CN117171606A - Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning - Google Patents
Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning Download PDFInfo
- Publication number
- CN117171606A CN117171606A CN202311129673.2A CN202311129673A CN117171606A CN 117171606 A CN117171606 A CN 117171606A CN 202311129673 A CN202311129673 A CN 202311129673A CN 117171606 A CN117171606 A CN 117171606A
- Authority
- CN
- China
- Prior art keywords
- layer
- node
- representing
- adaptive fuzzy
- gait
- 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.)
- Granted
Links
- 238000012567 pattern recognition method Methods 0.000 title claims abstract description 12
- 230000005021 gait Effects 0.000 claims abstract description 40
- 238000001514 detection method Methods 0.000 claims abstract description 15
- 238000010606 normalization Methods 0.000 claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 238000012549 training Methods 0.000 claims abstract description 7
- 238000012360 testing method Methods 0.000 claims abstract description 6
- 230000003044 adaptive effect Effects 0.000 claims description 17
- 238000000034 method Methods 0.000 claims description 17
- 230000001133 acceleration Effects 0.000 claims description 14
- 238000005259 measurement Methods 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 abstract description 2
- 210000002683 foot Anatomy 0.000 description 14
- 230000003993 interaction Effects 0.000 description 5
- 238000012544 monitoring process Methods 0.000 description 5
- 230000036541 health Effects 0.000 description 4
- 238000007796 conventional method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 210000003423 ankle Anatomy 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The invention discloses a pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning, which belongs to the technical field of satellite navigation, and extracts foot sensor raw data of pedestrians in various gait patterns, wherein the foot sensor raw data comprises three-axis gyroscope and three-axis accelerometer sensor data; then, establishing a five-layer structure self-adaptive fuzzy inference system with a fuzzification layer, a regularization layer, a normalization layer, a defuzzification layer and a summation output layer for data pre-training; finally, constructing a condition test type based on generalized likelihood ratio estimation as the input of a fuzzy inference system, so as to identify and output various gait modes such as walking, running, stairs, elevators and the like; the invention can effectively detect and identify various gait modes such as walking, running, riding an elevator, stairs and the like by introducing a lightweight self-adaptive fuzzy inference system, has simple model, low calculation complexity and high detection precision, and can be applied to various intelligent terminals and wearable equipment.
Description
Technical Field
The invention belongs to the field of intelligent recognition of motion modes, and particularly relates to a pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning.
Background
The rapid, accurate and reliable pedestrian gait pattern detection and recognition plays an important role in the services of navigation positioning, exercise training, health monitoring, man-machine interaction and the like of users. Currently, pedestrian gait recognition methods mainly include two categories, vision-based and sensor-based. Although the vision-based method has higher recognition accuracy, the calculation amount is large, the calculation speed is low, and the privacy problem of the user is often involved. With the development of mems technology, pedestrian motion state and gait pattern detection based on inertial sensors becomes easier to implement.
The inertial sensor can accurately capture the motion information of acceleration, speed, displacement, attitude angle and the like of the pedestrian, and the inertial sensor arranged on the foot can further reflect the motion state of the pedestrian through the gait change of the foot step. However, since individual differences of pedestrians are obvious and gait types are wide, it is difficult for the conventional method to timely detect and recognize the current gait pattern. Once gait pattern recognition is wrong, the accumulated error of the inertial sensor is gradually increased along with the time, and various service activities such as pedestrian positioning track, health monitoring, posture correction and the like are affected.
Disclosure of Invention
In order to solve the technical problems, the invention provides a pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning, which solves the problems in the prior art, and adopts the following technical scheme:
a pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning comprises the following steps:
s1: extracting the original data of a triaxial gyroscope and an accelerometer sensor on the foot of a pedestrian;
s2: establishing a self-adaptive fuzzy inference system for gait detection and recognition;
s3: and constructing a conditional test formula based on the generalized likelihood ratio estimation as an input value of the self-adaptive fuzzy inference system so as to identify and output various gait patterns of pedestrians.
Further, in the step S1, the three-axis gyroscope and the accelerometer sensor are integrated on the foot inertial sensor, and the step S1 includes:
assume that the raw data at the kth time of the accelerometer sensor and the triaxial gyroscope are respectivelyAndconstructional data pair->The observation sequence of the foot inertial sensor within the time window w is expressed by the following formula:
wherein k represents a sampling point at the kth time, X W Representing the total data set within the observed time window,representing the real number domain.
