CN111539507B - Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm - Google Patents

Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm Download PDF

Info

Publication number
CN111539507B
CN111539507B CN202010201690.2A CN202010201690A CN111539507B CN 111539507 B CN111539507 B CN 111539507B CN 202010201690 A CN202010201690 A CN 202010201690A CN 111539507 B CN111539507 B CN 111539507B
Authority
CN
China
Prior art keywords
particle
optimal
optimization algorithm
objective function
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010201690.2A
Other languages
Chinese (zh)
Other versions
CN111539507A (en
Inventor
张辉
石谦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
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 Beihang University filed Critical Beihang University
Priority to CN202010201690.2A priority Critical patent/CN111539507B/en
Publication of CN111539507A publication Critical patent/CN111539507A/en
Application granted granted Critical
Publication of CN111539507B publication Critical patent/CN111539507B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/004—Artificial life, i.e. computing arrangements simulating life
    • G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
    • 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
    • A61B5/1118—Determining activity level
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6802—Sensor mounted on worn items
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
    • A61B5/6801—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient specially adapted to be attached to or worn on the body surface
    • A61B5/6813—Specially adapted to be attached to a specific body part
    • A61B5/6824—Arm or wrist

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Molecular Biology (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • General Health & Medical Sciences (AREA)
  • Veterinary Medicine (AREA)
  • Medical Informatics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Public Health (AREA)
  • Pathology (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Physiology (AREA)
  • Dentistry (AREA)
  • Oral & Maxillofacial Surgery (AREA)
  • Rehabilitation Tools (AREA)

Abstract

The invention discloses a recovery movement speed calculation model parameter identification method based on a particle swarm optimization algorithm, and relates to a model parameter identification strategy for detecting recovery movement speed by utilizing portable movement parameter measurement equipment. Firstly, establishing a walking speed model containing unknown parameters; secondly, establishing an optimization algorithm target function; and finally, optimizing the objective function by utilizing a particle swarm optimization algorithm to obtain the optimal model parameters. The wrist-wearing type exercise measuring equipment is introduced to measure the rehabilitation exercise parameters, and compared with the foot-wearing type exercise measuring equipment, the wrist-wearing type exercise measuring equipment brings convenience to training personnel. The rehabilitation exercise speed calculation model parameter identification method based on the particle swarm optimization algorithm establishes a wrist part exercise and walking speed relation model, and provides a training speed calculation method for wrist wearable exercise measurement equipment. Meanwhile, model parameter identification is carried out by utilizing a particle swarm optimization algorithm, and an accurate and rapid model parameter solving method is provided for a relation model of wrist movement and walking speed.

