CN108720841A - Wearable lower extremity movement correction system based on cloud detection - Google Patents
Wearable lower extremity movement correction system based on cloud detection Download PDFInfo
- Publication number
- CN108720841A CN108720841A CN201810492265.6A CN201810492265A CN108720841A CN 108720841 A CN108720841 A CN 108720841A CN 201810492265 A CN201810492265 A CN 201810492265A CN 108720841 A CN108720841 A CN 108720841A
- Authority
- CN
- China
- Prior art keywords
- data
- module
- movement state
- human body
- lower extremity
- 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.)
- Pending
Links
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
- A61B5/1121—Determining geometric values, e.g. centre of rotation or angular range of movement
-
- 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
- A61B5/103—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
- A61B5/1121—Determining geometric values, e.g. centre of rotation or angular range of movement
- A61B5/1122—Determining geometric values, e.g. centre of rotation or angular range of movement of movement trajectories
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/74—Details of notification to user or communication with user or patient ; user input means
- A61B5/7465—Arrangements for interactive communication between patient and care services, e.g. by using a telephone network
Abstract
A kind of wearable lower extremity movement correction system based on cloud detection, including lower extremity movement state acquisition module, data transmission module, the movement state information visualization for being set to mobile terminal and human body attitude three-dimensional reconstruction module and comprising the data processing module of cloud platform, wherein:Lower extremity movement state acquisition module is connected with data transmission module by bus and transmission sensor signal, wirelessly it is connected with human body attitude three-dimensional reconstruction module with movement state information visualization after the syndicated format that sorts after data transmission module receiving sensor signal and packing and transmits motion pose and muscular force status data, movement state information is visualized carries out big data training and storage based on database by network with human body attitude three-dimensional reconstruction module by real-time data transmission to cloud platform, data processing module will train, human body three-dimensional posture that treated, gait analysis data are back to mobile terminal to show action correction information by way of wireless network.
Description
Technical field
The present invention relates to a kind of technology in rehabilitation appliances field, specifically a kind of wearable lower limb based on cloud detection
Move correction system.
Background technology
Human recovery process is a dynamic, continually changing process.Either Physical Therapist or healing robot, control
The essence for the treatment of process is all to correct the abnormal behaviour posture of patient during the motion to get a desired effect.Therefore, rehabilitation
It is very necessary to the real-time tracking of patient physiological condition and scientific analysis in the process.Existing apparatus is mostly based on vision system,
Equipment is huge, not portable enough, and high cost.
Invention content
The present invention in view of the drawbacks of the prior art, proposes a kind of wearable lower extremity movement correction system based on cloud detection,
Lower extremity movement situation can be monitored in real time with function monitoring and according to sensor data analysis operation realization to abnormal movement shape
The detection and correction of condition have the patient of obstacle, the sportsman and soldier of needs action correction suitable for lower extremity motor function.
The present invention is achieved by the following technical solutions:
The present invention includes:Lower extremity movement state acquisition module, data transmission module, the motion state for being set to mobile terminal
Information visualization and human body attitude three-dimensional reconstruction module and the data processing module comprising cloud platform, wherein:Lower extremity movement state
Acquisition module is connected with data transmission module by bus and transmission sensor signal, data transmission module receiving sensor signal
The syndicated format that sorts afterwards simultaneously wirelessly visualizes and human body attitude three-dimensional reconstruction module with movement state information after packing
It is connected and transmits motion pose and muscular force status data, movement state information visualization and human body attitude three-dimensional reconstruction module
Real-time data transmission to cloud platform is subjected to big data training and storage based on database, data processing module by network
Training, treated human body three-dimensional posture, gait analysis data are back to mobile terminal with aobvious by way of wireless network
Show action correction information.
The sensor signal includes but not limited to:Sensor number, quaternary number information, is adopted at 3-axis acceleration information
The sample period, as shown:
1 agreement of table includes data information
Content | Number | ax | ay | az | Q0 | Q1 | Q2 | Q3 | T |
Example | 1 | 0.12 | -0.97 | -0.16 | 0.51 | -0.55 | -0.53 | -0.40 | 31 |
The data transmission module includes:Embedded system and the wireless transmission unit being attached thereto, wherein:It is embedded
System receives the sensor signal acquired from lower extremity movement state acquisition module by way of iic bus, pretreated
Motion pose and muscular force status data are exported with bluetooth approach to mobile terminal by wireless transmission unit.
The pretreatment refers to:Consider limitation of the transport protocol to size of data, to sensor signal initial data into
Row divided stator frame and reception, are 20B per bag data size, while register data is converted to standard unit's data.
The motion pose and muscular force status data include:Joint of lower extremity number measures hip joint, knee pass
Section, ankle-joint three dimensions corner, normal condition hip joint, knee joint, ankle-joint three dimensions corner, root joint space position
It sets, muscle group number and muscle group activation value.
The movement state information is visualized with human body attitude three-dimensional reconstruction module:Display interface unit, user
Administrative unit and connecting interface unit, wherein:Display interface unit by connecting interface unit respectively with wireless transmission unit and
Cloud platform, which is connected and transmits the reproduction of user movement posture, muscle activation grade and movement correction information, service management unit, to be passed through
Connecting interface unit is connected with wireless transmission unit and transmits individual subscriber contact details and corresponding device numbering information.