Further, the self-adaptive fuzzy inference system comprises five layers of structures which are sequentially carried out, wherein the five layers of structures are respectively as follows:
blurring layer: performing fuzzy clustering on all original data;
regularization layer: calculating the triggering intensity of each data set of the fuzzy cluster;
normalization layer: normalizing each trigger intensity;
deblurring layer: calculating a weighted value of each rule;
summation output layer: the weighted value of each rule is summed.
Further, in the fuzzification layer, the fuzzy clustering of the original data is obtained by using a membership function, and the following formula is adopted:
wherein,is the input variable x i Is a fuzzy set of (2); the gbellmf is a generalized bell-shaped membership function; { a, b, c } is a set of precondition parameters; />Is the membership of the membership function.
Further, in the regularization layer, each fuzzy clustering data set is recorded as a pi-shaped circular node, partial fuzzy set operation is realized in the regularization layer, and the following formula is adopted:
wherein w is i Representing the trigger intensity of each fuzzy data set,a fuzzy dataset representing each node of the input.
Further, in the normalization layer, the ratio of the trigger intensity of each node in the layer to the sum of the trigger intensities of all the normalization layers is counted, and the following formula is adopted:
wherein w is i Is the trigger strength of each node,representing the total trigger strength of n nodes, < ->Representing the normalized value for each node.
Further, in the defuzzification layer, when calculating the weights on the node rules, the following formula is adopted:
wherein,representing normalized value of each node, f i Representing the input of each node, { x 1 ,x 2 ,…,x n The observation sequence of the inertial sensor input in equation (1), is +.>The back-piece parameters representing fuzzy rules, the number of result parameters per rule being one more than the number of inputs.
Further, in the summation output layer, the following formula is adopted for summation:
wherein,representing the final training result of the output,/->Defuzzified value for each node, w i Is the trigger strength of each node, f i Input value representing each node, +.>Representing the normalized value for each node.
Further, in the step S3, the method includes:
the method for jointly judging the acceleration information and the angular velocity information is selected as a judging condition of the motion state:
wherein T is k (α n ,ω n ) For the kth moment acceleration alpha n And angular velocity omega n W is the width of the time window, σ ω Standard deviation, sigma, of angular velocity measurement error a G represents gravitational acceleration, and II is a 2-norm calculation formula; mu (mu) a The acceleration average value of all samples in the time window;
when the adaptive fuzzy inference system detects T k After meeting any gait pattern, the user can pass through different detection thresholds gamma k To represent the corresponding gait pattern and to output the corresponding recognized gait pattern.
Further, in the formula (7), μ a The following formula is used for calculation:
the invention has the following beneficial effects:
(1) And improving the detection accuracy: the pedestrian multi-gait pattern recognition method based on the self-adaptive fuzzy reasoning can effectively combine the output of the accelerometer and the gyroscope, thereby greatly improving the accuracy of gait pattern detection.
(2) The operation complexity is reduced: compared with the conventional method, the method can reduce the operation complexity, effectively solve the recognition problems of obvious individual differences and various gait types of pedestrians, and can be conveniently applied to various low-cost intelligent terminals.
(3) The practicability is strong: the method can be widely applied to services such as navigation positioning, exercise training, health monitoring, man-machine interaction and the like of the user, and has the characteristic of strong practicability. At the same time, the use concerns are reduced as sensor-based methods are less involved in the privacy of the user.
(4) The invention can effectively detect and identify various gait modes such as walking, running, riding an elevator, stairs and the like by introducing the light self-adaptive fuzzy inference system, has simple model, low calculation complexity and high detection precision, and can be applied to various intelligent terminals and wearable equipment.
Drawings
FIG. 1 is a flow chart of a pedestrian multi-gait pattern recognition method based on adaptive fuzzy reasoning provided by an embodiment of the invention;
fig. 2 is a schematic diagram of an adaptive fuzzy inference system with a five-layer structure according to an embodiment of the present invention;
fig. 3 is a schematic diagram of a system for identifying multiple gait patterns based on adaptive fuzzy inference according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to fig. 1 to 3 in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, but not all embodiments, and technical means used in the embodiments are conventional means known to those skilled in the art unless specifically indicated.