Description

Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm
Technical Field
The invention relates to a motion model parameter method, in particular to a model parameter identification strategy for detecting rehabilitation motion speed by utilizing portable motion parameter measuring equipment, and belongs to the field of sports medicine.
Background
Motion tracking may be used in the rehabilitation field. For the disabled with leg or the old with inconvenient walking, the walking assisting device is a necessity for rehabilitation training. Usually rehabilitation training needs to be done in a specialized rehabilitation center under the supervision of a physician. However, due to cost and manpower limitations, the number and availability of rehabilitation centers often cannot meet the needs of all people requiring rehabilitation training. And the portable motion acquisition equipment can acquire the motion data of the rehabilitation people at home in real time and transmit the data to a doctor.
The motion acquisition equipment internally comprises an acceleration sensor, an angular velocity sensor and a magnetometer. As the foot movement can reflect various states of the human body during walking, most of the existing movement acquisition equipment needs to be worn on the feet or shoes by a person to be tested, and the wearing part is inconvenient for the old or people with walking inconvenience. While wearing the motion capture device on the wrist may avoid this inconvenience. However, the movement of the wrist part is less related to the state of the human body when walking, and particularly, the movement of the hand is less noticeable when the walking aid is used than when walking normally, and therefore the present patent is directed to establishing the relationship between the wrist part and the walking state.
Disclosure of Invention
The invention provides a calculation model and a parameter identification method, aiming at the problem that the walking speed of the portable motion measurement equipment worn on the wrist part is difficult to calculate when a walking auxiliary device is arranged.
The invention provides a rehabilitation exercise speed model parameter identification method, which comprises the following steps:
step 1, establishing a walking speed model containing unknown parameters;
step 2, establishing an optimization algorithm objective function, and solving unknown parameters in the step 1;
and 3, optimizing the objective function by utilizing a particle swarm optimization algorithm to obtain the optimal model parameters.
The invention has the advantages that:
(1) according to the rehabilitation exercise speed calculation model parameter identification method based on the particle swarm optimization algorithm, the wrist-worn exercise measurement equipment is introduced to measure rehabilitation exercise parameters, and compared with foot wearing, convenience is brought to training personnel.
(2) The invention relates to a recovery movement velocity calculation model parameter identification method based on a particle swarm optimization algorithm, which establishes a relation model of wrist position movement and walking velocity and provides a training velocity calculation method for wrist wearing type movement measurement equipment.
(3) The rehabilitation exercise speed calculation model parameter identification method based on the particle swarm optimization algorithm utilizes the particle swarm optimization algorithm to identify the model parameters, and provides an accurate and rapid model parameter solving method for a relation model of wrist position exercise and walking speed.
Drawings
FIG. 1 is a general flow chart of the rehabilitation exercise velocity calculation model parameter identification method based on the particle swarm optimization algorithm.
FIG. 2 is a flow chart of parameter identification objective function calculation in the rehabilitation exercise velocity calculation model parameter identification method based on particle swarm optimization.
FIG. 3 is a calculation flow chart of a particle swarm optimization algorithm in the rehabilitation exercise velocity calculation model parameter identification method based on the particle swarm optimization algorithm.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings.
The rehabilitation exercise speed calculation model parameter identification method based on the particle swarm optimization algorithm is realized by the following steps as shown in figure 1:
the method comprises the following steps: and establishing a motion relation model containing unknown parameters.
As shown in FIG. 2, a rehabilitation person using the walking assistance device wears a data acquisition device (e.g., Apple Watch) on the wrist, and the data acquisition device acquires acceleration signals (a) in three directions of x, y and z at a certain sampling periodx，ay，az) And an angular velocity signal (g)x，gy，gz). The motion relation model is the mathematical relation between the motion data of the wrist part and the walking speed. The walking speed v, the frequency f of the periodic motion of the wrist part and the angular velocity g in the z direction can be obtained from the correlation analysis when the walking assistance device is usedzAbout, therefore, the kinematic relationship model is built as follows:
