the acceleration information;
then, according to the angular velocity information, an integral algorithm is adopted to obtain the angular velocity posture information of the thigh and the crus limb section,
then, carrying out information fusion according to the posture information of the two large and small leg limb sections to obtain the posture information of the processed large and small leg limb sections; the information fusion method is used for establishing a Kalman filtering model aiming at the attitude information characteristics of acceleration and angular velocity calculation respectively, and the specific implementation mode is as follows: (1) establishing a state equation aiming at the calculation of the angular velocity attitude
Figure BDA0001253118590000071
Wherein X1Calculating attitude angle, X, for a gyro2Is the gyro output error, ukOutputting angular velocity information for the gyroscope at the moment k, and T is a calculation time interval; (2) establishing a measurement equation for an accelerometer computing attitude
Figure BDA0001253118590000072
Wherein Y iskAttitude information calculated for accelerometer at time k, RkOutputting an error for the accelerometer; (3) and carrying out recursion calculation according to the output values of the gyroscope and the accelerometer at each k moment, and calculating accurate attitude information.
And finally, obtaining knee joint angle information according to the difference between the posture information of the limbs of the upper leg and the lower leg.
2. Carrying out differential operation according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist and transmitting the posture information to an upper computer; the determination of the attitude information comprises the following steps:
firstly, defining a coordinate system o-xyz of the back of a human body, wherein the upward direction of the back of the human body along the spine is the positive direction of an x axis, the left side direction of the intersection line of the plane of the back of the human body and a horizontal plane is the positive direction of a z axis, and the forward direction vertical to the x axis and the z axis is the positive direction of a y axis;
then, according to the triaxial acceleration information of the 2 accelerometer sensors, obtaining triaxial acceleration information a after filtering out dynamic errors caused by movement by adopting a differential calculation method, wherein the acceleration information is a component a under a coordinate system of the back of the human bodyx、ay、az，ax、azIs the gravity vector component under the posture of the back of the human body;
the difference calculation method specifically includes: the acceleration collected by 2 accelerometer sensors coaxially and equidirectionally arranged on a steel plate is a1、a2With acceleration of gravity g, mounting vector R of two accelerometer sensors1、R2And a mounting position r1、r2Angular velocity ω of rotation1、ω2In the context of a correlation, the correlation,
Figure BDA0001253118590000073
Figure BDA0001253118590000074
because the two sensors are arranged on the same steel plate, the angular velocities sensed by the two sensors are the same, and the values obtained by filtering out the motion disturbance acceleration by adopting a differential calculation method are as follows:
Figure BDA0001253118590000081
and finally, calculating the size of the posture information of the back of the human body according to the gravity vector component information: pitch angle
Figure BDA0001253118590000082
And roll angle
Figure BDA0001253118590000083
g is the acceleration of gravity; the pitch angle is the angle of rotation about the z-axis and the roll angle is the angle of rotation about the y-axis. Pitch angle direction according tox、ayDirection determination, roll angle direction according tozAnd (4) determining the direction.
(III) upper computer
The upper computer obtains the step frequency and the step length information of the wearer according to the received knee joint angle information and the leg length of the wearer, obtains the step frequency and the step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information, and determines and stores the more reliable gait information of the wearer finally according to the two step frequency and step length information; after the gait information is stored (the gait condition required by each exoskeleton gait control is acquired for at least 10 walking cycles), a gait classification model is established by adopting a support vector machine algorithm according to the gait information at each moment, and different gait is classified to provide a theoretical basis for the exoskeleton gait classification control; and finally, establishing a human body gait prediction model by adopting a neural network according to the gait data at the current moment and the gait information at the previous moment so as to provide a theoretical basis for exoskeleton gait prediction control. The specific steps for establishing the human gait prediction model are as follows:
(3.1) storing gait information at each moment;
(3.2) calculating the input quantity of the neural network model at each moment:
Figure BDA0001253118590000084
wherein ImIn order to input the quantity of the input,
Figure BDA0001253118590000085
gait information at the previous N moments of the current moment k;
(3.3) calculating the output quantity of the neural network model at each moment:
Figure BDA0001253118590000086
wherein O ismIn order to be an output quantity,
Figure BDA0001253118590000087
the gait information at the current moment; (ii) a
(3.4) selecting a proper neural network model, loading the input quantity and the output quantity of each moment into the RBF neural network model by adopting the RBF neural network, and training;
and (3.5) deriving the trained parameters of the RBF neural network model for gait information prediction.
As shown in FIG. 2, the human gait acquisition and analysis method based on inertial sensor information fusion of the invention comprises the following steps:
(1) two identical IMU sensors are arranged on the upper leg and the lower leg on at least one side of the human body, are ensured to be coaxially and homodromous arranged at the positions outside the upper leg and the lower leg and are used for sensing the acceleration and angular velocity information of the limb sections of the upper leg and the lower leg;
the method comprises the steps that two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped steel plate, the x-axis direction of a sensitive axis of each sensor is parallel to the long edge of the steel plate, the y-axis direction of the sensitive axis of each sensor is parallel to the short edge of the steel plate, the strip-shaped steel plate is arranged in the center of the back of the waist, the long edge of the steel plate is parallel to the spine direction of a human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration information in the spine;
(2) processing the acceleration and angular velocity information of the large leg limb segment and the small leg limb segment to obtain knee joint angle information; carrying out differential operation according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist;
(3) obtaining the step frequency and step length information of the wearer according to the knee joint angle information and the leg length of the wearer; obtaining the step frequency and step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information;
(4) determining and storing more reliable gait information of the final wearer according to the two step frequencies and the step length information in the step (3);
(5) after the gait information in the preset time period is stored, a gait classification model is established by adopting a support vector machine algorithm according to the gait information at each moment, different gait is classified, and a theoretical basis is provided for exoskeleton gait classification control;
(6) and finally, establishing a human body gait prediction model by adopting a neural network according to the gait data at the current moment and the gait information at the previous moment, and providing a theoretical basis for exoskeleton gait prediction control.
The implementation of the specific steps in the method may use the same processing method as the system, and will not be described herein.
The invention has not been described in detail in part of the common general knowledge of those skilled in the art.

Claims (9)

1. Human gait acquisition and analysis system based on inertial sensor information fusion is characterized in that: the device comprises at least 2 IMU sensors, 2 triaxial accelerometer sensors, a controller and an upper computer;
the two same IMU sensors are arranged on the thigh and the shank on at least one side of a human body, are coaxially and homodromous arranged at the positions outside the thigh and the shank and are used for transmitting the information of the acceleration and the angular velocity of the limb sections of the thigh and the shank which are sensitive to the sensors to the controller;
the two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped fixing plate, the x-axis direction of a sensitive axis of the sensor is parallel to the long edge of the fixing plate, the z-axis direction of the sensitive axis of the sensor is parallel to the short edge of the fixing plate, the strip-shaped fixing plate is arranged at the center position of the back of the waist, the long edge of the fixing plate is parallel to the spine direction of the human body, the distance from the installation position to a hip joint rotating shaft is recorded, and the acceleration information in the spine direction of the waist of the human body;
the controller obtains knee joint angle information according to the received acceleration and angular velocity information of the limbs of the large leg and the small leg and transmits the knee joint angle information to the upper computer, and differential operation is carried out according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist and transmits the posture information to the upper computer;
the upper computer obtains the step frequency and the step length information of the wearer according to the received knee joint angle information and the leg length of the wearer, obtains the step frequency and the step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information, and determines and stores the more reliable gait information of the wearer finally according to the two step frequency and step length information; after the gait information is stored, classifying the dyssynchrony according to the gait information at each moment; finally, according to the gait data at the current moment and the gait information at the previous moment, a human body gait prediction model is established for carrying out gait prediction;
the controller obtains knee joint angle information according to the received acceleration and angular velocity information of the large leg limb segment and the small leg limb segment, and the specific implementation mode is as follows:
firstly, calculating acceleration posture information of the large leg limb segment and the small leg limb segment by adopting a gravity vector projection method according to the acceleration information; then, obtaining the angular velocity posture information of the large leg limb segment and the small leg limb segment by adopting an integral algorithm according to the output angular velocity information of the gyroscope; then, carrying out information fusion according to the angular velocity posture information and the acceleration posture information of the large and small leg limb segments to obtain the processed large and small leg limb segment posture information; and finally, obtaining knee joint angle information according to the difference of the posture information of the fused thigh and leg limb sections.
2. The system of claim 1, wherein: the controller obtains human waist spine direction posture information according to the received acceleration information of the human waist spine direction of the 2 acceleration sensors, and the specific implementation mode is as follows:
firstly, defining a coordinate system o-xyz of the back of a human body, wherein the upward direction of the back of the human body along the spine is the positive direction of an x axis, the left side direction of the intersection line of the plane of the back of the human body and a horizontal plane is the positive direction of a z axis, and the forward direction vertical to the x axis and the z axis is the positive direction of a y axis;
then obtaining triaxial acceleration information a after filtering dynamic errors caused by movement by adopting a differential calculation method according to triaxial acceleration information of 2 accelerometer sensors, wherein the component of the acceleration information in a human back coordinate system is ax、ay、az，ax、azIs the gravity vector component under the posture of the back of the human body;
and finally, calculating the size of the posture information of the back of the human body according to the gravity vector component information: pitch angle
Figure FDA0002299601960000021
And roll angle
Figure FDA0002299601960000022
g is the acceleration of gravity; the pitch angle is the angle of rotation about the z-axis and the roll angle is the angle of rotation about the y-axis.
3. The system of claim 2, wherein: the difference calculating method specifically comprises the following steps: the acceleration collected by 2 accelerometer sensors coaxially and equidirectionally arranged on a fixed plate is a1、a2With acceleration of gravity g, mounting vector R of two accelerometer sensors1、R2And a mounting position r1、r2Angular velocity ω of rotation of 2 accelerometer sensors1、ω2In the context of a correlation, the correlation,