As shown in FIG. 1, the invention extracts foot MEMS-IMU sensor raw data of pedestrians in various gait modes, including triaxial gyroscope and triaxial accelerometer sensor data. Then, a five-layer structure self-adaptive fuzzy inference system with a fuzzification layer, a regularization layer, a normalization layer, a defuzzification layer and a summation output layer is established for data pre-training. Finally, a conditional test formula based on generalized likelihood ratio estimation is constructed to be used as the input of a fuzzy inference system, so that various gait modes such as walking, running, stairs, elevators and the like are identified and output.
It is understood that the foot sensor in the present invention may be worn on the foot of a pedestrian, preferably on the ankle part, through a smart wearable device, a man-machine interaction device, a smart terminal, etc. The intelligent wearable device, the man-machine interaction device and the intelligent terminal are macroscopic concepts in the prior art. As long as the electronic equipment can be worn on feet, the electronic equipment for realizing the functions of various sports monitoring, health monitoring, navigation positioning and the like can be called as intelligent wearing equipment, man-machine interaction equipment and intelligent terminal.
Specifically, the method for identifying the pedestrian multi-gait pattern based on the adaptive fuzzy reasoning provided by the embodiment is described, and the method for identifying the pattern comprises the following steps:
s1: extracting foot triaxial gyroscope and accelerometer sensor original data of pedestrians in various gait modes;
s2: establishing an adaptive fuzzy inference system with a five-layer structure for gait detection and recognition;
s3: constructing a condition test type based on generalized likelihood ratio estimation as an input value of a fuzzy inference system, so as to identify and output various gait patterns of pedestrians;
specifically, the step S1 includes:
extracting foot triaxial gyroscope and accelerometer sensor original data of pedestrians in various gait modes; among other things, low cost Inertial Sensors (IMUs) typically include tri-axis gyroscopes and tri-axis accelerometers for measuring states of acceleration, velocity, vibration, and rotation of an object. Since foot-mounted inertial sensors have significant stance and swing phases, we choose foot inertial devices to extract raw sensory data of pedestrians.
The raw data at the kth moment of the accelerometer and gyroscope output by the foot inertial sensor are assumed to be respectivelyAnd->Constructional data pair->The foot inertial sensor observation sequence within the time window W is:
wherein k represents a sampling point at the kth time, X W Representing the total data set within the observed time window,representing the real number domain.
Specifically, in step S2, the adaptive fuzzy inference system includes sequentially performed five layers of structures, which are respectively: a blurring layer, a regularization layer, a normalization layer, a defuzzification layer and a summation output layer. In the invention, the adaptive fuzzy inference system is a mathematical model for identifying gait patterns.
For step S2, the following flow is specifically included:
s2-1: all data are fuzzy clustered in a first fuzzification layer.
As shown in fig. 2, to adaptively detect the asynchronous mode of pedestrians, we construct a fuzzy inference system with five layers structure based on an adaptive network framework, construct an input-output map in the form of fuzzy rules, and specify input-output data pairs. Firstly, blurring all data in a first blurring layer, and obtaining fuzzy clustering of original data input information by using a membership function:
wherein,is the input variable x i Is a fuzzy set of (2); the gbellmf is called a generalized bell membership function; { a, b, c } is a set of preconditions that determine the form of the membership function. By using these parametersThe membership degree +.>
S2-2: calculating each trigger intensity of the fuzzy clustering dataset in a second regularization layer;
each fuzzy data set is recorded as a pi-shaped circular node, and partial fuzzy set operation is realized in a second regularization layer:
wherein w is i Representing the trigger strength of each node,each node representing an input ambiguous data set.
S2-3: each trigger intensity is normalized in a third normalization layer.
The third layer structure is a normalization layer, i.e. the ratio of the trigger intensity of each node in the layer to the sum of the trigger intensities of all the normalization layers is counted:
wherein w is i Is the trigger strength of each node,representing the total trigger strength of n nodes, < ->Representing a normalized value for each node;
s2-4: calculating a weighted value of each rule in a fourth defuzzification layer;
the fourth layer structure is a defuzzification layer, which is mainly used for calculating the weights on node rules:
wherein f i Representing the input of each layer, { x 1 ,x 2 ,…,x n The observation sequence of the inertial sensor input in equation (1),a back-piece parameter representing fuzzy rules, the number of result parameters of each rule being one more than the number of inputs;
s2-5: calculating the sum of all rules in a fifth layer of summation;
the fifth layer structure is a summation layer, which sums the weighted values of each node:
wherein,representing the final training result of the output based on the adaptive fuzzy inference system, < >>Defuzzified value for each node, w i Is the trigger strength of each node, f i Input value representing each node, +.>Representing a normalized value for each node; thus, the original inertial measurement data can be successfully classified through a fuzzy inference system with a five-layer structure.
In addition, referring to fig. 2, in the fuzzy inference system with the five-layer structure, the first four layers respectively comprise n nodes, namely data nodes, so as to realize data calculation, and the node in the invention corresponds to the data node in the layer structure. The summing layer performs weighted summation of the n nodes. The data of each node is transferred to the corresponding node of the next layer.
Specifically, in the step S3, the specific flow is as follows:
constructing a condition test type based on generalized likelihood ratio estimation as an input value of a fuzzy inference system, so as to identify and output various gait patterns of pedestrians;
in order to ensure the accuracy and reliability of pattern recognition, a method for jointly judging acceleration information and angular velocity information is selected as a judging condition of a motion state:
wherein T is k (α n ,ω n ) For the kth moment acceleration alpha n And angular velocity omega n W is the width of the time window, σ ω Standard deviation, sigma, of angular velocity measurement error a G represents gravitational acceleration, and II is a 2-norm calculation formula; mu (mu) a Acceleration averages for all samples over a time window:
when the adaptive fuzzy inference system detects T k After meeting any gait modes such as walking, running, elevator riding, stair riding and the like, the user can pass through different detection threshold values gamma k To represent the corresponding gait pattern, i.e. to output the identified gait pattern.
Referring to FIG. 3, it can be seen that the detection sequence T is trained by the data of the fuzzy inference system k Can be precisely divided into different motion states such as walking, running, elevator and stairs, and the function value T in the formula (7) k And a detection threshold gamma k By contrast, the detection threshold gamma can be different k To distinguish the corresponding gait patterns.
The above embodiments are only illustrative of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention, and various modifications, variations, alterations, substitutions made by those skilled in the art to the technical solution of the present invention should fall within the protection scope defined by the claims of the present invention without departing from the spirit of the design of the present invention.
Claims (10)
1. The pedestrian multi-gait pattern recognition method based on the self-adaptive fuzzy reasoning is characterized by comprising the following steps of:
s1: extracting the original data of a triaxial gyroscope and an accelerometer sensor on the foot of a pedestrian;
s2: establishing a self-adaptive fuzzy inference system for gait detection and recognition;
s3: and constructing a conditional test formula based on the generalized likelihood ratio estimation as an input value of the self-adaptive fuzzy inference system so as to identify and output various gait patterns of pedestrians.
2. The method for identifying multiple gait patterns of a pedestrian based on adaptive fuzzy inference according to claim 1, wherein in the step S1, the three-axis gyroscope and the accelerometer sensor are integrated on the foot inertial sensor, and the step S1 comprises:
assume that the raw data at the kth time of the accelerometer sensor and the triaxial gyroscope are respectivelyAndconstructional data pair->The observation sequence of the foot inertial sensor within the time window W is expressed by the following formula:
wherein k represents a sampling point at the kth time, X W Representing the total data set within the observed time window,representing the real number domain.
3. The pedestrian multi-gait pattern recognition method based on the self-adaptive fuzzy inference as claimed in claim 2, wherein the self-adaptive fuzzy inference system comprises sequentially performed five layers of structures, respectively:
blurring layer: performing fuzzy clustering on all original data;
regularization layer: calculating the triggering intensity of each data set of the fuzzy cluster;
normalization layer: normalizing each trigger intensity;
deblurring layer: calculating a weighted value of each rule;
summation output layer: the weighted value of each rule is summed.
4. The pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning of claim 3, wherein in the fuzzification layer, fuzzy clustering of original data is obtained by using a membership function, and the following formula is adopted:
wherein,is the input variable x i Is a fuzzy set of (2); the gbellmf is a generalized bell-shaped membership function; { a, b, c } is a set of precondition parameters; />Is the membership of the membership function.
5. The method for identifying the pedestrian multi-gait pattern based on the self-adaptive fuzzy reasoning according to claim 3, wherein in the regularization layer, a data set of each fuzzy cluster is recorded as a pi-shaped circular node, partial fuzzy set operation is realized in the regularization layer, and the following formula is adopted:
wherein w is i Representing the trigger strength of each node,a fuzzy dataset representing each node of the input.
6. The method for identifying the pedestrian multi-gait pattern based on the self-adaptive fuzzy inference as claimed in claim 3, wherein in the normalization layer, the ratio of the trigger intensity of each node in the layer to the sum of the trigger intensities of all the normalization layers is counted, and the following formula is adopted:
wherein w is i Is the trigger strength of each node,representing the total trigger strength of n nodes, < ->Representing the normalized value for each node.
7. The method for identifying multiple gait patterns of a pedestrian based on adaptive fuzzy inference as claimed in claim 3, wherein the following formula is adopted when the weights on the node rules are calculated in the defuzzification layer:
wherein,representing normalized value of each node, f i Representing the input of each node, { x 1 ,x 2 ,…,x n The observation sequence of the inertial sensor input in equation (1), is +.>The back-piece parameters representing fuzzy rules, the number of result parameters per rule being one more than the number of inputs.
8. The pedestrian multi-gait pattern recognition method based on adaptive fuzzy reasoning of claim 3, wherein the summation output layer sums with the following formula:
wherein,representing the final training result of the output,/->For each layer of defuzzified values, w i Is the trigger strength of each node, f i Input value representing each node, +.>Representing the normalized value for each node.
9. The method for identifying multiple gait patterns of a pedestrian based on adaptive fuzzy inference according to claim 1, wherein the step S3 comprises:
the method for jointly judging the acceleration information and the angular velocity information is selected as a judging condition of the motion state:
wherein T is k (α n ,ω n ) For the kth moment acceleration alpha n And angular velocity omega n W is the width of the time window, σ ω Standard deviation, sigma, of angular velocity measurement error a G represents gravitational acceleration, and II is a 2-norm calculation formula; mu (mu) a The acceleration average value of all samples in the time window;
when the adaptive fuzzy inference system detects T k After meeting any gait pattern, the user can pass through different detection thresholds gamma k To represent the corresponding gait pattern and to output the corresponding recognized gait pattern.
10. The method for identifying multiple gait patterns of a pedestrian based on adaptive fuzzy inference of claim 9, wherein μ is μ in the formula (7) a The following formula is used for calculation:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311129673.2A CN117171606B (en) | 2023-09-04 | 2023-09-04 | Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311129673.2A CN117171606B (en) | 2023-09-04 | 2023-09-04 | Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117171606A true CN117171606A (en) | 2023-12-05 |
| CN117171606B CN117171606B (en) | 2024-05-28 |
Family
ID=88929271
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311129673.2A Active CN117171606B (en) | 2023-09-04 | 2023-09-04 | Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117171606B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217107A (en) * | 2014-08-27 | 2014-12-17 | 华南理工大学 | Method for detecting tumbling state of humanoid robot based on multi-sensor information |
| CN107958221A (en) * | 2017-12-08 | 2018-04-24 | 北京理工大学 | A kind of human motion Approach for Gait Classification based on convolutional neural networks |
| CN109297485A (en) * | 2018-08-24 | 2019-02-01 | 北京航空航天大学 | A kind of personal inertial navigation height accuracy method for improving in interior based on height from observation algorithm |
| CN110464315A (en) * | 2019-07-23 | 2019-11-19 | 闽南理工学院 | It is a kind of merge multisensor the elderly fall down prediction technique and device |
| CN111568432A (en) * | 2020-04-06 | 2020-08-25 | 沈阳工业大学 | Fall detection method of omnibearing walking training robot based on fuzzy inference |
| US20210052196A1 (en) * | 2019-08-21 | 2021-02-25 | The Swatch Group Research And Development Ltd | Method and system for gait detection of a person |
| CN112440267A (en) * | 2020-11-27 | 2021-03-05 | 北京精密机电控制设备研究所 | Gait phase identification method based on inertial sensor |
| CN115167234A (en) * | 2022-07-29 | 2022-10-11 | 哈尔滨工业大学 | Multi-sensor information fusion gait recognition method of humanoid foot and foot plate biped robot |
-
2023
- 2023-09-04 CN CN202311129673.2A patent/CN117171606B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104217107A (en) * | 2014-08-27 | 2014-12-17 | 华南理工大学 | Method for detecting tumbling state of humanoid robot based on multi-sensor information |
| CN107958221A (en) * | 2017-12-08 | 2018-04-24 | 北京理工大学 | A kind of human motion Approach for Gait Classification based on convolutional neural networks |
| CN109297485A (en) * | 2018-08-24 | 2019-02-01 | 北京航空航天大学 | A kind of personal inertial navigation height accuracy method for improving in interior based on height from observation algorithm |
| CN110464315A (en) * | 2019-07-23 | 2019-11-19 | 闽南理工学院 | It is a kind of merge multisensor the elderly fall down prediction technique and device |
| US20210052196A1 (en) * | 2019-08-21 | 2021-02-25 | The Swatch Group Research And Development Ltd | Method and system for gait detection of a person |
| CN111568432A (en) * | 2020-04-06 | 2020-08-25 | 沈阳工业大学 | Fall detection method of omnibearing walking training robot based on fuzzy inference |
| CN112440267A (en) * | 2020-11-27 | 2021-03-05 | 北京精密机电控制设备研究所 | Gait phase identification method based on inertial sensor |
| CN115167234A (en) * | 2022-07-29 | 2022-10-11 | 哈尔滨工业大学 | Multi-sensor information fusion gait recognition method of humanoid foot and foot plate biped robot |
Non-Patent Citations (3)
| Title |
|---|
| XU SU ETC.: "A novel gait analysis system based on adaptive neuro-fuzzy inference system", EXPERT SYSTEMS WITH APPLICATIONS, 1 March 2010 (2010-03-01) * |
| 赵东辉;杨俊友;白殿春;姜银来;: "基于节点迭代模糊Petri网的非接触异常步态识别方法", 仪器仪表学报, no. 04, 15 April 2019 (2019-04-15) * |
| 钱兴: "基于多传感器信息的步态识别方法研究", 中国优秀硕士论文电子期刊网, 15 January 2023 (2023-01-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117171606B (en) | 2024-05-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP4243684B2 (en) | Walking motion detection processing device and walking motion detection processing method | |
| CN109522793A (en) | More people's unusual checkings and recognition methods based on machine vision | |
| AU2020103811A4 (en) | Method for recognizing human knee motion postures based on extreme learning machine | |
| CN106846729A (en) | A kind of fall detection method and system based on convolutional neural networks | |
| CN107092894A (en) | A kind of motor behavior recognition methods based on LSTM models | |
| CN108245172A (en) | It is a kind of not by the human posture recognition method of position constraint | |
| CN104586402B (en) | A kind of feature extracting method of physical activity | |
| CN108256238A (en) | A kind of optic fiber grating wavelength demodulation method and device based on deep learning | |
| Yang et al. | Human posture recognition in intelligent healthcare | |
| CN112464738A (en) | Improved naive Bayes algorithm user behavior identification method based on mobile phone sensor | |
| Shi et al. | Fall detection system based on inertial mems sensors: Analysis design and realization | |
| Al Jeroudi et al. | Online sequential extreme learning machine algorithm based human activity recognition using inertial data | |
| Al Abiad et al. | Smartstep: A robust STEP detection method based on SMARTphone inertial signals driven by gait learning | |
| CN117171606B (en) | Pedestrian multi-gait pattern recognition method based on self-adaptive fuzzy reasoning | |
| CN114881079A (en) | Human body movement intention abnormity detection method and system for wearable sensor | |
| Shi et al. | Sensor‐based activity recognition independent of device placement and orientation | |
| CN111603169A (en) | Pedestrian gait recognition method based on MEMS inertial sensor | |
| CN111623797A (en) | Step number measuring method based on deep learning | |
| Omae et al. | Detection of swimming stroke start timing by deep learning from an inertial sensor | |
| Maldonado-Mendez et al. | Fall detection using features extracted from skeletal joints and SVM: Preliminary results | |
| CN116465412A (en) | Improved PDR indoor positioning method based on LSTM and attention mechanism | |
| Imran et al. | Harresnext: An efficient resnext inspired network for human activity recognition with inertial sensors | |
| CN109029432B (en) | Human body action detection and identification method based on six-axis inertial sensing signal | |
| Lin et al. | Deep learning-based multifloor indoor tracking scheme using smartphone sensors | |
| Lin et al. | LocMe: Human locomotion and map exploitation based indoor localization |
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 |