v＝af+bgz+c (1)
wherein, a, b and c are unknown parameters, which are parameters to be identified in the model.
Step two: and establishing an objective function for parameter solution.
In order to solve unknown parameters a, b and c, the method converts the parameter solving problem into an optimization problem, and firstly, an objective function for parameter solving needs to be established, which is specifically as follows:
if the time required by the walking distance L is T when the rehabilitation training personnel wear the portable movement measuring equipment, the walking distance L is carried out, and when the same movement of the duration T is carried out, the frequency measured by the movement measuring equipment is f, and the angular speed in the z direction is gz. The method establishes a target letterThe numbers are as follows:
Figure GDA0003142093470000021
as shown in FIG. 3, the optimal values of the unknown parameters a, b, and c are determined by the walking speed v estimated by the relational model and the speed calculated by the speed definitional formula when the trainees perform the same exercise
Figure GDA0003142093470000031
The sum of the squares of the differences between is minimal.
Step three: the parameter value that minimizes the objective function is solved.
As shown in fig. 2, the variables to be optimized include three variables a, b, and c, and the optimization objective function is cost (a, b, c). Parameter identification is realized by utilizing a particle swarm optimization algorithm, so that an objective function is optimal, namely the sum of squares of the difference between v and L/T is minimized, namely a, b and c which enable the objective function to be minimum are solved, and the method specifically comprises the following steps:
(a) setting the number of population particles as m and the particle dimension 3;
(b) randomly initializing the speed and position of each particle in the population to obtain the initial population position
Figure GDA0003142093470000032
The lower corner marks indicate particle labels; u. ofmThe method is a three-dimensional vector which represents the position of the mth particle and is a possible solution in the optimization process; the superscript "1" indicates that the 1 st iteration is currently performed; setting initial population velocity
Figure GDA0003142093470000033
(c) In an iterative process, the objective function values for each possible solution are compared. Setting the optimal position (the minimum value of the objective function) of each particle as the optimal position p of the particle in the iterative processbest,i(ii) a The optimal positions in all the particles are obtained through comparison and are set as the optimal positions g of the particle swarmbest；
(d) And updating the speed and the position of the particle, wherein the speed and the position of the ith particle are respectively as follows:
Figure GDA0003142093470000034
wherein the content of the first and second substances,
Figure GDA0003142093470000035
for the velocity of the kth iteration of the ith particle,
Figure GDA0003142093470000036
the velocity of the (k + 1) th iteration of the ith particle,
Figure GDA0003142093470000037
for the position of the ith particle before the kth iteration,
Figure GDA0003142093470000038
is the position of the ith particle before the (k + 1) th iteration; w is the inertial weight, r1And r2Is distributed in the interval [0,1]The random number in (c), k is the current iteration number, the initial value is 1,
Figure GDA0003142093470000039
for the individual optimal particle position of the ith particle at the kth iteration,
Figure GDA00031420934700000310
is the global optimal particle position at the k generation, c1And c2Is a constant.
And further obtaining the position of the k +1 iteration population:
Figure GDA00031420934700000311
(e) calculating the objective function value of each particle in the step k +1 and comparing the objective function value with the previous optimal position
Figure GDA00031420934700000312
Comparing the corresponding objective function values, if the current position isPreferably, the current position is used as the optimal position of the particle
Figure GDA00031420934700000313
The objective function of each particle and the optimal particle position of the particle swarm are compared
Figure GDA00031420934700000314
Comparing, if the current position is better, updating the optimal particle position
Figure GDA00031420934700000315
(f) Checking a final value condition, stopping iteration if the precision meets a preset condition or the iteration frequency exceeds a limit, and otherwise, repeating the steps (c) - (f);
(g) population optimization with optimal output
Figure GDA0003142093470000041
The optimal a, b and c are obtained.
Finally, after a, b and c are identified, the relationship between the movement speed, the frequency and the angular speed of the rehabilitation personnel is established. During actual measurement, the frequency and the angular speed are brought into the relational expression, so that the movement speed of the rehabilitation personnel can be measured, and the complicated steps of calculating the movement speed and establishing a movement model by using the traditional method are avoided.