Figure FDA0002299601960000023
Figure FDA0002299601960000024
then the value after filtering out the motion disturbance acceleration by adopting a differential calculation method is as follows:
Figure FDA0002299601960000025
4. the system of claim 1, wherein: the information fusion steps are as follows:
(1) establishing a state equation aiming at the calculation of the angular velocity attitude information
Figure FDA0002299601960000026
Wherein X1For outputting calculated angular velocity attitude information from the gyroscope of the IMU sensor, X2Gyroscope output error, u, for IMU sensorkOutputting angular velocity information for a gyroscope of the IMU sensor, wherein T is a calculation time interval;
(2) establishing a measurement equation for accelerometer calculation attitude information
Figure FDA0002299601960000031
Wherein Y iskFor outputting calculated angular velocity attitude information from accelerometers, RkOutputting an error for the accelerometer;
(3) and performing recursion calculation according to the output values of the gyroscope and the accelerometer at each k moment, and calculating the posture information of the limbs of the big leg and the small leg after processing.
5. The system of claim 1, wherein: and the upper computer establishes a gait classification model by adopting a support vector machine algorithm according to the gait information at each moment, and classifies the different gait by using the gait classification model.
6. The system of claim 1, wherein the gait information of the previous time is data of at least four time points before the current time.
7. The system of claim 6, wherein 4-7 point-in-time data are selected.
8. The system of claim 1, wherein the human gait prediction model is established by a neural network model, wherein the neural network has the following inputs:
Figure FDA0002299601960000032
wherein ImIn order to input the quantity of the input,
Figure FDA0002299601960000033
the gait information is the gait information of the previous N moments of the current moment.
9. A gait collection and analysis method of a human body gait collection and analysis system according to claim 1, characterized by the steps of:
(1) two identical IMU sensors are arranged on the upper leg and the lower leg on at least one side of the human body, are ensured to be coaxially and homodromous arranged at the positions outside the upper leg and the lower leg and are used for sensing the acceleration and angular velocity information of the limb sections of the upper leg and the lower leg;
the method comprises the following steps that two same accelerometer sensors are coaxially and unidirectionally arranged on a strip-shaped fixing plate, the x-axis direction of a sensitive axis of each sensor is parallel to the long edge of the fixing plate, the z-axis direction of the sensitive axis of each sensor is parallel to the short edge of the fixing plate, the strip-shaped fixing plate is arranged in the center of the back of the waist, the long edge of the fixing plate is parallel to the spine direction of a human body, the distance from the mounting position to a hip joint rotating shaft is recorded, and the acceleration information in the;
(2) processing the acceleration and angular velocity information of the large leg limb segment and the small leg limb segment to obtain knee joint angle information; carrying out differential operation according to the received acceleration information of the human waist in the spine direction to obtain posture information of the human waist;
(3) obtaining the step frequency and step length information of the wearer according to the knee joint angle information and the leg length of the wearer; obtaining the step frequency and step length information of the wearer according to the waist posture information and the waist accelerometer sensor mounting position information;
(4) determining and storing more reliable gait information of the final wearer according to the two step frequencies and the step length information in the step (3);
(5) after the gait information in the preset time period is stored, a gait classification model is established by adopting a support vector machine algorithm according to the gait information at each moment, and different gait is classified;
(6) and finally, establishing a human body gait prediction model by adopting a neural network according to the gait data at the current moment and the gait information at the previous moment so as to predict the gait.
CN201710178883.9A 2017-03-23 2017-03-23 Human body gait acquisition and analysis system and method based on inertial sensor information fusion Active CN106821391B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710178883.9A CN106821391B (en) 2017-03-23 2017-03-23 Human body gait acquisition and analysis system and method based on inertial sensor information fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710178883.9A CN106821391B (en) 2017-03-23 2017-03-23 Human body gait acquisition and analysis system and method based on inertial sensor information fusion

Publications (2)

Publication Number Publication Date
CN106821391A CN106821391A (en) 2017-06-13
CN106821391B true CN106821391B (en) 2020-03-24

Family

ID=59130411

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710178883.9A Active CN106821391B (en) 2017-03-23 2017-03-23 Human body gait acquisition and analysis system and method based on inertial sensor information fusion

Country Status (1)

Country Link
CN (1) CN106821391B (en)

Families Citing this family (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107029409A (en) * 2017-06-16 2017-08-11 河北工业大学 A kind of running training system based on virtual reality technology
CN109387220B (en) * 2017-08-03 2022-07-08 北京自动化控制设备研究所 Method for correcting azimuth installation deviation of walking micro-inertial navigation equipment
CN109492362A (en) * 2017-09-13 2019-03-19 腾讯科技（深圳）有限公司 Verification method, device, mobile terminal and the computer storage medium of sensitive operation
CN107928678B (en) * 2017-12-06 2020-08-28 中北大学 Wearable patella instability testing device and method based on micro-inertia measurement array
CN107918492A (en) * 2017-12-22 2018-04-17 安庆师范大学 A kind of human motion in face of Intelligent lower limb artificial limb is intended to precognition recognition methods
CN108245164B (en) * 2017-12-22 2021-03-26 北京精密机电控制设备研究所 Human body gait information acquisition and calculation method for wearable inertial device
CN108334827B (en) * 2018-01-23 2024-03-26 大连乾函科技有限公司 Gait identity authentication method based on intelligent shoe and intelligent shoe
CN108520242A (en) * 2018-04-11 2018-09-11 上海理工大学 The method and apparatus that a kind of heel for Gait Recognition detects
CN109091151B (en) * 2018-09-06 2021-07-16 中国人民解放军战略支援部队信息工程大学 Pedestrian falling detection method and device based on MIMU
CN110193830B (en) * 2019-05-24 2022-10-11 上海大学 Ankle joint gait prediction method based on RBF neural network
CN110755070B (en) * 2019-08-28 2022-07-05 北京精密机电控制设备研究所 Multi-sensor fusion-based lower limb movement pose rapid prediction system and method
CN110522458A (en) * 2019-10-15 2019-12-03 北京理工大学 A kind of gait real-time identification method suitable for knee joint ectoskeleton
CN111076748A (en) * 2020-01-06 2020-04-28 重庆邮电大学 Horizontal inclinometer error compensation method and system based on MEMS accelerometer
CN111241983B (en) * 2020-01-07 2023-09-26 京东科技信息技术有限公司 Gesture detection method, device and system, electronic equipment and storage medium
CN111469117B (en) * 2020-04-14 2022-06-03 武汉理工大学 Human motion mode detection method of rigid-flexible coupling active exoskeleton
RU2020116716A (en) * 2020-05-07 2021-11-08 Общество С Ограниченной Ответственностью "Экзоатлет" METHOD AND DEVICE FOR DIAGNOSING THE FUNCTIONAL STATE OF THE SUPPORT-MOTOR EQUIPMENT
CN111544005B (en) * 2020-05-15 2022-03-08 中国科学院自动化研究所 Parkinson's disease dyskinesia quantification and identification method based on support vector machine
CN111938656A (en) * 2020-07-28 2020-11-17 南京审计大学 Real-time walking mode identification method and system based on knee joint exoskeleton
CN113143256B (en) * 2021-01-28 2023-09-26 上海电气集团股份有限公司 Gait feature extraction method, lower limb evaluation and control method, device and medium
CN113284601A (en) * 2021-06-10 2021-08-20 上海大学 Cloud platform management system based on laser gait detection
CN116442202B (en) * 2023-06-19 2023-08-18 贵州航天控制技术有限公司 Waist boosting equipment control method based on back posture information
CN118370534B (en) * 2024-06-21 2024-09-17 浙江大学 Gait analysis method and system