The data processing module includes:Cloud platform, Database Unit, neural network unit and motion rendition unit,
Wherein:Cloud platform is connected with service management unit and transmits userspersonal information and historical movement data information, Database Unit
It is connected with cloud platform and transmits individual subscriber and historical movement data information, neural network unit is connected with cloud platform and returns to biography
The defeated classification of motion and movement accuracy information, motion rendition unit are connected with cloud platform and transmit the three-dimensional people of exercise data driving
Body Model posture information.
Technique effect
Compared with prior art, the technology of the present invention effect includes:
1) perception of the nine axis movable informations such as gyroscope, accelerometer, magnetometer is integrated, different joints is can measure and is being moved through
Time And Space Parameters and the signals such as movement locus, displacement, speed, acceleration, angular displacement, angular speed, angular acceleration in journey;It can be real
Existing repeated multiple times wearable, lightweight, it is not affected by environment, daytime measurement may be implemented, while should have preferable real
When property and measurement accuracy.Sensory perceptual system and execution system combination are had higher degree of intelligence, can pass through cloud by the present invention
Platform pre-processes measurement data, compared and is evaluated, and the displaying of measurement result will visualize.
2) present invention is realized in mobile terminal and must be restored and act in the human body three-dimensional posture that computer end carries out originally
It captures, improves the portability of system, motion function evaluation and action correcting function can be carried out whenever and wherever possible;
3) present invention resolves human body attitude in three dimensions, can obtain bending and stretching at hip joint, take down the exhibits, rotates
Angle, bends and stretches angle at knee joint, and ankle bends and stretches, overturns, rotation angle, has more detailed analysis indexes;The present invention designs
Based on generalized complementary filtering human body motion capture optimization method output position repeatability error it is smaller, reappeared well by
The track of examination person's straight line moving disclosure satisfy that the demand that human body motion capture spatial location is reappeared.
Description of the drawings
Fig. 1 is present system schematic diagram;
Fig. 2 is root node position solver structure figure;
Fig. 3 is minimum measuring unit schematic diagram;
Fig. 4 is embodiment configuration diagram;
Fig. 5 is embodiment lower extremity movement state acquisition module module diagram;
Fig. 6 is embodiment iic bus connection diagram;
Fig. 7 a and b are respectively human body reference frame figure and human body lower limbs tree structure schematic diagram;
Fig. 8 is that human body pose reappears schematic diagram in embodiment Matlab;
Fig. 9 is gait processes angle schematic diagram;
In figure:Knee joint bends and stretches angle in a gait processes, hip joint angle in b gait processes, ankle-joint in c gait processes
Bend and stretch angle;
Figure 10, which is left knee joint, cannot bend and stretch schematic diagram;
In figure:A is that right knee is normal, and b is the knee joint gait after correction;
Figure 11 a and b are that embodiment moves display interface and historical record interface schematic diagram;
Figure 12 is embodiment muscle display interface schematic diagram;
In figure:1 longitudinal spacing slot, 2 positioning columns, 3 cross spacing slots, 4IMU positioning columns, the 5 U-shaped holes of fluting, 6 Elastic buckles,
7IMU chips, 8 debugging windows.
Specific implementation mode
As shown in Figure 1 and Figure 4, it is that a kind of wearable lower extremity movement correction based on cloud detection that the present embodiment is related to is
System, wherein including:Lower extremity movement state acquisition module, data transmission module, be set to mobile terminal movement state information can
Depending on changing with human body attitude three-dimensional reconstruction module and comprising the data processing module of cloud platform, wherein:Lower extremity movement state acquisition mould
Block is connected with data transmission module by bus and transmission sensor signal, sorts after data transmission module receiving sensor signal
It wirelessly visualizes with movement state information after syndicated format and packing and is connected simultaneously with human body attitude three-dimensional reconstruction module
Motion pose and muscular force status data are transmitted, movement state information visualization passes through net with human body attitude three-dimensional reconstruction module
Real-time data transmission to cloud platform is carried out big data training and storage based on database by network, and data processing module will instruct
Practice, treated human body three-dimensional posture, that gait analysis data are back to mobile terminal by way of wireless network is dynamic to show
Make correction information.
The embedded system of data transmission module uses stm32F103VET6 microcontrollers in the present embodiment, the microcontroller
The Cortex-M3 that ARM is 32, possesses the working frequency of highest 72MHz, so as to meet the operation demand of the present embodiment.
The microcontroller is communicated using DC 5V power supplies with iic bus and Inertial Measurement Unit, and it is logical to carry out bluetooth serial ports with USART
News, and with the analog signal of I/O pin acquisition surface myoelectric sensor.
In the present embodiment, the data transmission module carries out data transfer using bluetooth, under the sample frequency of 30Hz
It is about 40Kbps to need wireless bandwidth, and the actual measurement of the bandwidth of low-power consumption bluetooth only has 3~10Kbps, it is difficult to be met the requirements.Still
Finally use conventional Bluetooth, although the mode Bluetooth power consumption is larger with low-power consumption bluetooth compared with, transmission rate actual measurement energy
Reach 100Kbps, so as to meet the wireless transmission demand of the present embodiment.