Claims (2)

1. A recovery movement velocity calculation model parameter identification method based on particle swarm optimization algorithm is characterized in that: the method is realized by the following steps:
step 1, establishing a walking speed model containing unknown parameters;
step 2, establishing an optimization algorithm objective function, and solving unknown parameters in the step 1; the specific solving method of the unknown parameters comprises the following steps:
if the time required by the walking distance L is T when the rehabilitation training personnel wear the portable movement measuring equipment, the walking distance L is carried out, and when the same movement of the duration T is carried out, the frequency measured by the movement measuring equipment is f, and the angular speed in the z direction is gz(ii) a The objective function is established as follows:
Figure FDA0003165055150000011
wherein a, b and c are parameters to be identified in the model;
step 3, optimizing a target function by utilizing a particle swarm optimization algorithm to obtain optimal model parameters; the method for obtaining the optimal model parameters comprises the following steps:
the variables to be optimized comprise three variables a, b and c, and the optimization objective function is cost (a, b and c); parameter identification is realized by utilizing a particle swarm optimization algorithm, so that an objective function is optimal, namely the sum of squares of the difference between v and L/T is minimized, namely a, b and c which enable the objective function to be minimum are solved, and the method specifically comprises the following steps:
(a) setting the number of population particles as m and the particle dimension 3;
(b) randomly initializing the speed and position of each particle in the population to obtain the initial population position
Figure FDA0003165055150000012
The lower corner marks indicate particle labels; u. ofmThe method is a three-dimensional vector which represents the position of the mth particle and is a possible solution in the optimization process; the superscript "1" indicates that the 1 st iteration is currently performed; setting initial population velocity
Figure FDA0003165055150000013
(c) In the iterative process, comparing objective function values of all possible solutions; setting each particle optimal position as the particle optimal position p in the iterative processbest,i(ii) a The optimal positions in all the particles are obtained through comparison and are set as the optimal positions g of the particle swarmbest；
(d) And updating the speed and the position of the particle, wherein the speed and the position of the ith particle are respectively as follows:
Figure FDA0003165055150000014
wherein the content of the first and second substances,
Figure FDA0003165055150000015
for the velocity of the kth iteration of the ith particle,
Figure FDA0003165055150000016
the velocity of the (k + 1) th iteration of the ith particle,
Figure FDA0003165055150000017
for the position of the ith particle before the kth iteration,
Figure FDA0003165055150000018
is the position of the ith particle before the (k + 1) th iteration; w is the inertial weight, r1And r2Is distributed in the interval [0,1]The random number in (c), k is the current iteration number, the initial value is 1,
Figure FDA0003165055150000021
for the individual optimal particle position of the ith particle at the kth iteration,
Figure FDA0003165055150000022
is the global optimal particle position at the k generation, c1And c2Is a constant;
and further obtaining the position of the k +1 iteration population:
Figure FDA0003165055150000023
(e) calculating the objective function value of each particle in the step k +1 and comparing the objective function value with the previous optimal position
Figure FDA0003165055150000024
Comparing the obtained objective function values, and if the current position is better, taking the current position as the optimal position of the particle
Figure FDA0003165055150000025
The objective function of each particle and the optimal particle position of the particle swarm are compared
Figure FDA0003165055150000026
Comparing, if the current position is better, updating the optimal particle position
Figure FDA0003165055150000027
(f) Checking a final value condition, stopping iteration if the precision meets a preset condition or the iteration frequency exceeds a limit, and otherwise, repeating the steps (c) - (f);
(g) population optimization with optimal output
Figure FDA0003165055150000028
The optimal a, b and c are obtained.
2. The particle swarm optimization algorithm-based rehabilitation exercise velocity calculation model parameter identification method as claimed in claim 1, wherein: the method for establishing the walking speed model in the step 1 comprises the following steps:
the rehabilitation personnel using the walking assisting device wears the data acquisition equipment on the wrist part, and the data acquisition equipment acquires acceleration signals (a) in the x, y and z directions in a certain sampling periodx，ay，az) And an angular velocity signal (g)x，gy，gz) (ii) a The motion relation model is a mathematical relation between the motion data of the wrist part and the walking speed, and comprises the following steps:
v＝af+bgz+c (3)
wherein, a, b and c are parameters to be identified in the model.
CN202010201690.2A 2020-03-20 2020-03-20 Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm Active CN111539507B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010201690.2A CN111539507B (en) 2020-03-20 2020-03-20 Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010201690.2A CN111539507B (en) 2020-03-20 2020-03-20 Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm

Publications (2)

Publication Number Publication Date
CN111539507A CN111539507A (en) 2020-08-14
CN111539507B true CN111539507B (en) 2021-08-31

Family

ID=71969195

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010201690.2A Active CN111539507B (en) 2020-03-20 2020-03-20 Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm

Country Status (1)