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101558996B (en) * 2009-05-15 2010-11-10 天津大学 Gait recognition method based on orthogonal projection three-dimensional reconstruction of human motion structure
CN104567933B (en) * 2015-01-19 2017-07-21 华侨大学 Suppress the common-mode error of three axle inertial sensors and the method and apparatus of coherent noise
CN104613963B (en) * 2015-01-23 2017-10-10 南京师范大学 Pedestrian navigation system and navigation locating method based on human cinology's model
CN104729507B (en) * 2015-04-13 2018-01-26 大连理工大学 A kind of gait recognition method based on inertial sensor
CN104864884B (en) * 2015-05-12 2018-03-06 上海大学 A kind of pedestrian's stride measuring system
EP3297520B1 (en) * 2015-05-18 2022-11-02 Vayu Technology Corp. Devices for measuring human gait and related methods of use

Also Published As

Publication number Publication date
CN106821391A (en) 2017-06-13

Similar Documents

Publication Publication Date Title
CN106821391B (en) Human body gait acquisition and analysis system and method based on inertial sensor information fusion
JP7342864B2 (en) Positioning program, positioning method, and positioning device
Yan et al. Ronin: Robust neural inertial navigation in the wild: Benchmark, evaluations, and new methods
CN109579853B (en) Inertial navigation indoor positioning method based on BP neural network
CN108245164B (en) Human body gait information acquisition and calculation method for wearable inertial device
KR101751760B1 (en) Method for estimating gait parameter form low limb joint angles
CN110146077A (en) Pose of mobile robot angle calculation method
EP2850392B1 (en) Method for step detection and gait direction estimation
Ribeiro et al. Inertial measurement units: A brief state of the art on gait analysis
Liu et al. A wearable human motion tracking device using micro flow sensor incorporating a micro accelerometer
CN111024126B (en) Self-adaptive zero-speed correction method in pedestrian navigation positioning
CN114224325A (en) Gait analysis system and method for calculating joint moment and angle by using inertial sensor
Guidolin et al. On the accuracy of imus for human motion tracking: a comparative evaluation
CN114663913A (en) Human body gait parameter extraction method based on Kinect
Wang et al. A yaw correction method for pedestrian positioning using two low-cost MIMUs
CN113229806A (en) Wearable human body gait detection and navigation system and operation method thereof
CN115607146B (en) Wearable single-node device for leg posture estimation and measurement method
Xu et al. Measuring human joint movement with IMUs: Implementation in custom-made low cost wireless sensors
Comotti et al. Inertial based hand position tracking for future applications in rehabilitation environments
CN104913772B (en) A kind of pedestrian movement's detection method based on leg posture information
Liu et al. Measurement of human lower limb orientations and ground reaction forces using wearable sensor systems
Petruska et al. Non-drifting limb angle measurement relative to the gravitational vector during dynamic motions using accelerometers and rate gyros
Alonge et al. Use of accelerometers and gyros for hip and knee angle estimation
Yi et al. G-constellations: G-sensor motion tracking systems
Salehi et al. Validation of a Low-cost Inertial Exercise Tracker.

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