Eight sensor numbers are 0x50~0x51, with embedded system by the communication mode of the iic bus
Stm32 is host, and each minimum measuring unit in lower extremity movement state acquisition module is slave.Host is in a manner of poll
Slave is called one by one, and data are obtained from each slave.Bus is as shown in fig. 6, wherein serial time clock line (SCL) and serial data line
(SDA) it needs to be pulled upward to 3.3V by 4.7K resistance, to make up bus current.
Lower extremity movement state acquisition module described in the present embodiment is made of 7 minimum measuring units, as shown in figure 5,
The foot of detection abdomen and left and right sides, the motion state of shank, thigh are corresponded to respectively, are passed through between each minimum measuring unit
Spring wire connects, and each minimum measuring unit includes:Shell fixed module and to be set to it internal comprising accelerometer, top
Spiral shell instrument, nine axle sensors of magnetometer and muscle electric transducer.
Each nine axle sensor is respectively installed to specified position in the present embodiment.By bus by each sensor
Signal transmission to embedded system.
The minimum measuring unit is bundled in the specific limbs of human body and is approximated as a rigid body, preferably by that will sense
Muscle and the place that skin is indeformable or deformation is smaller when device is bundled in each limb motion.The inertia for measuring thigh posture passes
Sensor is bound to centre position of the big leg outer side at hip joint about 4cm~5cm, and the inertial sensor for measuring shank is bound to
Centre position at one lateral extent knee joint about 4cm of shank.
By sensor attitude correct method, can eliminate sensor fixed position and corresponding skeletal joint coordinate system it
Between deviation, make sensor movement can represent lower extremity movement.
Due to the tree-shaped hierarchical structure of skeleton, the movement in upper layer joint can influence posture and the position in following level joint
It sets, is rotated based on joint length and quaternary number from root joint using the thought of direct kinematics, sub- joint can be obtained
Position and sub- joint relative to father joint rotation quaternary number, successively calculate after, by obtain the end joint i.e. position of tiptoe with
Posture.
The preferred 3D printing molding of the shell fixed module.
As shown in figure 3, the shell fixed module is fixed by bandage, specifically include:Pedestal and upper cover, wherein:Base
It being set in seat there are two the positioning column for being used for fixing IMU sensors, positioning column center is equipped with the positioning through hole for close-fitting upper cover, from
And the sliding of upper cover and pedestal horizontal direction can be limited.
It is further provided with Elastic buckle in the pedestal, the groove to match is equipped in corresponding upper cover, to limit
Upper cover is detached with pedestal vertical direction.
The both sides of the pedestal are U-shaped hole design of slotting, so as to facilitate the installation of bandage.
The upper lid is equipped with the straight slot for magnetic field calibration.
The present embodiment course of work is as follows:
Step 1, acquisition nominal data, determine the deviation between sensor coordinate system and joint coordinate system, specific steps are such as
Under:
1) it presses channel and reads IMU exercise data files, 7 joint movements data are respectively stored in 7 cell arrays, take out
Take preceding 30 groups of data for the data-optimized of initial attitude calibration;
2) experimenter's height is inputted, according to Chinese adult human size statistics table, obtains lower limb skeletons Duan Lian in proportion
Pole length is brought into human body lower limbs local coordinate system, and standard gestures when human body is stood are obtained;
3) quaternary number mean operation is done to the 30 groups of data extracted, it is inclined to obtain sensor attitude quaternary number for smoothed data
Difference;
4) joint of lower extremity tree is traversed in the circulating cycle, and the quaternary number of each articulation nodes is calculated according to positive kinematics thinking
And the coordinate position in space coordinates, left and right hip joint bends and stretches angle, left and right knee joint bends and stretches angle for calculating, left and right ankle-joint is bent
Hade is simultaneously stored in a three-dimensional array;
5) cycle prints and connects each node of lower limb, obtains motion rendition animation, exports each joint angle image.
Step 2 is divided into static parameter and dynamic parameter by obtained body gait parameter after various types of signal Processing Algorithm,
Static parameter is human body bone parameters, i.e. length etc., and dynamic parameter is the motion process parameter of human body.It is to body gait evaluation
Comprehensive analysis based on both parameters.
The analytical judgment of segmentation for each phase of gait cycle, such as table, all kinds of lower-limb ailments is substantially by dividing
The gait phase cut carries out, and completes these steps, needs very large gait analysis equipment and place.And pass through this reality
The wearable device for applying example can divide gait phase and extraction gait feature parameter extraction via recognizer, generate report automatically
Doctor can be transferred to directly to diagnose after announcement.
Step 3, motion detection analysis:For the identification of different actions, the present embodiment uses the algorithm of neural network classification
Go to realize, acquire different people walking, go upstairs, the data for the movement postures such as stride is walked, heavy burden is walked, by data prediction it
Two layers LSTM neural network is applied to classify afterwards, rate of accuracy reached can be applied to motion analysis system to 97%
Meter.The feature that the acceleration according to foot accelerometer at the landing moment is zero simultaneously, can read the information such as walking step number,
In conjunction with neural network classification as a result, providing motion detection and analysis result.