Country Link
CN (1) CN111539507B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104586399A (en) * 2013-10-31 2015-05-06 卡西欧计算机株式会社 Training supporting apparatus and system for supporting training of walking and/or running
CN104888444A (en) * 2015-05-20 2015-09-09 深圳大学 Intelligent gloves, method and system for calorie consumption and hand posture recognition
CN105823483A (en) * 2016-05-11 2016-08-03 南京邮电大学 User walking location method based on inertial measurement unit
CN108564600A (en) * 2018-04-19 2018-09-21 北京华捷艾米科技有限公司 Moving object attitude tracking method and device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106846372B (en) * 2017-02-13 2020-04-03 南京升渡智能科技有限公司 Human motion quality visual analysis and evaluation system and method thereof
CN110543727B (en) * 2019-09-05 2023-01-03 北京工业大学 Improved particle swarm algorithm-based omnidirectional mobile intelligent wheelchair robot parameter identification method

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104586399A (en) * 2013-10-31 2015-05-06 卡西欧计算机株式会社 Training supporting apparatus and system for supporting training of walking and/or running
CN104888444A (en) * 2015-05-20 2015-09-09 深圳大学 Intelligent gloves, method and system for calorie consumption and hand posture recognition
CN105823483A (en) * 2016-05-11 2016-08-03 南京邮电大学 User walking location method based on inertial measurement unit
CN108564600A (en) * 2018-04-19 2018-09-21 北京华捷艾米科技有限公司 Moving object attitude tracking method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
赵汇东.基于位姿的人体运动模式识别.《中国优秀硕士学位论文全文数据库 信息科技辑》.2020,全文. *

Also Published As

Publication number Publication date
CN111539507A (en) 2020-08-14

Similar Documents

Publication Publication Date Title
Zhang et al. Accurate ambulatory gait analysis in walking and running using machine learning models
Soltani et al. Real-world gait speed estimation using wrist sensor: A personalized approach
US20140066816A1 (en) Method, apparatus, and system for characterizing gait
CN110236550B (en) Human gait prediction device based on multi-mode deep learning
Song et al. Speed estimation from a tri-axial accelerometer using neural networks
CN110811553B (en) Detection method for assistance efficiency of load exoskeleton
US20200297243A1 (en) System for estimating body motion of person or the like
Bonnet et al. Fast determination of the planar body segment inertial parameters using affordable sensors
Joukov et al. Online tracking of the lower body joint angles using IMUs for gait rehabilitation
Panero et al. Comparison of different motion capture setups for gait analysis: Validation of spatio-temporal parameters estimation
Wang et al. Swimming motion analysis and posture recognition based on wearable inertial sensors
Tien et al. Results of using a wireless inertial measuring system to quantify gait motions in control subjects
CN107194193A (en) A kind of ankle pump motion monitoring method and device
Hindle et al. Inertial-based human motion capture: A technical summary of current processing methodologies for spatiotemporal and kinematic measures
Anwary et al. Validity and consistency of concurrent extraction of gait features using inertial measurement units and motion capture system
CN110313918A (en) A kind of gait phase recognition methods and system based on plantar pressure
Florentino-Liaño et al. Human activity recognition using inertial sensors with invariance to sensor orientation
Qiu et al. Ambulatory estimation of 3D walking trajectory and knee joint angle using MARG Sensors
Ladetto et al. Human walking analysis assisted by DGPS
Lueken et al. Estimation of stride time variability in unobtrusive long-term monitoring using inertial measurement sensors
CN115655265A (en) Pedestrian navigation indoor positioning method based on inertial sensing
Alahakone et al. Smart wearable device for real time gait event detection during running
JP2020000854A (en) Methods for computing real-time step length and speed of running or walking individual
Digo et al. Gait Phases Detection in Elderly using Trunk-MIMU System.
CN111539507B (en) Recovery movement speed calculation model parameter identification method based on particle swarm optimization algorithm

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant