WO2018132999A1 - 一种用于可穿戴式设备的人体步长测量方法及其测量设备 - Google Patents

一种用于可穿戴式设备的人体步长测量方法及其测量设备 Download PDF

Info

Publication number
WO2018132999A1
WO2018132999A1 PCT/CN2017/071674 CN2017071674W WO2018132999A1 WO 2018132999 A1 WO2018132999 A1 WO 2018132999A1 CN 2017071674 W CN2017071674 W CN 2017071674W WO 2018132999 A1 WO2018132999 A1 WO 2018132999A1
Authority
WO
WIPO (PCT)
Prior art keywords
calf
axis
sensor
acceleration
standing
Prior art date
Application number
PCT/CN2017/071674
Other languages
English (en)
French (fr)
Inventor
刘涛
王磊
李庆国
Original Assignee
浙江大学
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 浙江大学 filed Critical 浙江大学
Priority to PCT/CN2017/071674 priority Critical patent/WO2018132999A1/zh
Publication of WO2018132999A1 publication Critical patent/WO2018132999A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/10—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration
    • G01C21/12—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning
    • G01C21/16—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 by using measurements of speed or acceleration executed aboard the object being navigated; Dead reckoning by integrating acceleration or speed, i.e. inertial navigation
    • G01C21/166—Mechanical, construction or arrangement details of inertial navigation systems
    • 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
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B21/00—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant
    • G01B21/02—Measuring arrangements or details thereof, where the measuring technique is not covered by the other groups of this subclass, unspecified or not relevant for measuring length, width, or thickness
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20—Instruments for performing navigational calculations

Definitions

  • the invention belongs to the field of wearable sensors, and in particular relates to a human body step measuring method for a wearable device and a measuring device thereof.
  • Gait parameters are physical parameters during human walking, which can reflect human exercise ability and have great application value. Therefore, many researchers are currently studying the measurement of gait parameters. Wearable sensors include inertial measurement units, ultrasonic sensors, miniature cameras, etc., compared to large laboratory equipment that measure gait parameters such as optical motion capture systems, force gauges, etc., which are small, inexpensive, and free of time. The advantages of space limitation and easy promotion are widely used in the field of gait parameter measurement. At present, there are many studies using wearable sensors placed on the legs, waist, wrists, chest, etc. to measure gait time parameters, such as gait events, gait cycles, etc.; some also measure gait space parameters, such as walking speed. , step and so on. For example, the invention patent No.
  • CN201510887154.1 discloses an indoor positioning step calculation method, which places the inertial measurement unit at the waist of the person, and the person obtains the inertial sensor data of the person during the indoor walking process, and then calculates the step size.
  • the invention patent of the application No. CN201310007945.1 discloses a step calculation method and device for acquiring the acceleration values of the respective axes in the three-dimensional coordinate system; and calculating the step compensation coefficient according to the acceleration values of the respective axes; The step compensation coefficient and the preset preset step size determine the final step size of the carrier movement range.
  • the determination of the step size is of great significance.
  • Some diseases such as Parkinson's syndrome can reduce the body's ability to exercise, and the pace and stride length are reduced. Therefore, the step and pace can be used to reflect the lower limbs' ability to move.
  • the step and pace cannot reflect the difference in the ability of the two sides of the human body caused by hemiplegia, that is, the gait is not correct, and we can quantify the asymmetry of the gait on both sides of the human body according to the ratio of the two-leg steps, and The length can also be used to find parameters such as step size and pace, so the step size has greater clinical application value.
  • the existing method of measuring the step size using a wearable device mainly utilizes the up and down displacement of the center of gravity of the waist or the advantage.
  • Mathematical models based on variables such as step frequency and acceleration have defects such as poor precision and poor adaptability. Therefore, it is necessary to propose a new type of step measurement method with high precision and strong adaptability in different populations.
  • the object of the present invention is to solve the defects in the prior art that the gait measurement accuracy is low, the single step size cannot be measured, and the human body step measurement method and the measuring device thereof for the wearable device are provided.
  • the step distance refers to the distance between the places where the same foot is adjacent during the walking process.
  • the left leg step refers to the distance between the adjacent positions of the left foot during walking, and the right leg step refers to the right during the walking process.
  • the distance between the feet adjacent to the foot; the step length refers to the distance between the adjacent places of the feet during the walking of the person, and the left leg step refers to the step size of the left foot in front of the walking process, right
  • the leg step is the step size in which the right foot is in front of the walking process. In general, the step size is equal to the sum of the two leg steps.
  • Gait events refer to important moments in each gait cycle of a person during walking. There are four gait events: mid-swing, foot landing, middle middle, and foot off the ground.
  • the walking gait cycle of a person is shown in Figure 4. Taking the right leg as an example, the left leg supports the human body, and the right leg takes the forward phase as the swinging phase of the right leg; when the right leg swings to the vicinity of the left leg, it swings.
  • the gait event can be detected by the angular velocity characteristics of the leg calf, as shown in FIG.
  • the angular velocity of the calf usually has two relatively large peaks in one cycle, one high and one short, the higher peak corresponds to the middle of the swing, the shorter one corresponds to the middle of the middle; the middle of the swing has a small number of small negative peaks. In the area, this is the vibration caused by the landing of the foot.
  • the first negative peak is the moment when the foot is landing. After the middle of the standing, there is a trough before the middle of the next swing, which corresponds to the moment when the foot is off the ground.
  • the standing phase starts from the landing of the leg and the foot ends, and it can be approximated that the area where the angular velocity of the lower leg is negative is the standing phase; the swing phase starts from the leg and ends at the end of the foot, and can be approximated as a figure.
  • the region where the angular velocity of the middle and lower legs is positive is the oscillating phase.
  • the sagittal plane and the plumb line are shown in Figure 1.
  • the sagittal plane is perpendicular to the horizontal plane and divides the human body into two parts. The main movement occurs in the sagittal plane when walking.
  • the plumb line is A line perpendicular to the ground.
  • the horizontal advancement direction is marked in Figure 2, which refers to the side of the sagittal plane where the human body advances in the horizontal direction. to.
  • the acceleration and the moving speed of the calf in the horizontal advancing direction respectively refer to the acceleration of the calf and the component of the moving speed in the horizontal advancing direction.
  • the three-dimensional direction of the calf is defined, as shown in FIG. 2: the X-axis is parallel to the calf; the Y-axis is in the sagittal plane, perpendicular to the calf; and the Z-axis is perpendicular to the sagittal plane.
  • the axis of the sensor placed on the lower leg should be consistent with the calf, but because the sagittal plane is invisible, and the calf is not a regular cylinder, the X-axis, Y-axis, and Z-axis will appear when the sensor is placed.
  • the X-axis, Y-axis, and Z-axis angular velocities measured by the sensor can roughly calculate the three-axis offset angles of the sensors X, Y, and Z, and use this angle to calculate the acceleration and angular velocity of the corresponding axis of the lower leg.
  • the distance from the sensor to the sole is the height from the origin of the sensor axis to the ground when the user is standing still.
  • the present invention solves the technical problem, and the specific technical solutions adopted are as follows:
  • a method for measuring a human body step size for a wearable device comprising the following steps:
  • the present invention also provides several preferred implementations, and the technical features in each preferred mode can be combined with each other without conflict.
  • the above-mentioned wearable device includes sensors for detecting three-axis accelerations of the left and right lower legs X, Y, and Z, and three-axis angular velocities of X, Y, and Z, one on each of the left and right calves.
  • the X, Y axis acceleration and Z axis angular velocity during the walking of the calf on both sides of the user are obtained by the sensor.
  • part of the steps may be implemented in the following specific manner:
  • the gait events during walking are detected, including the landing of the foot and the middle of the standing;
  • each of the left or right calf calculates the angle between the lower leg and the vertical line (the following formula for calculating the left or right calf is not calculated at the same time, but is selected according to the current target to be tested. For example, when it is detected that the left lower leg is in the middle of standing, ⁇ lms is calculated; when it is detected that the right lower leg is in the middle of standing, ⁇ rms is calculated.
  • the formula in the subsequent step also adopts a similar method:
  • ⁇ lms and ⁇ rms are the angles between the left calf and the right calf at the mid-term of their standing and the plumb line, and a lyms and a ryms are the Y-axis accelerations of the left calf and the right calf at the middle of their standing;
  • the angle between the calf and the plumb line in the sagittal plane is calculated by angular velocity integration at various times before the next middle moment of standing:
  • t l , t r are the time of the last standing middle moment of the left and right calves at the current time
  • ⁇ l (t l ) are the left calf at time t l , right calf
  • the angle between the sagittal plane and the plumb line at time t r , ⁇ lz ( ⁇ ), ⁇ rz ( ⁇ ) are the instantaneous Z-axis angular velocity of the left lower leg and the right lower leg, respectively;
  • the acceleration of the two lower legs in the horizontal advancing direction is calculated by the X and Y axis accelerations, and the acceleration at each moment is calculated by the following formula:
  • a hl (t l ) -a lx (t l ) ⁇ sin ⁇ l (t l )+a ly (t l ) ⁇ cos ⁇ l (t l )
  • a hr (t r ) -a rx (t r ) ⁇ sin ⁇ r (t r )+a ry (t r ) ⁇ cos ⁇ r (t r )
  • a hl (t l ) and a hr (t r ) are the accelerations in the horizontal advancing direction of the left calf at time t l and the right calf at time t r
  • a rx (t r ) is the X-axis acceleration of the left calf at time t l and the right calf at time t r
  • a ly (t l ) and a ry (t r ) are respectively the left calf at time t l and right Y-axis acceleration of the calf at time t r .
  • v lms and v rms are the speeds of the left and right calves in the horizontal advancement direction in the middle of their standing
  • ⁇ lzms , ⁇ rzms are the Z-axis angular velocities of the left and right calves in the middle of their standing
  • d is the calf The distance between the placement position of the sensor for measuring acceleration and angular velocity from the sole of the foot;
  • the initial displacement of the calf in the horizontal advancement direction is calculated at the middle of each of the left or right calves:
  • s lms and s rms are the initial displacements of the left and right calves in the horizontal advancement direction during the middle of their standing;
  • v l (t l ) and v r (t r ) are the speeds of movement of the left lower leg at the time t l and the right lower leg at the time t r
  • a hl ( ⁇ ), a hr ( ⁇ ) is the instantaneous acceleration in the horizontal advancing direction of the left lower leg and the right lower leg respectively;
  • the step size of the leg that is generated between the two lower standing moments of the lower leg is calculated by:
  • LSDL, RSDL are the left leg step
  • s lmsl , s rmsl are the initial displacement of the left and right calves in the horizontal advance direction of the middle of the leg before the middle
  • s lmsn is the initial displacement of the left and right calves in the horizontal advancing direction of the middle of the leg
  • T l and T r are the lengths of time between the two adjacent mid-term moments of the left and right calves, respectively.
  • Lmsn and v rmsn are the movement speeds of the left and right calves in the horizontal advancing direction of the middle of the leg, v l (T l ) and v r (T r ) are respectively the left calf at time T l , The speed of the horizontal advancement direction calculated by the acceleration integral of the right lower leg at time T r ;
  • the step size of the leg that is generated between the two lower legs of the calf is calculated by the following formula:
  • LSL is the left leg step length
  • RSL is the right leg step length
  • ⁇ llc and ⁇ rlc are the left and right of the left foot landing time occurring between the middle of the standing position and the middle of the standing position.
  • the angle between the lower leg and the plumb line, ⁇ lrc and ⁇ rrc are the angles between the left and right calves and the plumb line at the moment of landing of the right foot between the middle of the standing position and the middle of the second standing position. .
  • the X, Y axis acceleration and Z axis angular velocity data of each sensor need to be corrected in advance before the acceleration and angular velocity data as the corresponding axis of the calf are substituted into the above formulas.
  • the method of correction is as follows:
  • the straight-line walking process within a certain distance of the target user is monitored in advance, and the angle of the actual direction of the X and Y axes of each sensor deviates from the sagittal plane is determined:
  • the angle at which the actual direction of the X-axis of the sensor deviates from the sagittal plane The angle at which the actual direction of the Y-axis of the sensor deviates from the sagittal plane, The average value of the angular velocity of the X-axis of the sensor during standing phase during the user's walking, The average value of the angular velocity of the Y-axis of the sensor during standing phase during the user's walking, The average value of the Z-axis angular velocity of the sensor when the user is standing during walking;
  • the three-axis acceleration and triaxial angular velocity data measured by each sensor are pre-corrected for the first time before use.
  • the correction formula is:
  • a x , a y , a z are the X, Y, and Z axis accelerations measured by the sensor
  • a x1 , a y1 , and a z1 are the X, Y, and Z axes obtained by the sensor's measurement data after the correction.
  • Acceleration; ⁇ x , ⁇ y , ⁇ z are the angular velocities of the X, Y, and Z axes measured by the sensor
  • ⁇ x1 , ⁇ y1 , ⁇ z1 are the angular velocities of the X, Y, and Z axes obtained by the sensor's measurement data after the correction.
  • the three-axis acceleration and triaxial angular velocity data obtained by each sensor after the first correction must be corrected a second time before use.
  • the correction formula is as follows:
  • a xc , a yc , a zc are the X, Y, and Z axis accelerations obtained by the sensor's measurement data after the second correction
  • ⁇ xc , ⁇ yc , ⁇ zc are the data of the sensor after the second correction Obtained X, Y, and Z axis angular velocities
  • a xc , a yc , a zc are the X, Y, and Z axis accelerations of the calf
  • ⁇ xc , ⁇ yc , and ⁇ zc are the angular velocity of the calf X, Y, and Z axes.
  • Another object of the present invention is to provide a wearable device for implementing the above-described human body step measuring method, comprising two inertial sensors and a host computer, each inertial sensor comprising a three-dimensional accelerometer and a three-dimensional angular velocity meter, an inertial sensor and a host computer Connected for data transmission, the host computer is used to control the two inertial sensors to collect acceleration and angular velocity data, and collect and store the data.
  • a fixing strap for fixing the inertial sensor is also included.
  • the inertial sensor is an inertial sensor based on the MPU6050 chip.
  • the inertial sensor sampling frequency is not less than 100 Hz.
  • the host computer stores the data measured by the inertial sensor into the SD card.
  • the invention has the following beneficial effects:
  • the step size measurement by the wearable device can conveniently and effectively quantify the movement ability of the human body and their differences, and can be conveniently applied to the clinic.
  • Figure 1 is a schematic view showing a sagittal plane and a plumb line in the present invention
  • FIG. 2 is a schematic view showing a sensor placement position, a calf coordinate system, and a horizontal advancing direction in the present invention
  • FIG. 3 is a structural diagram of a wearable device for implementing a measurement step method in the present invention
  • Figure 4 is a schematic view showing the walking cycle of the human body in the present invention.
  • Figure 5 is a schematic diagram of gait event detection in the present invention.
  • Figure 6 is a schematic view showing a two-dimensional geometric model of a lower limb single pendulum in the present invention.
  • Figure 7 is a schematic view showing the mid-angle of the right lower leg swing in the present invention.
  • Figure 8 is a calculation of the mid-speed and displacement of the right lower leg sensor in the present invention.
  • 1, 2 are the inertial sensor units placed on the left lower leg and the right lower leg, respectively, and 3 is the upper unit;
  • a to D are gait events of the right leg in a gait cycle, where A is the mid-swing gait event, B is the foot landing gait event, and C is the standing mid-gait event, D For the foot off the ground gait event.
  • the present invention uses a wearable device comprising two inertial measurement sensors and an acceleration based algorithm to measure the step size of the user while walking.
  • a wearable device comprising two inertial measurement sensors and an acceleration based algorithm to measure the step size of the user while walking.
  • the target user wears a wearable device, and two inertial sensors are symmetrically placed on the left and right lower legs, and the distance d between the origin of the sensor coordinate axis and the sole is 13 cm.
  • both inertial sensor units include an inertial measurement sensor module based on the MPU6050 chip.
  • the module includes a three-dimensional accelerometer and a three-dimensional gyroscope for collecting three-dimensional acceleration and three-dimensional angular velocity during the user's walking. Data, sampling frequency is 100Hz.
  • the two sensor units are respectively fixed to the outside of the user's two lower legs by a resilient strap.
  • the axis of the sensor placed on the lower leg should be consistent with the calf to collect data for the corresponding axis.
  • the upper computer unit structure includes a single chip microcomputer, a button, a battery, and an SD card storage module.
  • the MCU communicates with two inertial sensors through the IIC communication protocol, controls them to sample and collect the acceleration and angular velocity data collected by them, and then stores them into the SD card.
  • the sensor data used in subsequent processes are the data stored in the SD card. And based on MATLAB to write the corresponding program to complete the subsequent calculation process.
  • the X axis is parallel to the calf; the Y axis is in the sagittal plane, perpendicular to the calf; the Z axis is perpendicular to the sagittal plane), and the sensor detects the data before use.
  • Dynamic correction and static correction are required.
  • Subsequent steps require step size calculations using the unfiltered X, Y-axis acceleration of the modified sensor unit, the filtered Y-axis acceleration data, and the unfiltered Z-axis angular velocity data.
  • the dynamic correction is to continuously correct the three-axis data of the X, Y and Z axes of the sensor on the calf by using the swing of the lower leg during the walking of the human body.
  • the sensor placement position is prone to deviation, so that the X and Y axes deviate from the sagittal plane, causing the X, Y, and Z axis deviations of the sensor. Shift, so you need to determine the angle of the X and Y axes off the sagittal plane Dynamic correction is the first correction of sensor data correction, as follows:
  • the straight-line walking process within a certain distance of the target user is monitored in advance, assuming that the calf moves only in the sagittal plane, that is, the calf has an angular velocity only in the Z-axis direction thereof, thereby determining the X of each sensor.
  • the angle of the Y-axis from the sagittal plane is a certain distance of the target user.
  • the angle at which the actual direction of the X-axis of the sensor deviates from the sagittal plane The angle at which the actual direction of the Y-axis of the sensor deviates from the sagittal plane,
  • the average value of the X-axis angular velocity of the sensor when the user is in the standing phase during the walking process (approximate time period in which the angular velocity of the calf Z-axis is negative)
  • the average value of the angular velocity of the Y-axis of the sensor during standing phase during the user's walking The average value of the Z-axis angular velocity of the sensor during standing phase during the user's walking.
  • the above average values can be obtained by averaging data of a plurality of standing phase periods during a straight walking process.
  • the triaxial acceleration measured by each sensor and the triaxial angular velocity data are offset from the sagittal plane by the X and Y axes before use.
  • Dynamic correction is performed in advance.
  • the correction process can be understood as changing the spatial coordinate system of the sensor so that the X and Y axes are placed in the sagittal plane, that is, the Y axis is used as the rotation axis, and the X axis is corrected to the sagittal shape by the rotation of the coordinate system in the XZ plane.
  • In-plane (as shown in Figure 9, where X, Y, and Z are the sensor's original coordinate axes, X 1 and Z' are the corrected X and Z coordinate axes), and then the corrected X-axis is the rotation axis.
  • the Y-axis is corrected into the sagittal plane by the rotation of the coordinate system in the YZ plane, so that the corrected X-axis and Y-axis are both in the sagittal plane (as shown in FIG.
  • Y 1 and Z 1 are The corrected Y and Z coordinate axes), because the X, Y, and Z axes have a spatial relationship (that is, perpendicular to each other), so the corrected Z axis is perpendicular to the sagittal plane.
  • the formula for calculating the acceleration and angular velocity of each axis after correction is:
  • a x , a y , a z are the X, Y, and Z axis accelerations measured by the sensor
  • a x1 , a y1 , and a z1 are the X, Y, and Z axes obtained by the sensor's measurement data after the correction.
  • Acceleration; ⁇ x , ⁇ y , ⁇ z are the angular velocities of the X, Y, and Z axes measured by the sensor
  • ⁇ x1 , ⁇ y1 , ⁇ z1 are the angular velocities of the X, Y, and Z axes obtained by the sensor's measurement data after the correction.
  • the static correction is the second correction after the dynamic correction, and the data of the sensor is corrected by the standing state of the human body.
  • the first dynamic correction although the corrected X and Y axes are transferred to the sagittal plane, the Z axis is transferred to a position perpendicular to the sagittal plane, but there are also deviations in the X and Y axes in the sagittal plane. Move, so a correction is needed.
  • the static correction is as follows:
  • the target user's static standing state is monitored in advance, and the two lower legs of the human body are considered to be perpendicular to the ground, and the state determines the angle of each sensor after the first correction of the X axis from the lower leg, that is, The angle between the X-axis and the plumb line after the first correction:
  • a ys1 is the Y-axis acceleration after the first correction measured by the sensor when the user is standing still
  • g is the gravitational acceleration
  • the three-axis acceleration and triaxial angular velocity data obtained by each sensor after the first dynamic correction need to be statically corrected before use, that is, on the basis of the first correction
  • the corrected Z axis is the rotation axis
  • the X axis is corrected to be parallel with the calf by the rotation of the coordinate system in the XY plane
  • the Y axis is perpendicular to the calf due to the spatial relationship of the X and Y axes (as shown in FIG. 11 )
  • X c , Y c , and Z c are the corrected X, Y, and Z coordinate axes)
  • the acceleration and angular velocity calculation formulas of each coordinate axis after static correction are as follows:
  • a xc , a yc , a zc are the X, Y, and Z axis accelerations obtained by statically correcting the measured data of the sensor;
  • ⁇ xc , ⁇ yc , and ⁇ zc are X obtained by statically correcting the data of the sensor.
  • a xc , a yc , a zc are the X, Y, and Z axis accelerations of the calf
  • ⁇ xc , ⁇ yc , and ⁇ zc are the angular velocity of the X, Y, and Z axes of the calf.
  • the walking gait cycle of a person is shown in Figure 4.
  • the gait event to be detected in this method is the foot. Landing and standing in the middle.
  • the Z-axis angular velocity of the lower leg can detect gait events in each gait cycle of the user's leg, as shown in FIG.
  • the foot landing event is the moment of landing before the forefoot during the walking process, occurring at the first negative trough in the angular velocity vibration region after the highest peak of the calf angular velocity in each gait cycle, that is, the first after the highest peak Negative turning point, and can be detected immediately after it occurs;
  • the middle of the standing is the leg as the supporting leg, moving to the position close to the vertical and the ground, at the small short peak of the angular velocity of the lower leg, and the peak value is negative .
  • the angular acceleration of the lower leg is close to 0, the sole of the foot is in contact with the ground, and the calf is in a state of constant rotation with the sole of the foot as the center of rotation, as shown in Fig. 7, so the acceleration of gravity can be utilized in Y.
  • the component of the axis calculates the angle between the lower leg and the vertical line (the following formula for calculating the left or right lower leg is not calculated at the same time, but is selected according to the current target to be tested. For example, when the left lower leg is detected In the middle of standing, ⁇ lms is calculated; when the right lower leg is detected to be in the middle of standing, ⁇ rms is calculated.
  • the formula in the subsequent steps is also similar):
  • ⁇ lms and ⁇ rms are the angles between the left calf and the right calf at the mid-term of their standing and the plumb line, respectively.
  • a lyms and a ryms are respectively filtered by the left calf and the right calf at the middle of their standing.
  • the angle between the calf and the plumb line in the sagittal plane is calculated by angular velocity integration at various times before the next middle moment of standing:
  • t l , t r are the time of the last standing middle moment of the left and right calves at the current time
  • ⁇ l (t l ) are the left calf at time t l , right calf
  • ⁇ lz ( ⁇ ), ⁇ rz ( ⁇ ) are the instantaneous Z-axis angular velocity of the left lower leg and the right lower leg, respectively.
  • the acceleration of the calf in the horizontal advancing direction is calculated by using the X and Y axis acceleration of the calf, that is, the component of the X and Y axis acceleration in the horizontal advancing direction is calculated, and the acceleration of each moment is calculated by the following formula:
  • a hl (t l ) -a lx (t l ) ⁇ sin ⁇ l (t l )+a ly (t l ) ⁇ cos ⁇ l (t l )
  • a hr (t r ) -a rx (t r ) ⁇ sin ⁇ r (t r )+a ry (t r ) ⁇ cos ⁇ r (t r )
  • each filtered no acceleration a hl (t l), a hr (t r) are the left leg at the time t l, the right leg when the acceleration time t r in the horizontal forward direction, a lx (t l), a rx ( t r) are the left leg at the time t l time, right leg X-axis acceleration at the time t r time, a ly (t l), a ry (t r) are the left leg The Y-axis acceleration of the right calf at time t r at time t l .
  • the step can be calculated.
  • the specific calculation process is as follows:
  • the moving speed of the lower leg (sensor) in the horizontal advancing direction is calculated as the initial speed of the second acceleration of the subsequent acceleration:
  • v lms and v rms are the speeds of the left and right calves in the horizontal advancement direction in the middle of their standing
  • ⁇ lzms , ⁇ rzms are the Z-axis angular velocities of the left and right calves in the middle of their standing
  • d is the calf
  • the position of the upper sensor is located at a distance from the sole of the foot.
  • the sensor is regarded as a point on the lower leg. In the middle of each of the left or right calf, the sensor has a certain horizontal distance from the lateral sole. The calculation of the step should also be considered, so the calculation is performed.
  • This horizontal displacement can also be referred to as the initial displacement of the lower leg (sensor) in the horizontal advancement direction:
  • s lms and s rms are the initial displacements of the left and right calves in the horizontal advancement direction at the mid-term of their standing, that is, the horizontal distance of the sensor on the calf from the sole of the foot in the middle of standing.
  • v l (t l ) and v r (t r ) are the speeds of movement of the left lower leg at the time t l and the right lower leg at the time t r
  • a hl ( ⁇ ) is the instantaneous acceleration in the horizontal advancing direction of the left lower leg and the right lower leg, respectively.
  • T l and T r are the lengths of time between the two adjacent standing intermediate moments of the left and right calves, respectively, and s l and s r are respectively between the left and right calf sensors between two adjacent standing mid-term moments. Displacement distance.
  • s lc and s rc are the displacement distances between the two adjacent standing middle moments after the correction of the left and right calf sensors
  • v lmsn and v rmsn are respectively the left and right calves in the middle of the middle of the leg.
  • v l (T l) are the left leg at time T l, a right leg movement speed in the forward direction when the horizontal time T r.
  • the displacement of the sensor on the lower leg between the two adjacent middle stages can be obtained between the two adjacent standing stages.
  • the displacement that is, the step of the leg that is created between the two adjacent middle stages of standing, is also the distance between the adjacent points of the leg, calculated by:
  • LSDL, RSDL are the left leg step and the right leg step respectively
  • s lmsl and s rmsl are the initial displacements of the left and right calves in the middle of a standing position before the leg
  • s lmsn and s rmsn are respectively left
  • Step size is obtained by splitting step:
  • the thighs and calves on both sides of the human body are basically in a straight state, simplifying the lower limbs of the human body into a two-dimensional geometric model of the pendulum, and simplifying the user's motion into a plane motion in the sagittal plane, as shown in Fig. 6.
  • Simplify the legs into a rod and the hip joint is simplified as a hinge.
  • the angle between the whole leg and the vertical line can be approximately equal to the angle between the lower leg and the vertical line, thereby theoretically calculating the geometrical relationship between the two adjacent standing moments of the leg.
  • RSL L (sin ⁇ rrc -sin ⁇ lrc )
  • LSL is the left leg step length
  • RSL is the right leg step length
  • ⁇ llc and ⁇ rlc are the left and right of the left foot landing time occurring between the middle of the standing position and the middle of the standing position.
  • the angle between the lower leg and the plumb line, ⁇ lrc and ⁇ rrc are the angles between the left and right calves and the plumb line at the moment of landing of the right foot between the middle of the standing position and the middle of the second standing position.
  • L is the length of the entire leg.
  • the step size of the left leg and the right leg is obtained by splitting the step by using the two-leg step ratio estimated by the above geometric relationship.
  • the step size is equal to the sum of the left leg step and the right leg step of the previous step:
  • the right leg step has the same calculation method:
  • the user has no lower extremity dyskinesia, and a total of 26 steps are collected during the step measurement.
  • the root mean square error of all steps measured by the process is 3.2 cm, which is 5.6% of the actual average step size (57.2 cm) of the user.
  • the embodiments described above are only some of the preferred embodiments of the present invention, but are not intended to limit the present invention. Various changes and modifications can be made by those skilled in the art without departing from the spirit and scope of the invention.
  • the above embodiment may also use other algorithms or use other sensors to calculate the angle of the lower leg, such as using an accelerometer, a gyroscope, and a magnetic field sensor fusion to calculate the angles of the left and right lower legs by the Kalman filter algorithm, and then used for the horizontal advancement direction.
  • the acceleration on the solution is solved, and the step size is calculated by using the geometric model of Fig. 6 to calculate the step size.
  • the above-mentioned wearable device may also adopt other structures in the prior art or modify the device shown in the drawing, such as removing the wired connection in the original device, and transmitting data using wireless communication for more convenient use.
  • the host computer unit can also be in the form of a remote PC or the like.
  • the step of correcting the data collected by the sensor placed on the leg may be omitted, and the sensing is directly performed.
  • the acceleration and angular velocity data are used as the process of the step size measurement as the corresponding data of the lower leg.

Abstract

一种用于可穿戴式设备的人体步长测量方法，使用两个惯性传感器（1，2）单元采集并存储的人体行走过程中的加速度、角速度数据，随后利用相应算法计算使用者的步长信息，并可用于估计用户的步态不对称度，两个惯性传感器（1，2）分别放置在用户的两条小腿上，穿戴简单、方便、轻巧。一种实现人体步长测量方法的可穿戴设备，包括两个惯性传感器（1，2）和上位机（3），每个惯性传感器（1，2）包含三维加速度计以及三维角速度计，惯性传感器（1，2）与上位机（3）相连进行数据传输，上位机（3）用于控制两个惯性传感器（1，2）采集加速度、角速度数据并将数据收集和存储。

Description

一种用于可穿戴式设备的人体步长测量方法及其测量设备 技术领域
本发明属于可穿戴传感器领域，具体涉及一种用于可穿戴式设备的人体步长测量方法及其测量设备。
背景技术
步态参数是人类行走过程中的物理参数，可以反映人类运动能力，拥有较大的应用价值，因此目前有许多研究者在研究步态参数的测量。可穿戴传感器包括惯性测量单元，超声波传感器，微型摄像头等等，相比于大型实验室测量步态参数的设备如光学式运动捕捉系统、测力台等等，以其小巧、廉价，不受时间、空间限制、易于推广等优点被广泛应用于步态参数测量领域。目前有很多研究使用放置在人体腿部、腰部、手腕、胸部等部位的可穿戴传感器测量步态时间参数，如步态事件、步态周期等等；也有的测量步态空间参数，如步行速度、步距等等。如申请号为CN201510887154.1的发明专利公开了一种室内定位步长计算方法，将惯性测量单元置于人员的腰部，人员在室内行走过程中，获得人员惯性传感器数据，然后计算步长。而申请号为CN201310007945.1的发明专利公开了一种步长计算方法和装置，获取载体在三维坐标系中的各轴加速度值；根据所述各轴加速度值计算步长补偿系数；根据所述步长补偿系数和预先设定的预设步长确定载体移动幅度的最终步长。但上述方法再实际使用过程中，均存在步态测量精度较低、在不同人群中的适应能力差等缺陷。
步长的测定具有重要的意义。现在有很多研究用可穿戴传感器测量步速、步距等空间参数，比较成熟，但是测量步长的研究不多。一些疾病如帕金森综合征会使人体运动能力下降，步速、步距减小，因此步距和步速可以用来反映人体下肢运动能力。但是步距、步速不能反映出偏瘫造成的人体两侧运动能力的差异，即步态不对性，而我们可以根据两腿步长的比值量化人体两侧步态的不对称性，而且通过步长也可以求出步距、步速等参数，因此步长拥有更大的临床应用价值。现有的使用可穿戴设备测量步长的方法主要是利用腰部重心的上下位移或者利 用基于步频、加速度等变量的数学模型，均存在精度差，适应能力差等缺陷。因此有必要提出一种新型的高精度、在不同人群中适应能力强的步长测量方法。
发明内容
本发明的目的在于解决现有技术中步态测量精度较低、无法测量单步步长等缺陷，并提供一种用于可穿戴式设备的人体步长测量方法及其测量设备。
本发明中所涉及的部分名词含义如下：
步距是指人行走过程中同一只脚相邻着地点之间的距离，左腿步距是指行走过程中左脚相邻着地点之间的距离，右腿步距是指行走过程中右脚相邻着地点之间的距离；步长是指人行走过程中双脚的相邻着地点之间的距离，左腿步长是指行走过程中左脚着地点在前的步长，右腿步长是指行走过程中右脚着地点在前的步长。通常来说，步距等于两腿步长之和。
步态事件是指人在行走过程中的每个步态周期中的重要时刻，主要有摆动中期、脚落地、站立中期、脚离地四个步态事件。人的走路步态周期如图4所示，以右腿为例，左腿支撑人体，右腿向前迈出的这段时间为右腿的摆动相；右腿摆到左腿附近时为摆动中期；接着右脚落地，右腿开始支撑人体，左腿向前迈出的这段时间，为右腿的站立相；右腿以右踝关节为旋转中心，向前移动，到达接近于与地面垂直的位置，此时为站立中期；随后右脚离地，右腿向前迈出，完成一个步态周期。
步态事件可以该腿小腿的角速度特征进行检测，如图5所示。小腿角速度在一个周期内通常主要有一高一矮的两个比较大的波峰，较高的峰对应着摆动中期，较矮的对应着站立中期；摆动中期之后有一段有很多小的负向的波峰的区域，这是脚落地造成的震动，第一个负向峰为脚落地时刻；在站立中期后，下一个摆动中期前，有一处波谷，此时对应着脚离地时刻。站立相从该腿脚落地开始，到脚离地结束，可以近似的认为图中小腿角速度为负值的区域为站立相；摆动相从该腿脚离地开始，到脚落地结束，可以近似的认为图中小腿角速度为正值的区域为摆动相。
矢状面和铅垂线如图1所示，矢状面是垂直于水平面并将人体分为左右两部分的面，人在行走时，主要运动都发生在矢状面内；铅垂线是垂直于地面的线。
水平前进方向在图2中有标注，指的是矢状面内，人体沿水平方向前进的方 向。小腿在水平前进方向上的加速度、运动速度，分别指小腿的加速度、运动速度在水平前进方向上的分量。
另外为方便描述，定义小腿的三维方向，如图2所示：X轴与该小腿平行；Y轴在矢状面内，与该小腿垂直；Z轴垂直于矢状面。放置在小腿上的传感器的坐标轴应该与该小腿保持一致，但是因为矢状面是不可见的，另外小腿也不是规则的圆柱体，在安放传感器时会出现X轴、Y轴、Z轴偏移的情况，根据传感器测得的X轴、Y轴、Z轴角速度可以大致计算传感器X、Y、Z三轴偏移角度，并利用此角度计算该小腿的相应轴的加速度、角速度。
传感器到脚底距离为：用户静止站立时，传感器坐标轴原点到地面的高度。
本发明为解决技术问题，所采用的具体技术方案如下：
用于可穿戴式设备的人体步长测量方法，包括以下步骤：
通过在用户两侧小腿上安装传感器，测量两侧小腿行走过程中的X、Y轴加速度和Z轴角速度，根据Z轴角速度检测行走过程中的步态事件；根据步态事件以及Z轴角速度积分确定用户行走过程中的各个时刻小腿在矢状面内与铅垂线的夹角；根据该夹角利用X、Y轴加速度计算小腿在水平前进方向上的加速度；根据步态事件对该水平前进方向上加速度进行二次积分并根据传感器位于小腿上的位置关系计算步距；通过利用几何关系估算的步长比例拆分步距，得到左腿和右腿的步长。
基于该方案，本发明还提供了几种优选实现方式，而且各优选方式中的技术特征，若没有冲突均可进行相互组合。
作为一种优选方式，上述所说的可穿戴式设备中包含用于检测左、右小腿X、Y、Z三轴加速度以及X、Y、Z三轴角速度的传感器，左、右小腿上各一个，用户两侧小腿行走过程中的X、Y轴加速度和Z轴角速度通过该传感器获得。
作为另一种优选方式，上述人体步长测量方法中，部分步骤可采用如下具体方式实现：
(1)确定用户行走过程中的各个时刻小腿在矢状面内与铅垂线的夹角具体方法为：
根据Z轴角速度检测行走过程中的步态事件，包括脚落地以及站立中期；
在左小腿或右小腿的各个站立中期时刻，计算该小腿与铅垂线的夹角(下述 左小腿或右小腿的计算公式，并非同时进行计算，而是根据当前待测的目标择一选用。例如，当检测到左小腿处于站立中期，则计算θlms；当检测到右小腿处于站立中期，则计算θrms。后续步骤中的公式也采用类似方法)：
Figure PCTCN2017071674-appb-000001
Figure PCTCN2017071674-appb-000002
式中：θlms、θrms分别为左小腿、右小腿在其站立中期时刻与铅垂线的夹角，alyms、aryms分别为左小腿、右小腿在其站立中期时刻的Y轴加速度；
在到下一个站立中期时刻之前的各个时刻，通过角速度积分计算该小腿在矢状面内与铅垂线的夹角：
Figure PCTCN2017071674-appb-000003
Figure PCTCN2017071674-appb-000004
式中：tl、tr分别为当前时刻距离左、右小腿的上一个站立中期时刻的时间，θl(tl)、θr(tr)分别为左小腿在tl时刻、右小腿在tr时刻在矢状面内与铅垂线的夹角，ωlz(δ)、ωrz(δ)分别为左小腿、右小腿瞬时Z轴角速度；
由此可以得到用户行走过程中的各个时刻的两小腿与铅垂线的夹角。
(2)计算小腿在水平前进方向上的加速度的具体方法为：
利用X、Y轴加速度计算两小腿在水平前进方向上的加速度，每个时刻的加速度均通过下式计算：
ahl(tl)＝-alx(tl)·sinθl(tl)+aly(tl)·cosθl(tl)
ahr(tr)＝-arx(tr)·sinθr(tr)+ary(tr)·cosθr(tr)
式中，ahl(tl)、ahr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时在水平前进方向上的加速度，alx(tl)、arx(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的X轴加速度，aly(tl)、ary(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的Y轴加速度。
(3)所述计算步距的具体方法为：
在左小腿或右小腿的各个站立中期时刻，计算该小腿此时刻在水平前进方向上的运动速度：
vlms＝-d·ωlzms·cosθlms
vrms＝-d·ωrzms·cosθrms
式中，vlms、vrms为左、右小腿在其站立中期时刻在水平前进方向的运动速度，ωlzms、ωrzms为左、右小腿在其站立中期时刻的Z轴角速度，d为该小腿上用于测量加速度、角速度的传感器的摆放位置距离脚底的距离；
在左小腿或右小腿的各个站立中期时刻，计算该小腿的在水平前进方向上的初始位移：
slms＝-d·sinθlms
srms＝-d·sinθrms
式中，slms、srms为左、右小腿在其站立中期时刻在水平前进方向上的初始位移；
在到下一个站立中期时刻之前的各个时刻，该小腿的水平前进方向上的运动速度通过该方向上的加速度积分而得：
Figure PCTCN2017071674-appb-000005
Figure PCTCN2017071674-appb-000006
式中，vl(tl)、vr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时水平前进方向上的运动速度，ahl(δ)、ahr(δ)分别为左小腿、右小腿瞬时的水平前进方向上的加速度；
该小腿在相邻两个站立中期时刻之间产生的该腿的步距，通过下式计算：
Figure PCTCN2017071674-appb-000007
Figure PCTCN2017071674-appb-000008
式中：LSDL、RSDL分别为左腿步距、右腿步距，slmsl、srmsl分别为左、右小腿在该腿前一个站立中期时刻的水平前进方向上的初始位移，slmsn、srmsn分别为左、右小腿在该腿后一个站立中期时刻的水平前进方向上的初始位移，Tl、Tr分别为左、右小腿的相邻两个站立中期时刻之间的时间长度，vlmsn、vrmsn分别为左、右小腿在该腿后一个站立中期时刻的水平前进方向上的运动速度，vl(Tl)、vr(Tr)分别为左小腿在Tl时刻时、右小腿在Tr时刻时通过加速度积分计算的水平 前进方向上的运动速度；
(4)拆分步距得到步长的具体方法为：
该小腿相邻两个站立中期时刻之间产生的该腿的步长，通过下式计算：
Figure PCTCN2017071674-appb-000009
Figure PCTCN2017071674-appb-000010
式中，LSL为左腿步长，RSL为右腿步长，θllc、θrlc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的左脚落地时刻的左、右小腿与铅垂线的夹角，θlrc、θrrc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的右脚落地时刻的左、右小腿与铅垂线的夹角。
作为另一种优选方式，测量步长过程中的，每个传感器的X、Y轴加速度、Z轴角速度数据在作为该小腿相应轴的加速度、角速度数据代入上述各公式前，都需要预先进行修正，修正的方法如下：
测量步长之前，预先监测目标用户的一定距离内的直线行走过程，确定每个传感器的X、Y轴实际方向偏离矢状面的角度：
Figure PCTCN2017071674-appb-000011
Figure PCTCN2017071674-appb-000012
式中：
Figure PCTCN2017071674-appb-000013
为该传感器X轴实际方向偏离矢状面的角度，
Figure PCTCN2017071674-appb-000014
为该传感器Y轴实际方向偏离矢状面的角度，
Figure PCTCN2017071674-appb-000015
为用户行走过程中在站立相时该传感器X轴角速度的平均值，
Figure PCTCN2017071674-appb-000016
为用户行走过程中在站立相时该传感器Y轴角速度的平均值，
Figure PCTCN2017071674-appb-000017
为用户行走过程中在站立相时该传感器Z轴角速度的平均值；
在测量步长过程中，每个传感器测得的三轴加速度，三轴角速度数据在使用前都预先进行第一次修正，修正公式为：
Figure PCTCN2017071674-appb-000018
Figure PCTCN2017071674-appb-000019
式中：ax、ay、az为该传感器测量的X、Y、Z轴加速度，ax1、ay1、az1为 传感器的测量数据经该次修正后得到的X、Y、Z轴加速度；ωx、ωy、ωz为该传感器测量的X、Y、Z轴角速度，ωx1、ωy1、ωz1为传感器的测量数据经该次修正后得到的X、Y、Z轴角速度；
测量步长之前，还需预先监测目标用户的静止站立状态，确定此状态下每个传感器X轴实际方向偏离铅垂线的角度：
Figure PCTCN2017071674-appb-000020
式中：
Figure PCTCN2017071674-appb-000021
为静止站立状态下该传感器X轴实际方向偏离铅垂线的角度，ays1为用户静止站立时该传感器测得的经过第一次修正后的Y轴加速度，g为重力加速度；
在测量步长过程中，每个传感器经过第一次修正后得到的三轴加速度、三轴角速度数据，在使用前还需进行第二次修正，修正公式如下：
Figure PCTCN2017071674-appb-000022
Figure PCTCN2017071674-appb-000023
式中：axc、ayc、azc为传感器的测量数据经第二次修正后得到的X、Y、Z轴加速度；ωxc、ωyc、ωzc为传感器的数据经第二次修正后得到的X、Y、Z轴角速度；
在测量步长过程中，axc、ayc、azc即为所述的小腿X、Y、Z轴加速度，ωxc、ωyc、ωzc即为所述的小腿X、Y、Z轴角速度。
本发明的另一目的是提供一种实现上述人体步长测量方法的可穿戴式设备，包括两个惯性传感器和上位机，每个惯性传感器包含三维加速度计以及三维角速度计，惯性传感器与上位机相连进行数据传输，上位机用于控制两个惯性传感器采集加速度、角速度数据，并将数据收集和存储。
作为一种优选方式，还包括用于固定惯性传感器的固定带。
作为一种优选方式，惯性传感器为基于MPU6050芯片的惯性传感器。
作为一种优选方式，惯性传感器采样频率不低于100Hz。
作为一种优选方式，上位机将惯性传感器测量的数据存储入SD卡中。
上述各优选方式中的技术特征在不相互冲突的前提下，均可进行相互组合，不构成限制。
本发明相对于现有技术而言，其有益效果是：
1)使用本发明计算下肢步长，廉价、方便，不受场地限制，易于推广。
2)使用一种基于小腿加速度积分的方法计算步长，可以适应多种病态的步态，拥有较高的精度、较好的应用价值以及广泛的应用范围。
3)通过可穿戴式设备进行步长测量，可以方便、有效的量化人体两侧运动能力以及它们的差异，可以方便应用于临床。
本发明中部分步骤的具体效果将通过后续的具体实施方式进行详细说明。
附图说明
图1本发明中矢状面、铅垂线示意图；
图2本发明中传感器放置位置，小腿坐标系及水平前进方向示意图；
图3本发明中用于实现测量步长方法的可穿戴设备结构图；
图4本发明中人体行走周期示意图；
图5本发明中步态事件检测示意图；
图6本发明中下肢单摆二维几何模型示意图；
图7本发明中右小腿摆动中期角度标定示意图；
图8本发明中右小腿传感器摆动中期速度以及位移的计算；
图9、10、11本发明中传感器坐标系的修正示意图；
上述图1、2、7、8、9、10、11中P表示铅垂线(Plumb line)，S表示矢状面(Sagittal Plane)，H表示水平前进方向(Horizontal Direction of Progression)；
上述图2、3中，1、2分别为放置在左小腿、右小腿上的惯性传感器单元，3为上位机单元；
上述图4、5中，A～D为右腿在一个步态周期内的步态事件，其中A为摆动中期步态事件，B为脚落地步态事件，C为站立中期步态事件，D为脚离地步态事件。
具体实施方式
下面结合附图对本发明进行进一步说明，因便于更好地理解。本发明以下实施例仅用于提供一种优选的方式，但其中技术特征在不相互冲突的前提下，均可进行相互组合，不构成对本发明保护范围的限制。
本发明使用包括两个惯性测量传感器的可穿戴设备以及一种基于加速度积分的算法，测量使用者在行走时的步长。以某一用户为例，本发明具体实施过程如下：
(1)准备工作：
本实施例中，目标用户穿戴有可穿戴设备，两个惯性传感器对称放置在左右小腿上，传感器坐标轴原点至脚底距离d为13cm。
整套设备结构如图3所示，包含两个惯性传感器单元(左小腿惯性传感器1、右小腿惯性传感器2)以及一个上位机单元3。本发明中各传感器及其他电子元件的具体型号，可以根据实际需要进行选型。在本实施例中，两个惯性传感器单元中都包含一个基于MPU6050芯片的惯性测量传感器模块，模块包括一个三维加速度计以及一个三维陀螺仪，用于采集使用者行走过程中的三维加速度以及三维角速度数据，采样频率为100Hz。两个传感器单元分别通过带弹性的固定带固定在用户的两条小腿外侧。定义小腿的三维方向，如图2所示：X轴与该小腿平行；Y轴在矢状面内，与该小腿垂直；Z轴垂直于矢状面。放置在小腿上的传感器的坐标轴应该与该小腿保持一致，以用来采集相应坐标轴的数据。上位机单元结构上包括单片机、按键、电池以及SD卡存储模块。单片机通过IIC通信协议与两个惯性传感器通信，控制它们采样并收集它们采集的加速度、角速度数据，随后存储入SD卡中，后续各个过程中使用的传感器数据均为存入SD卡中的数据，并基于MATLAB编写相应程序，完成后续计算过程。
由于传感器放置的位置通常不能达到标准的理想状态(X轴与该小腿平行；Y轴在矢状面内，与该小腿垂直；Z轴垂直于矢状面)，传感器检测到的数据在使用前需要先后经过动态修正和静态修正。为了减少行走过程中传感器震动带来的误差，需要再使用截止频率为3Hz左右的低通滤波器对修正后的Y轴加速度进行滤波，以便后面步骤使用。后续步骤需要使用修正过后的传感器单元的未滤波的X、Y轴加速度、滤波后的Y轴加速度数据以及未滤波的Z轴角速度数据进行步长计算。
动态修正是利用人体行走过程中小腿的摆动对小腿上的传感器的X、Y、Z轴三轴数据不断进行修正。在用户安放传感器时，由于矢状面是不可见的，小腿不是规则的圆柱体形状，传感器放置位置容易出现偏离，使得X、Y轴偏离矢状 面，造成传感器的X、Y、Z轴偏移，因此需要确定X、Y轴偏离矢状面的角度
Figure PCTCN2017071674-appb-000024
Figure PCTCN2017071674-appb-000025
动态修正是传感器数据修正的第一次修正，具体如下：
测量步长之前，预先监测目标用户的一定距离内的直线行走过程，假定该过程小腿仅在矢状面内运动，即小腿仅在其Z轴方向上有角速度，由此确定每个传感器的X、Y轴实际方向偏离矢状面的角度：
Figure PCTCN2017071674-appb-000026
Figure PCTCN2017071674-appb-000027
式中：
Figure PCTCN2017071674-appb-000028
为该传感器X轴实际方向偏离矢状面的角度，
Figure PCTCN2017071674-appb-000029
为该传感器Y轴实际方向偏离矢状面的角度，
Figure PCTCN2017071674-appb-000030
为用户行走过程中在站立相(近似的取小腿Z轴角速度为负值的时段)时该传感器X轴角速度的平均值，
Figure PCTCN2017071674-appb-000031
为用户行走过程中在站立相时该传感器Y轴角速度的平均值，
Figure PCTCN2017071674-appb-000032
为用户行走过程中在站立相时该传感器Z轴角速度的平均值。上述各平均值可通过直线行走过程中，多个站立相时段的数据取平均后得到。
在测量步长过程中，每个传感器测得的三轴加速度，三轴角速度数据在使用前都利用X、Y轴偏离矢状面的角度
Figure PCTCN2017071674-appb-000033
预先进行动态修正。该修正过程可以理解为变换传感器的空间坐标系从而将X、Y轴放置在矢状面内，即先以Y轴为旋转轴，通过坐标系在X-Z平面内的旋转将X轴修正到矢状面内(如图9所示，其中X、Y、Z为传感器原始坐标轴，X1、Z'为经过该修正后的X、Z坐标轴)，再以修正后的X轴为旋转轴，通过坐标系在Y-Z平面内的旋转将Y轴修正到矢状面内，使得修正后的X轴、Y轴均在矢状面内(如图10所示，其中Y1、Z1为经过该修正后的Y、Z坐标轴)，因为X、Y、Z三轴存在空间关系(即互相垂直)，所以修正后的Z轴与矢状面垂直。修正后各轴加速度、角速度计算公式为：
Figure PCTCN2017071674-appb-000034
Figure PCTCN2017071674-appb-000035
式中：ax、ay、az为该传感器测量的X、Y、Z轴加速度，ax1、ay1、az1为传感器的测量数据经该次修正后得到的X、Y、Z轴加速度；ωx、ωy、ωz为该传感器测量的X、Y、Z轴角速度，ωx1、ωy1、ωz1为传感器的测量数据经该次修正后得到的X、Y、Z轴角速度。
静态修正是在动态修正后的第二次修正，通过人体静止站立状态对传感器的数据进行修正。经过第一次动态修正后，虽然修正过的X、Y轴被转移至矢状面内，Z轴被转移至垂直于矢状面的位置，但是在矢状面内X、Y轴还存在偏移，因此还需要一次修正。静态修正具体如下：
测量步长之前，预先监测目标用户的静止站立状态，并认为静止站立状态人体的两条小腿与地面垂直，由此状态确定每个传感器经第一次修正后的X轴偏离小腿的角度，即经第一次修正后的X轴与铅垂线的夹角：
Figure PCTCN2017071674-appb-000036
式中：
Figure PCTCN2017071674-appb-000037
为该传感器经第一次修正后的X轴与铅垂线的夹角，ays1为用户静止站立时该传感器测得的经过第一次修正后的Y轴加速度，g为重力加速度。
在之后的测量步长过程中，每个传感器经过第一次动态修正后得到的三轴加速度、三轴角速度数据，在使用前还需进行静态修正，即在第一次修正的基础上，以修正后的Z轴为旋转轴，通过坐标系在X-Y平面内的旋转修正X轴使之与小腿平行，同时由于X、Y轴的空间关系，Y轴与小腿垂直(如图11所示，其中其中Xc、Yc、Zc为经过该修正后的X、Y、Z坐标轴)，静态修正后的各个坐标轴的加速度、角速度计算公式如下：
Figure PCTCN2017071674-appb-000038
Figure PCTCN2017071674-appb-000039
式中：axc、ayc、azc为传感器的测量数据经静态修正后得到的X、Y、Z轴 加速度；ωxc、ωyc、ωzc为传感器的数据经静态修正后得到的X、Y、Z轴角速度。
在测量步长过程中，axc、ayc、azc即为小腿X、Y、Z轴加速度，ωxc、ωyc、ωzc即为小腿X、Y、Z轴角速度。
(2)基于Z轴角速度检测步态事件，并通过Z轴角速度积分计算左右小腿在各个时刻与铅垂线的夹角：
以上各项准备工作完成后，可以开始测量用户的步长。用户在平坦的地面上行走，两个惯性传感器采集加速度、角速度数据，并用于后续处理。
人的走路步态周期如图4所示，一个周期中主要有摆动中期A、脚落地B、站立中期C、脚离地D四个步态事件，本方法中需要检测的步态事件为脚落地以及站立中期。小腿的Z轴角速度可以检测用户该腿的每一个步态周期内的步态事件，如图5所示。脚落地事件是走路过程中前脚后落地的时刻，发生在每个步态周期内小腿角速度的最高波峰后的角速度震动区域中的第一个负值波谷处，即在最高波峰后的第一个负值转折点，并且在其发生后能够被立刻检测；站立中期为该腿作为支撑腿，移动到接近垂直与地面的位置的时刻，位于小腿角速度的较大的矮波峰处，且峰值为负值。
在左小腿或右小腿的各个站立中期时刻，小腿角加速度接近于0，脚底与地面接触，小腿处于以脚底为旋转中心做匀速转动的状态，如图7所示，因此可以利用重力加速度在Y轴的分量计算该小腿与铅垂线的夹角(下述左小腿或右小腿的计算公式，并非同时进行计算，而是根据当前待测的目标择一选用。例如，当检测到左小腿处于站立中期，则计算θlms；当检测到右小腿处于站立中期，则计算θrms。后续步骤中的公式也采用类似方法)：
Figure PCTCN2017071674-appb-000040
Figure PCTCN2017071674-appb-000041
式中：θlms、θrms分别为左小腿、右小腿在其站立中期时刻与铅垂线的夹角，alyms、aryms分别为左小腿、右小腿在其站立中期时刻的经滤波后的Y轴加速度。
在到下一个站立中期时刻之前的各个时刻，通过角速度积分计算该小腿在矢状面内与铅垂线的夹角：
Figure PCTCN2017071674-appb-000042
Figure PCTCN2017071674-appb-000043
式中：tl、tr分别为当前时刻距离左、右小腿的上一个站立中期时刻的时间，θl(tl)、θr(tr)分别为左小腿在tl时刻、右小腿在tr时刻在矢状面内与铅垂线的夹角，ωlz(δ)、ωrz(δ)分别为左小腿、右小腿瞬时Z轴角速度。
由此可以得到用户行走过程中的各个时刻的两小腿与铅垂线的夹角。
(3)计算各个时刻两小腿在水平前进方向上的加速度：
利用小腿X、Y轴加速度计算该小腿在水平前进方向上的加速度，即计算X、Y轴加速度在水平前进方向上的分量和，每个时刻的加速度均通过下式进行相应的计算：
ahl(tl)＝-alx(tl)·sinθl(tl)+aly(tl)·cosθl(tl)
ahr(tr)＝-arx(tr)·sinθr(tr)+ary(tr)·cosθr(tr)
式中，各加速度均未经过滤波，ahl(tl)、ahr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时在水平前进方向上的加速度，alx(tl)、arx(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的X轴加速度，aly(tl)、ary(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的Y轴加速度。
(4)根据步态事件对水平前进方向上加速度进行二次积分计算步距：
通过对相邻两个站立中期之间的时段的水平前进方向加速度进行二次积分，再结合传感器在小腿上的位置，可以计算出步距，具体计算过程如下：
如图8所示，在左小腿或右小腿的各个站立中期时刻，计算该小腿(传感器)此时刻的水平前进方向上的运动速度，作为后续加速度第二次积分的初始速度：
vlms＝-d·ωlzms·cosθlms
vrms＝-d·ωrzms·cosθrms
式中，vlms、vrms为左、右小腿在其站立中期时刻在水平前进方向的运动速度，ωlzms、ωrzms为左、右小腿在其站立中期时刻的Z轴角速度，d为该小腿上的传感器的摆放位置距离脚底的距离。
如图8所示，将传感器视为小腿上的一点，在左小腿或右小腿的各个站立中期时刻，传感器与该侧脚底有一定的水平距离，计算步距时也应考虑，因此计算 该水平位移，也可称为该小腿(传感器)的在水平前进方向上的初始位移：
slms＝-d·sinθlms
srms＝-d·sinθrms
式中，slms、srms为左、右小腿在其站立中期时刻在水平前进方向上的初始位移，即站立中期时该小腿上传感器距离脚底的水平距离。
在到下一个站立中期时刻之前的各个时刻，该小腿(传感器)的水平前进方向上的运动速度通过该方向上的加速度积分而得：
Figure PCTCN2017071674-appb-000044
Figure PCTCN2017071674-appb-000045
式中，vl(tl)、vr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时水平前进方向上的运动速度，ahl(δ)、ahr(δ)分别为左小腿、右小腿瞬时的水平前进方向上的加速度。
通过加速度积分可以得到该小腿上传感器在相邻两个站立中期时刻之间的位移距离：
Figure PCTCN2017071674-appb-000046
Figure PCTCN2017071674-appb-000047
式中：Tl、Tr分别为左、右小腿的两个相邻站立中期时刻之间的时间长度，sl、sr分别为左、右小腿传感器两个相邻站立中期时刻之间的位移距离。
再两个相邻站立中期时刻之前，从前一个站立中期积分计算的在后一个站立中期时该小腿在水平前进方向上的速度，与后一个站立中期时直接计算的初速度之间存在一定差距，因此对传感器加速度积分得的位移距离有一定的误差，需要利用这个速度差距进行修正：
Figure PCTCN2017071674-appb-000048
Figure PCTCN2017071674-appb-000049
式中：slc、src分别为左、右小腿传感器修正后的两个相邻站立中期时刻之 间的位移距离，vlmsn、vrmsn分别为左、右小腿在该腿后一个站立中期时刻的水平前进方向上的运动速度，vl(Tl)、vr(Tr)分别为左小腿在Tl时刻时、右小腿在Tr时刻时水平前进方向上的运动速度。
再利用前文所述的站立中期时传感器与脚底的水平距离(小腿初始位移)，可以通过两个相邻站立中期之间小腿上传感器的位移得到该腿脚底在两个相邻站立中期之间的位移，即在两个相邻站立中期之间产生的该腿的步距，也是该腿相邻着地点之间的距离，通过下式计算：
Figure PCTCN2017071674-appb-000050
Figure PCTCN2017071674-appb-000051
式中：LSDL、RSDL分别为左腿步距、右腿步距，slmsl、srmsl分别为左、右小腿在该腿前一个站立中期时刻的初始位移，slmsn、srmsn分别为左、右小腿在该腿后一个站立中期时刻的初始位移。
(5)拆分步距得到步长：
在脚落地时刻，人体两侧大腿、小腿基本处于伸直状态，将人体下肢简化为单摆二维几何模型，并将用户的运动简化为矢状面内的平面运动，如图6所示，将两腿简化为杆，髋关节简化为铰链。
根据图6，整条腿与铅垂线的夹角可以近似等于小腿与铅垂线的夹角，由此从理论上可以跟据几何关系近似计算该腿相邻两个站立中期时刻之间产生的步长：
LSL＝L(sinθllc-sinθrlc)
RSL＝L(sinθrrc-sinθlrc)
式中，LSL为左腿步长，RSL为右腿步长，θllc、θrlc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的左脚落地时刻的左、右小腿与铅垂线的夹角，θlrc、θrrc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的右脚落地时刻的左、右小腿与铅垂线的夹角，L为整条腿的长度。
但是，这样近似的计算会引入较大误差，因为在脚落地时刻腿并不是完全伸直的，因此这样计算得出步长不可取。但是左右腿步长的比例却没有很大误差，可以加以利用，本实施例中通过利用上述几何关系估算的两腿步长比例拆分步距，得到左腿和右腿的步长。。根据图6，对于左腿，步距等于左腿步长与之前一步的右腿步长之和：
LSDL＝LSL+RSL
因而我们可以根据左右腿步长的比例准确拆分左腿步距计算左腿步长：
Figure PCTCN2017071674-appb-000052
而右腿步长也有相同的计算方式：
Figure PCTCN2017071674-appb-000053
(6)步长测量效果验证：
通过上述步骤(1)～(5)完成步长测量后，我们利用实测数据进一步验证了测量精度。
本例中用户无下肢运动障碍，在步长测量过程中一共采集了26步。通过与每一步的步长的实际值对比，该过程测量的所有步长的均方根误差为3.2cm，为该用户实际平均步长(57.2cm)的5.6％。由此可见，本发明的方法相对于现有技术而言，拥有较高的精度；另一方面，由于使用加速度二次积分获得距离，本方法受个人行走习惯影响较小，因此拥有较好的在不同人群中的适应性。
以上所述的实施例只是本发明的一些较佳的方案，然而其并非用以限制本发明。有关技术领域的普通技术人员，在不脱离本发明的精神和范围的情况下，还可以做出各种变化和变型。例如，上述实施例也可以使用其他算法或者使用其他传感器来计算小腿的角度，如使用加速度计、陀螺仪、磁场传感器融合进行卡尔曼滤波算法计算左、右小腿的角度，进而用于水平前进方向上的加速度求解，并利用如图6的几何模型进行步距拆分，计算步长。而上述可穿戴设备也可以采用现有技术中的其他结构或对附图中展示的设备进行改动，如去除原设备中的有线连接，使用无线通讯传输数据，以更方便地使用。上位机单元也可以采用远程的PC机等形式。另外，在本发明的方法中，在传感器安装角度和位置准确的情况下，还可以省略对放置在腿上的传感器采集的数据进行校正的步骤，直接将传感 器加速度、角速度数据作为小腿的相应数据用于步长测量的过程。
由此可见，凡采取等同替换或等效变换的方式所获得的技术方案，均落在本发明的保护范围内。

Claims (10)

  1. 一种用于可穿戴式设备的人体步长测量方法，其特征在于，包括以下步骤：通过在用户两侧小腿上安装传感器，测量两侧小腿行走过程中的X、Y轴加速度和Z轴角速度，根据Z轴角速度检测行走过程中的步态事件；根据步态事件以及Z轴角速度积分确定用户行走过程中的各个时刻小腿在矢状面内与铅垂线的夹角；根据所述的夹角利用X、Y轴加速度计算小腿在水平前进方向上的加速度；根据步态事件对所述水平前进方向上加速度进行二次积分并根据传感器位于小腿上的位置关系计算步距；通过利用几何关系估算的步长比例拆分步距，得到左腿和右腿的步长。
  2. 如权利要求1所述的用于可穿戴式设备的人体步长测量方法，其特征在于，所述的可穿戴式设备中包含用于检测左、右小腿X、Y、Z三轴加速度以及X、Y、Z三轴角速度的传感器，左、右小腿上各一个，用户两侧小腿行走过程中的X、Y轴加速度和Z轴角速度通过该传感器获得。
  3. 如权利要求1所述的基于可穿戴式设备的人体步长测量方法，其特征在于：
    (1)确定用户行走过程中的各个时刻小腿在矢状面内与铅垂线的夹角具体方法为：
    所需要检测的步态事件包括脚落地以及站立中期；
    在左小腿或右小腿的各个站立中期时刻，计算该小腿与铅垂线的夹角：
    Figure PCTCN2017071674-appb-100001
    Figure PCTCN2017071674-appb-100002
    式中：θlms、θrms分别为左小腿、右小腿在其站立中期时刻与铅垂线的夹角，alyms、aryms分别为左小腿、右小腿在其站立中期时刻的Y轴加速度；
    在到下一个站立中期时刻之前的各个时刻，通过角速度积分计算该小腿在矢状面内与铅垂线的夹角：
    Figure PCTCN2017071674-appb-100003
    Figure PCTCN2017071674-appb-100004
    式中：tl、tr分别为当前时刻距离左、右小腿的上一个站立中期时刻的时间，θl(tl)、θr(tr)分别为左小腿在tl时刻、右小腿在tr时刻在矢状面内与铅垂线的夹角，ωlz(δ)、ωrz(δ)分别为左小腿、右小腿瞬时Z轴角速度；
    由此可以得到用户行走过程中的各个时刻的两小腿与铅垂线的夹角；
    (2)计算小腿在水平前进方向上的加速度的具体方法为：
    利用X、Y轴加速度计算两小腿在水平前进方向上的加速度，每个时刻的加速度均通过下式计算：
    ahl(tl)＝-alx(tl)·sinθl(tl)+aly(tl)·cosθl(tl)
    ahr(tr)＝-arx(tr)·sinθr(tr)+ary(tr)·cosθr(tr)
    式中，ahl(tl)、ahr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时在水平前进方向上的加速度，alx(tl)、arx(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的X轴加速度，aly(tl)、ary(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时的Y轴加速度；
    (3)所述计算步距的具体方法为：
    在左小腿或右小腿的各个站立中期时刻，计算该小腿此时刻在水平前进方向上的运动速度：
    vlms＝-d·ωlzms·cosθlms
    vrms＝-d·ωrzms·cosθrms
    式中，vlms、vrms为左、右小腿在其站立中期时刻在水平前进方向的运动速度，ωlzms、ωrzms为左、右小腿在其站立中期时刻的Z轴角速度，d为该小腿上用于测量加速度、角速度的传感器的摆放位置距离脚底的距离；
    在左小腿或右小腿的各个站立中期时刻，计算该小腿的在水平前进方向上的初始位移：
    slms＝-d·sinθlms
    srms＝-d·sinθrms
    式中，slms、srms为左、右小腿在其站立中期时刻在水平前进方向上的初始位移；在到下一个站立中期时刻之前的各个时刻，该小腿的水平前进方向上的运动速度通过该方向上的加速度积分而得：
    Figure PCTCN2017071674-appb-100005
    Figure PCTCN2017071674-appb-100006
    式中，vl(tl)、vr(tr)分别为左小腿在tl时刻时、右小腿在tr时刻时水平前进方向上的运动速度，ahl(δ)、ahr(δ)分别为左小腿、右小腿瞬时的水平前进方向上的加速度；
    该小腿在相邻两个站立中期时刻之间产生的该腿的步距，通过下式计算：
    Figure PCTCN2017071674-appb-100007
    Figure PCTCN2017071674-appb-100008
    式中：LSDL、RSDL分别为左腿步距、右腿步距，slmsl、srmsl分别为左、右小腿在该腿前一个站立中期时刻的水平前进方向上的初始位移，slmsn、srmsn分别为左、右小腿在该腿后一个站立中期时刻的水平前进方向上的初始位移，Tl、Tr分别为左、右小腿的相邻两个站立中期时刻之间的时间长度，vlmsn、vrmsn分别为左、右小腿在该腿后一个站立中期时刻的水平前进方向上的运动速度，vl(Tl)、vr(Tr)分别为左小腿在Tl时刻时、右小腿在Tr时刻时通过加速度积分计算的水平前进方向上的运动速度；
    (4)拆分步距得到步长的具体方法为：
    该小腿相邻两个站立中期时刻之间产生的该腿的步长，通过下式计算：
    Figure PCTCN2017071674-appb-100009
    Figure PCTCN2017071674-appb-100010
    式中，LSL为左腿步长，RSL为右腿步长，θllc、θrlc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的左脚落地时刻的左、右小腿与铅垂线的夹角，θlrc、θrrc分别为发生在该腿前一个站立中期时刻到后一个站立中期时刻之间的右脚落地时刻的左、右小腿与铅垂线的夹角。
  4. 如权利要求2所述的用于可穿戴式设备的人体步长测量方法，其特征在于，测量步长之前，预先监测目标用户的一定距离内的直线行走过程，确定每个传感 器的X、Y轴实际方向偏离矢状面的角度：
    Figure PCTCN2017071674-appb-100011
    Figure PCTCN2017071674-appb-100012
    式中：
    Figure PCTCN2017071674-appb-100013
    为该传感器X轴实际方向偏离矢状面的角度，
    Figure PCTCN2017071674-appb-100014
    为该传感器Y轴实际方向偏离矢状面的角度，
    Figure PCTCN2017071674-appb-100015
    为用户行走过程中在站立相时该传感器X轴角速度的平均值，
    Figure PCTCN2017071674-appb-100016
    为用户行走过程中在站立相时该传感器Y轴角速度的平均值，
    Figure PCTCN2017071674-appb-100017
    为用户行走过程中在站立相时该传感器Z轴角速度的平均值；
    在测量步长过程中，每个传感器测得的三轴加速度，三轴角速度数据在使用前都预先进行第一次修正，修正公式为：
    Figure PCTCN2017071674-appb-100018
    Figure PCTCN2017071674-appb-100019
    式中：ax、ay、az为该传感器测量的X、Y、Z轴加速度，ax1、ay1、az1为传感器的测量数据经该次修正后得到的X、Y、Z轴加速度；ωx、ωy、ωz为该传感器测量的X、Y、Z轴角速度，ωx1、ωy1、ωz1为传感器的测量数据经该次修正后得到的X、Y、Z轴角速度。
  5. 如权利要求4所述的用于可穿戴式设备的人体步长测量方法，其特征在于，测量步长之前，还需预先监测目标用户的静止站立状态，确定此状态下每个传感器X轴实际方向偏离铅垂线的角度：
    Figure PCTCN2017071674-appb-100020
    式中：
    Figure PCTCN2017071674-appb-100021
    为静止站立状态下该传感器X轴实际方向偏离铅垂线的角度，ays1为用户静止站立时该传感器测得的经过第一次修正后的Y轴加速度，g为重力加速度；
    在测量步长过程中，每个传感器经过第一次修正后得到的三轴加速度、三轴角速度数据，在使用前还需进行第二次修正，修正公式如下：
    Figure PCTCN2017071674-appb-100022
    Figure PCTCN2017071674-appb-100023
    式中：axc、ayc、azc为传感器的测量数据经第二次修正后得到的X、Y、Z轴加速度；ωxc、ωyc、ωzc为传感器的数据经第二次修正后得到的X、Y、Z轴角速度；
    在测量步长过程中，axc、ayc、azc即为所述的小腿X、Y、Z轴加速度，ωxc、ωyc、ωzc即为所述的小腿X、Y、Z轴角速度。
  6. 一种实现权利要求1所述人体步长测量方法的可穿戴式设备，其特征在于，包括两个惯性传感器和上位机，每个惯性传感器包含三维加速度计以及三维角速度计，惯性传感器与上位机相连进行数据传输，上位机用于控制两个惯性传感器采集加速度、角速度数据，并将数据收集和存储。
  7. 如权利要求6所述的可穿戴式设备，其特征在于，还包括用于固定惯性传感器的固定带。
  8. 如权利要求6所述的可穿戴式设备，其特征在于，所述的惯性传感器为基于MPU6050芯片的惯性传感器。
  9. 如权利要求6所述的可穿戴式设备，其特征在于，所述的惯性传感器采样频率不低于100Hz。
  10. 如权利要求6所述的可穿戴式设备，其特征在于，所述的上位机将惯性传感器测量的数据存储入SD卡中。
PCT/CN2017/071674 2017-01-19 2017-01-19 一种用于可穿戴式设备的人体步长测量方法及其测量设备 WO2018132999A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/071674 WO2018132999A1 (zh) 2017-01-19 2017-01-19 一种用于可穿戴式设备的人体步长测量方法及其测量设备