Step 4, gait correcting posture:By acquiring the sample of standard walking step state, data, will be neural after pretreatment
A certain joint angles input is adjusted to zero in network, and is changed into the mark value of network output, then neural metwork training goes out
As a result it is the desired value of the joint angles under standard gait.It, then can be at certain by the way that each joint is carried out similar process
When morbid state occurs in one joint, the rational joint angles are provided.Since the input of network is the normal joint angles of patient, because
This whole network contains patient information, has achieved the purpose that provide different correction solutions to different people.Simultaneously as network is
Precondition is good, and the correction in ill joint can reach real-time output;
Activity that the correction of lower limb posture is mainly investigated or hip knee ankle three joints and corresponding muscle, first separates here
It is corrected.By taking knee joint as an example, the present embodiment acquire about 10 people normally walk data, the stiff data of 1 group of right leg,
The data for selecting normal person to walk when training pattern.In order to achieve the purpose that knee joint is corrected, carried out in training network certain
Modification, the knee joint correlated inputs of normal person are adjusted to zero, are normal human knee joint next frame by network output modifications
Value, in brief, the model that this network training goes out infers normal person's knee joint of lower a moment with the athletic posture of current 20 frame
The case where should reaching.Due to sufferer, the joint is problematic, input such a trained model when, output will
It is the knee joint of the health for meeting patient itself.Such network amendment avoid different people cadence it is different bring ask
Topic, while also because adding the information in patient remaining joint and muscle in network inputs so that network output carries patient
Self information (as schemed).
By such method, to lower limb, same processing is done in remaining joint, can thus complete to rectify different parts
Just.After completing procedure above, classify in conjunction with lower limb movement, you can patient part is first judged, followed by the corresponding gait of calling
Program is corrected, this completes meet the different gait correction of different patients.
Step 5, human cinology's parameter identification based on acceleration signal:
During restoring human motion, in order to ensure the accuracy of human body joint angles resolving, human synovial length, half
The determination of the parameters such as diameter is most important, and the position offset that inertial sensor occurs during human motion may also influence to tie
Fruit, therefore update kinematics parameters are to ensure the basis of inertia-type motion capture device precision in real time.It is caught based on optical motion
The double bone segment models in joint and spherical model can be provided the length of gauger's four limbs by catching equipment, and inertia-type movement in the market is caught
Product is caught often to carry out by inputting human synovial length in host computer, or using the preceding straightforward procedure for executing specific action
Pose calibrating, this operating process is not only lengthy and jumbled, but also can not be carried out to the position offset of the inertial sensor in motion process
Error correction.Product establishes human body lower limbs kinematics model in conjunction with human body lower limbs joint feature first, mostly logical using human body lower limbs
Road inertial sensor information establishes acceleration, the equation between angular acceleration and angle, and carrying out parameter by least square method distinguishes
Know, and then realizes quick human body lower limbs kinematic calibration.
Step 6, lower limb movement classification:Lower limb movement classification is to carry out the basis of motion monitoring and gait analysis, needs elder generation
It identifies and is carrying out which type of action can just take different processing modes.The application of the present embodiment concentrates on motion monitoring
And rehabilitation medical, it is therefore desirable to identify some given poses, such as:For motion monitoring, action is divided into common movement side
Formula, including walk, jog, hurrying up, going upstairs, under squat up, bear a heavy burden (race);For rehabilitation medical, action is left in addition to identification
The usual positions such as road, it is also necessary to patient part is identified, such as right leg support skew is short.
Different with Conventional visual motion capture equipment, the present embodiment can not go judgement to act with the angle of image recognition, because
This needs to be used as the classification that characteristic quantity goes progress action recognition by joint angles, electromyography signal etc..Locate in advance by 5.1 data
After reason, data mark can be carried out according to action, it is contemplated that everyone posture otherness is very big, it is therefore desirable to enhance data,
For the certain processing of the carry out such as cadence and stride so that classification range expands, and the robustness of such whole network can obtain
To raising.
Step 7, gait cycle divide automatically:In a gait cycle, by common RLA partitionings in gait analysis,
One complete cycle can be divided into eight different phases, respectively:Heel contact (landing for the first time), foot are laid flat that (load-bearing is anti-
Answer), midstance, terminal stance (heel is liftoff), toes are liftoff, swing initial stage, swing mid-term, swing latter stage;Wherein first five
A is support phase, and latter three are swing phase.
Above divide gives a unified quantitative criteria, in each phase, hip, SCID Mice and corresponding flesh
The activity of meat is more single, thus for normal gait, can directly divide completion according to threshold value substantially;And for disease
Shape gait, illness can be refine to some by the overwhelming majority or certain phases go wrong, and such as be subtracted in pain gait, the branch of Ipsilateral
The time of support phase shortens, and to reduce suffering limb heavy burden to the greatest extent, monitoring that some phase goes wrong in this way can correspond to
Corresponding illness.Therefore it is to carry out the necessary links of intelligent auxiliary diagnosis to carry out gait phase segmentation, and accurate segmentation can assist curing
Most rational diagnosis is made in life, while can very easily store data, and basis is carried out for big data diagnosis in future, actually this
The segmentation of sample plays the role of extracting feature.
For medical diagnosis, the segmentation of previous phase is all to be shot by video camera, then carried out to video by doctor
Analysis.The present embodiment does not have camera to carry out lower extremity movement state acquisition module, it is therefore desirable to by the analysis for data
Reach the automatic target for dividing gait phase.But for part symptom gait, due in some phases gait be it is wrong, because
This directly can not divide gait phase according to threshold value, it is therefore desirable to design the sorting algorithm that a threshold decision is aided with neural network
To reach division target.
For threshold decision, can be calculated according to numerical value in table and gathered data according to 1 table of table, 1 joint threshold decision table
Joint angles be compared, can be completed using a decision tree structure.
1 joint threshold decision table of table
Mistake will produce according to threshold value since certain deviation occur in the joint angles of certain phases for symptom gait
Misclassification, thus need to go the feature for excavating some deep layers come subsidiary classification by neural network.Eight can be obtained according to table 1
First and third, five, seven phases are actually a static action in phase, remaining four are process phase, are asked so dividing phase
Topic can be converted into:To in a cycle, the identification problem of four key frames.Since the data of various symptom gaits are difficult to obtain
, it is currently the division carried out according to the collected data of energy, therefore the neural network is to need to constantly update to keep accurate
True property.
Step 7, the human motion spatial position based on generalized complementary filtering reappear:Due to human body pose reproduction in abdomen
Portion's node is root node, with root node spatial pose and clear out the space bit of human body in conjunction with the spatial attitude of each IMU
Appearance, so being determined that the spatial position of root node has determined that the spatial position of human body.However according to Human Body Gait Analysis, it is expert at
The time that the left side (right side) of people has 60 percent in the entire period of motion enough during walking is static, however abdomen root section
Point, in persistent movement, prodigious site error is easily caused to directly be integrated to its acceleration in whole cycle.
The characteristics of being moved for human body lower limbs and the characteristic for combining complementary filter, are filtered by generalized complementary to foot
Acceleration signal and abdomen acceleration signal extract analysis, however due to being all acceleration signal so sample frequency and dynamic
Step response is identical, is unsatisfactory for the condition of traditional complementary filter.
The generalized complementary filtering, letter is inputted using the acceleration signal of abdomen consecutive variations as the high frequency of filter
Number, to there is the foot signal of quiescent phase to pre-process, obtained foot movement displacement will be pre-processed and inputted as low frequency
Signal.It is filtered respectively with significant difference of two kinds of signals on frequency domain, the result in conjunction with filtering calibrates and inhibits abdomen
The drift that the integrated acceleration of root node is brought.
In human body walking, double-legged movement locus has significant periodicity, and when foot lands, has longer
A period of time it is static, by the way that with foot, to carve speed when being contacted with ground be zero this characteristic, foot acceleration can be believed
Number integrator drift effectively inhibited, specific steps include:Enable subject quiet three seconds vertical after system boot first, by this section
Null offset of the acceleration mean value of IMU three axis in reference frame as each axis in time, after subtracting null offset
Carry out next step operation.Resultant acceleration threshold value is set, is then judged to contacting to earth when detecting that foot IMU resultant accelerations are less than threshold value
State, by the speed zero setting of IMU.Since sampled data is discrete data, integrator drift error will necessarily be remained.At i-th week
It is t in phase, at the time of enabling foot start statici, foot duration of oscillation is Δ Ti, sampling period dt.It may be considered that
, due to the velocity error that integrator drift is brought, to subtract mean value of the velocity error to duration of oscillation in a foot movement period
It is integrated again afterwards, the error that integrator drift brings foot displacement can be significantly inhibited by being experimentally confirmed this method.
In order to obtain abdomen root node space displacement track, it is necessary to which drift of the rejection of acceleration in integral misses
Difference.However in the process of walking, web joint persistently keeps moving, and integral drift is easily generated without extraneous vision assisted calibration
It moves, thus can not the displacement of accurate recreation subject in space.Foot displacement is being obtained by above-mentioned foot signal processing
Afterwards, carry out assisted calibration abdomen signal with calibrated foot signal.
Above-mentioned specific implementation can by those skilled in the art under the premise of without departing substantially from the principle of the invention and objective with difference
Mode carry out local directed complete set to it, protection scope of the present invention is subject to claims and not by above-mentioned specific implementation institute
Limit, each implementation within its scope is by the constraint of the present invention.
Claims (9)
1. a kind of wearable lower extremity movement correction system based on cloud detection, which is characterized in that including:Lower extremity movement state acquisition
Module, data transmission module, be set to mobile terminal movement state information visualization with human body attitude three-dimensional reconstruction module and
Include the data processing module of cloud platform, wherein:Lower extremity movement state acquisition module is connected by bus with data transmission module
And transmission sensor signal, after data transmission module receiving sensor signal sequence syndicated format and after being packaged wirelessly
It is connected with human body attitude three-dimensional reconstruction module with movement state information visualization and transmits motion pose and muscular force status number
According to movement state information visualization is carried out real-time data transmission to cloud platform by network with human body attitude three-dimensional reconstruction module
Big data training and the storage based on database, data processing module will train, treated human body three-dimensional posture, gait point
Analysis data are back to mobile terminal to show action correction information by way of wireless network;
The sensor signal includes:Sensor number, 3-axis acceleration information, quaternary number information, sampling period;
The motion pose and muscular force status data include:Joint of lower extremity number measures hip joint, knee joint, ankle
Joint three dimensions corner, normal condition hip joint, knee joint, ankle-joint three dimensions corner, root joint space position, muscle
Group number and muscle group activation value.
2. system according to claim 1, characterized in that the data transmission module includes:Embedded system and with
Connected wireless transmission unit, wherein:Embedded system is received by way of iic bus from lower extremity movement state acquisition
The sensor signal of module acquisition, pretreated motion pose and muscular force status data are by wireless transmission unit with indigo plant
Tooth mode is exported to mobile terminal.
3. system according to claim 2, characterized in that the communication mode of the iic bus compiles eight sensors
Number it is 0x50~0x51, using embedded system as host, each minimum measuring unit in lower extremity movement state acquisition module is
Slave, host call slave one by one in a manner of poll, and data, wherein serial time clock line and serial data are obtained from each slave
Line is pulled upward to 3.3V to make up bus current by 4.7K resistance.
4. system according to claim 1, characterized in that the movement state information visualization and human body attitude is three-dimensional
Rendering module includes:Display interface unit, service management unit and connecting interface unit, wherein:Display interface unit passes through company
Connection interface unit be connected respectively with wireless transmission unit and cloud platform and transmit user movement posture reappear, muscle activation grade and
Movement correction information, service management unit, which is connected with wireless transmission unit by connecting interface unit and transmits individual subscriber, to be contacted
Information and corresponding device numbering information.
5. system according to claim 1, characterized in that the data processing module includes:Cloud platform, database list
Member, neural network unit and motion rendition unit, wherein:Cloud platform is connected with service management unit and transmits userspersonal information
With historical movement data information, Database Unit is connected with cloud platform and transmits individual subscriber and historical movement data information, god
It is connected with cloud platform through network element and returns to the transmission classification of motion and movement accuracy information, motion rendition unit and cloud platform
It is connected and transmits the three-dimensional (3 D) manikin posture information of exercise data driving.
6. system according to claim 1 or 3, characterized in that the lower extremity movement state acquisition module is described in 7
Minimum measuring unit composition, the motion state of the foot of corresponding detection abdomen and left and right sides, shank, thigh, each respectively
It is connected by spring wire between minimum measuring unit, each minimum measuring unit includes:Shell fixed module and it is set to it
Internal includes accelerometer, gyroscope, nine axle sensors of magnetometer and muscle electric transducer.
7. system according to claim 6, characterized in that when the minimum measuring unit is bundled in each limb motion
Muscle and the place that skin is indeformable or deformation is smaller, including:The inertial sensor for measuring thigh posture is bound to outside thigh
Centre position at lateral extent hip joint about 4cm~5cm, the inertial sensor for measuring shank are bound to one lateral extent knee of shank pass
Save the centre position at 4cm.
8. system according to claim 6, characterized in that the shell fixed module is fixed by bandage, specific to wrap
It includes:Pedestal and upper cover, wherein:It is set in pedestal there are two the positioning column for being used for fixing IMU sensors, positioning column center is equipped with and is used for
The positioning through hole of close-fitting upper cover, so as to limit the sliding of upper cover and pedestal horizontal direction.
9. system according to claim 8, characterized in that Elastic buckle is further provided in the pedestal, in correspondence
It is equipped with the groove to match in lid, is detached with pedestal vertical direction to limit upper cover;The both sides of the pedestal are fluting U
Type hole designs, so as to facilitate the installation of bandage;The upper lid is equipped with the straight slot for magnetic field calibration.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810492265.6A CN108720841A (en) | 2018-05-22 | 2018-05-22 | Wearable lower extremity movement correction system based on cloud detection |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810492265.6A CN108720841A (en) | 2018-05-22 | 2018-05-22 | Wearable lower extremity movement correction system based on cloud detection |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108720841A true CN108720841A (en) | 2018-11-02 |
Family
ID=63938815
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810492265.6A Pending CN108720841A (en) | 2018-05-22 | 2018-05-22 | Wearable lower extremity movement correction system based on cloud detection |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108720841A (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109737941A (en) * | 2019-01-29 | 2019-05-10 | 桂林电子科技大学 | A kind of human action method for catching |
CN109781094A (en) * | 2018-12-24 | 2019-05-21 | 上海交通大学 | Earth magnetism positioning system based on Recognition with Recurrent Neural Network |
CN110276380A (en) * | 2019-05-22 | 2019-09-24 | 杭州电子科技大学 | A kind of real time kinematics online direction system based on depth model frame |
CN110338952A (en) * | 2019-06-27 | 2019-10-18 | 浙江大学 | A kind of device of trained normal gait and its application |
CN110693501A (en) * | 2019-10-12 | 2020-01-17 | 上海应用技术大学 | Wireless walking gait detection system based on multi-sensor fusion |
CN111544006A (en) * | 2020-05-15 | 2020-08-18 | 中国科学院自动化研究所 | Wearable equipment for quantifying and identifying dyskinesia of Parkinson's disease people |
CN111588597A (en) * | 2020-04-22 | 2020-08-28 | 百年旭康医疗器械有限公司 | Intelligent interactive walking training system and implementation method thereof |
WO2020223944A1 (en) * | 2019-05-09 | 2020-11-12 | 深圳大学 | System and method for physiological function assessment |
CN112258555A (en) * | 2020-10-15 | 2021-01-22 | 佛山科学技术学院 | Real-time attitude estimation motion analysis method, system, computer equipment and storage medium |
CN112370048A (en) * | 2020-11-10 | 2021-02-19 | 南京紫金体育产业股份有限公司 | Movement posture injury prevention method and system based on joint key points and storage medium |
CN112842824A (en) * | 2021-02-24 | 2021-05-28 | 郑州铁路职业技术学院 | Training method for lower limb rehabilitation recovery |
CN113229806A (en) * | 2021-05-14 | 2021-08-10 | 哈尔滨工程大学 | Wearable human body gait detection and navigation system and operation method thereof |
CN113876316A (en) * | 2021-09-16 | 2022-01-04 | 河南翔宇医疗设备股份有限公司 | System, method, device, equipment and medium for detecting abnormal flexion and extension activities of lower limbs |
CN114271812A (en) * | 2021-12-06 | 2022-04-05 | 南京大学 | Three-dimensional gait analysis system and method based on inertial sensor |
CN114469079A (en) * | 2022-01-29 | 2022-05-13 | 北京中科深智科技有限公司 | Body joint measuring method using LightHouse |
CN114533039A (en) * | 2021-12-27 | 2022-05-27 | 重庆邮电大学 | Human body joint position and angle calculating method based on redundant sensors |
US11426098B2 (en) | 2020-03-02 | 2022-08-30 | PROVA Innovations Ltd. | System and method for gait monitoring and improvement |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011013099A1 (en) * | 2009-07-31 | 2011-02-03 | Koninklijke Philips Electronics N.V. | Method and system for providing a training program to a subject |
CN102670217A (en) * | 2012-05-04 | 2012-09-19 | 嘉兴市制衡精仪有限公司 | Wearable sensor measuring device and method for lower limb joint acting force and moment |
CN105286875A (en) * | 2015-11-24 | 2016-02-03 | 龙岩学院 | Gait analysis method for auxiliary screening of injury of anterior cruciate ligament |
CN105615890A (en) * | 2015-12-24 | 2016-06-01 | 西安交通大学 | Angle and myoelectricity continuous decoding method for human body lower limb walking joint |
CN106650687A (en) * | 2016-12-30 | 2017-05-10 | 山东大学 | Posture correction method based on depth information and skeleton information |
CN107243141A (en) * | 2017-05-05 | 2017-10-13 | 北京工业大学 | A kind of action auxiliary training system based on motion identification |
-
2018
- 2018-05-22 CN CN201810492265.6A patent/CN108720841A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2011013099A1 (en) * | 2009-07-31 | 2011-02-03 | Koninklijke Philips Electronics N.V. | Method and system for providing a training program to a subject |
CN102670217A (en) * | 2012-05-04 | 2012-09-19 | 嘉兴市制衡精仪有限公司 | Wearable sensor measuring device and method for lower limb joint acting force and moment |
CN105286875A (en) * | 2015-11-24 | 2016-02-03 | 龙岩学院 | Gait analysis method for auxiliary screening of injury of anterior cruciate ligament |
CN105615890A (en) * | 2015-12-24 | 2016-06-01 | 西安交通大学 | Angle and myoelectricity continuous decoding method for human body lower limb walking joint |
CN106650687A (en) * | 2016-12-30 | 2017-05-10 | 山东大学 | Posture correction method based on depth information and skeleton information |
CN107243141A (en) * | 2017-05-05 | 2017-10-13 | 北京工业大学 | A kind of action auxiliary training system based on motion identification |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109781094A (en) * | 2018-12-24 | 2019-05-21 | 上海交通大学 | Earth magnetism positioning system based on Recognition with Recurrent Neural Network |
CN109737941A (en) * | 2019-01-29 | 2019-05-10 | 桂林电子科技大学 | A kind of human action method for catching |
WO2020223944A1 (en) * | 2019-05-09 | 2020-11-12 | 深圳大学 | System and method for physiological function assessment |
CN110276380B (en) * | 2019-05-22 | 2021-08-17 | 杭州电子科技大学 | Real-time motion on-line guidance system based on depth model framework |
CN110276380A (en) * | 2019-05-22 | 2019-09-24 | 杭州电子科技大学 | A kind of real time kinematics online direction system based on depth model frame |
CN110338952A (en) * | 2019-06-27 | 2019-10-18 | 浙江大学 | A kind of device of trained normal gait and its application |
CN110338952B (en) * | 2019-06-27 | 2021-01-05 | 浙江大学 | Device for training normal gait and application thereof |
CN110693501A (en) * | 2019-10-12 | 2020-01-17 | 上海应用技术大学 | Wireless walking gait detection system based on multi-sensor fusion |
US11426098B2 (en) | 2020-03-02 | 2022-08-30 | PROVA Innovations Ltd. | System and method for gait monitoring and improvement |
CN111588597A (en) * | 2020-04-22 | 2020-08-28 | 百年旭康医疗器械有限公司 | Intelligent interactive walking training system and implementation method thereof |
CN111544006A (en) * | 2020-05-15 | 2020-08-18 | 中国科学院自动化研究所 | Wearable equipment for quantifying and identifying dyskinesia of Parkinson's disease people |
CN112258555A (en) * | 2020-10-15 | 2021-01-22 | 佛山科学技术学院 | Real-time attitude estimation motion analysis method, system, computer equipment and storage medium |
CN112370048A (en) * | 2020-11-10 | 2021-02-19 | 南京紫金体育产业股份有限公司 | Movement posture injury prevention method and system based on joint key points and storage medium |
CN112842824A (en) * | 2021-02-24 | 2021-05-28 | 郑州铁路职业技术学院 | Training method for lower limb rehabilitation recovery |
CN112842824B (en) * | 2021-02-24 | 2023-05-16 | 郑州铁路职业技术学院 | Training method for rehabilitation and recovery of lower limbs |
CN113229806A (en) * | 2021-05-14 | 2021-08-10 | 哈尔滨工程大学 | Wearable human body gait detection and navigation system and operation method thereof |
CN113876316A (en) * | 2021-09-16 | 2022-01-04 | 河南翔宇医疗设备股份有限公司 | System, method, device, equipment and medium for detecting abnormal flexion and extension activities of lower limbs |
CN113876316B (en) * | 2021-09-16 | 2023-10-10 | 河南翔宇医疗设备股份有限公司 | System, method, device, equipment and medium for detecting abnormal lower limb flexion and extension activities |
CN114271812A (en) * | 2021-12-06 | 2022-04-05 | 南京大学 | Three-dimensional gait analysis system and method based on inertial sensor |
CN114533039A (en) * | 2021-12-27 | 2022-05-27 | 重庆邮电大学 | Human body joint position and angle calculating method based on redundant sensors |
CN114533039B (en) * | 2021-12-27 | 2023-07-25 | 重庆邮电大学 | Human joint position and angle resolving method based on redundant sensor |
CN114469079A (en) * | 2022-01-29 | 2022-05-13 | 北京中科深智科技有限公司 | Body joint measuring method using LightHouse |
CN114469079B (en) * | 2022-01-29 | 2022-09-06 | 北京中科深智科技有限公司 | Body joint measurement method using LightHouse |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108720841A (en) | Wearable lower extremity movement correction system based on cloud detection | |
CN110916679B (en) | Human body lower limb pose gait detection device and method | |
Rapp et al. | Estimation of kinematics from inertial measurement units using a combined deep learning and optimization framework | |
EP3986266A1 (en) | Wearable joint tracking device with muscle activity and methods thereof | |
Wang et al. | Using wearable sensors to capture posture of the human lumbar spine in competitive swimming | |
Chen | Human motion analysis with wearable inertial sensors | |
Lu et al. | Evaluation of classification performance in human lower limb jump phases of signal correlation information and LSTM models | |
Lu et al. | Effective recognition of human lower limb jump locomotion phases based on multi-sensor information fusion and machine learning | |
US20220409098A1 (en) | A wearable device for determining motion and/or a physiological state of a wearer | |
Hossain et al. | Deepbbwae-net: A cnn-rnn based deep superlearner for estimating lower extremity sagittal plane joint kinematics using shoe-mounted imu sensors in daily living | |
Yi et al. | Continuous prediction of lower-limb kinematics from multi-modal biomedical signals | |
Yi et al. | Smart healthcare-oriented online prediction of lower-limb kinematics and kinetics based on data-driven neural signal decoding | |
Liang et al. | Synergy-based knee angle estimation using kinematics of thigh | |
CN114366559A (en) | Multi-mode sensing system for lower limb rehabilitation robot | |
Chen et al. | Gait recognition for lower limb exoskeletons based on interactive information fusion | |
Barua et al. | Deep learning techniques in estimating ankle joint power using wearable IMUs | |
CN111657951B (en) | Respiration monitoring device based on sensor array type arrangement | |
CN115644858B (en) | Wearable intelligent gait analyzer based on multi-sensor data fusion technology | |
Alemayoh et al. | LocoESIS: Deep-Learning-Based Leg-Joint Angle Estimation from a Single Pelvis Inertial Sensor | |
CN114831784A (en) | Lower limb prosthesis terrain recognition system and method based on multi-source signals | |
Huang et al. | Wearable sensors for Detecting and Measuring kinetic characteristics | |
Asif et al. | Analysis of Human Gait Cycle With Body Equilibrium Based on Leg Orientation | |
Wang et al. | A New Hidden Markov Model Algorithm to Detect Human Gait Phase Based on Information Fusion Combining Inertial with Plantar Pressure. | |
Tasca | A machine learning approach for spatio-temporal gait analysis based on a head-mounted inertial sensor | |
CN112107290A (en) | Systems, methods, and software applications for KAM predicting multiple gait cycles of a subject |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181102 |
|
RJ01 | Rejection of invention patent application after publication |