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/071674 WO2018132999A1 (zh) 2017-01-19 2017-01-19 一种用于可穿戴式设备的人体步长测量方法及其测量设备

Publications (1)

Publication Number Publication Date
WO2018132999A1 true WO2018132999A1 (zh) 2018-07-26

Family

ID=62907619

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/071674 WO2018132999A1 (zh) 2017-01-19 2017-01-19 一种用于可穿戴式设备的人体步长测量方法及其测量设备

Country Status (1)

Country Link
WO (1) WO2018132999A1 (zh)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110176028A (zh) * 2019-06-05 2019-08-27 中国人民解放军国防科技大学 一种基于事件相机的异步角点检测方法
CN111317480A (zh) * 2020-01-20 2020-06-23 深圳市丞辉威世智能科技有限公司 步态识别方法、装置、设备及可读存储介质
CN112418110A (zh) * 2020-11-26 2021-02-26 浙江大学 一种人体步态稳定性评估方法及装置
US11185735B2 (en) 2019-03-11 2021-11-30 Rom Technologies, Inc. System, method and apparatus for adjustable pedal crank
US11325005B2 (en) 2019-10-03 2022-05-10 Rom Technologies, Inc. Systems and methods for using machine learning to control an electromechanical device used for prehabilitation, rehabilitation, and/or exercise
US11471729B2 (en) 2019-03-11 2022-10-18 Rom Technologies, Inc. System, method and apparatus for a rehabilitation machine with a simulated flywheel
US11596829B2 (en) 2019-03-11 2023-03-07 Rom Technologies, Inc. Control system for a rehabilitation and exercise electromechanical device

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011152360A (ja) * 2010-01-28 2011-08-11 Nippon Telegr & Teleph Corp <Ntt> 歩行情報抽出装置、歩行情報抽出方法、及びコンピュータ読み取り可能な記憶媒体
JP2012000343A (ja) * 2010-06-18 2012-01-05 Tohoku Univ 歩行解析システムおよび歩行解析方法
CN102824177A (zh) * 2012-07-25 2012-12-19 王哲龙 一种三维人体步态定量分析系统和方法
CN103076023A (zh) * 2013-01-09 2013-05-01 上海大唐移动通信设备有限公司 一种步长计算方法和装置
CN106333692A (zh) * 2016-11-15 2017-01-18 上海市共进通信技术有限公司 检测人体步态周期和步长的腿部或足部穿戴装置及其方法
CN106705968A (zh) * 2016-12-09 2017-05-24 北京工业大学 基于姿态识别和步长模型的室内惯性导航算法

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011152360A (ja) * 2010-01-28 2011-08-11 Nippon Telegr & Teleph Corp <Ntt> 歩行情報抽出装置、歩行情報抽出方法、及びコンピュータ読み取り可能な記憶媒体
JP2012000343A (ja) * 2010-06-18 2012-01-05 Tohoku Univ 歩行解析システムおよび歩行解析方法
CN102824177A (zh) * 2012-07-25 2012-12-19 王哲龙 一种三维人体步态定量分析系统和方法
CN103076023A (zh) * 2013-01-09 2013-05-01 上海大唐移动通信设备有限公司 一种步长计算方法和装置
CN106333692A (zh) * 2016-11-15 2017-01-18 上海市共进通信技术有限公司 检测人体步态周期和步长的腿部或足部穿戴装置及其方法
CN106705968A (zh) * 2016-12-09 2017-05-24 北京工业大学 基于姿态识别和步长模型的室内惯性导航算法

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11185735B2 (en) 2019-03-11 2021-11-30 Rom Technologies, Inc. System, method and apparatus for adjustable pedal crank
US11471729B2 (en) 2019-03-11 2022-10-18 Rom Technologies, Inc. System, method and apparatus for a rehabilitation machine with a simulated flywheel
US11541274B2 (en) 2019-03-11 2023-01-03 Rom Technologies, Inc. System, method and apparatus for electrically actuated pedal for an exercise or rehabilitation machine
US11596829B2 (en) 2019-03-11 2023-03-07 Rom Technologies, Inc. Control system for a rehabilitation and exercise electromechanical device
US11904202B2 (en) 2019-03-11 2024-02-20 Rom Technolgies, Inc. Monitoring joint extension and flexion using a sensor device securable to an upper and lower limb
CN110176028A (zh) * 2019-06-05 2019-08-27 中国人民解放军国防科技大学 一种基于事件相机的异步角点检测方法
CN110176028B (zh) * 2019-06-05 2020-12-15 中国人民解放军国防科技大学 一种基于事件相机的异步角点检测方法
US11325005B2 (en) 2019-10-03 2022-05-10 Rom Technologies, Inc. Systems and methods for using machine learning to control an electromechanical device used for prehabilitation, rehabilitation, and/or exercise
CN111317480A (zh) * 2020-01-20 2020-06-23 深圳市丞辉威世智能科技有限公司 步态识别方法、装置、设备及可读存储介质
CN112418110A (zh) * 2020-11-26 2021-02-26 浙江大学 一种人体步态稳定性评估方法及装置
CN112418110B (zh) * 2020-11-26 2022-05-17 浙江大学 一种人体步态稳定性评估方法及装置

Similar Documents

Publication Publication Date Title
WO2018132999A1 (zh) 一种用于可穿戴式设备的人体步长测量方法及其测量设备
CN106908021B (zh) 一种用于可穿戴式设备的人体步长测量方法及其测量设备
WO2018081986A1 (zh) 一种可穿戴式设备及用于该设备的实时步长测量方法
CN104757976B (zh) 一种基于多传感器融合的人体步态分析方法和系统
KR101751760B1 (ko) 하지 관절 각도를 이용한 보행 인자 추정 방법
CN106419928B (zh) 一种可穿戴式设备及用于该设备的实时步长测量方法
CN108836346A (zh) 一种基于惯性传感器的人体步态分析方法和系统
CN106153077B (zh) 一种用于m-imu人体运动捕获系统的初始化校准方法
US20150149104A1 (en) Motion Tracking Solutions Using a Self Correcting Three Sensor Architecture
Tumkur et al. Modeling human walking for step detection and stride determination by 3-axis accelerometer readings in pedometer
TWI468646B (zh) Calculation method of step distance of computing device
Laudanski et al. A concurrent comparison of inertia sensor-based walking speed estimation methods
JP2015062654A (ja) 歩容推定装置とそのプログラム、転倒危険度算出装置とそのプログラム
CN106767790B (zh) 人体下肢运动模型与卡尔曼滤波融合估计行人移动跟踪的方法
CN109646009B (zh) 基于便携式步态分析系统的步态时空参数的计算方法
CN110609621B (zh) 姿态标定方法及基于微传感器的人体运动捕获系统
CN108836344A (zh) 步长步频估算方法和装置及步态检测仪
CN111895997B (zh) 一种无需标准矫正姿势的基于惯性传感器的人体动作采集方法
JP2013190370A (ja) 状態検出装置、電子機器、測定システム及びプログラム
CN110680335A (zh) 步长测量方法及其设备、系统、非易失性计算机存储介质
CN111012358B (zh) 一种人体踝关节运动轨迹测量方法及可穿戴式设备
Hu et al. A study on estimation of planar gait kinematics using minimal inertial measurement units and inverse kinematics
CN114287890A (zh) 基于mems传感器的帕金森患者的运动功能评估方法
CN114271812A (zh) 基于惯性传感器的三维步态分析系统及方法
CN105232053B (zh) 一种人体踝关节跖屈相位检测系统及方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17893193

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 19.12.2019)

122 Ep: pct application non-entry in european phase

Ref document number: 17893193

Country of ref document: EP

Kind code of ref document: A1