KR102085372B1 - Method and apparatus for estimating walking state - Google Patents

Abstract

Disclosed is a walking state estimation method and apparatus. The disclosed walking state estimation method includes: obtaining an angle value of the knee joint of the user through a sensor attached to the foot or leg of the user; Determining a pole change pattern for the knee joint based on the angle value of the knee joint; And estimating the walking step of the user using the pole change pattern and the state machine.

Description

METHOOD AND APPARATUS FOR ESTIMATING WALKING STATE}

The following description relates to a walking state estimation method and apparatus.

The prosthesis is an assistive device used by people whose legs have been cut off due to a congenital cause or an acquired accident. The prosthesis is divided into femoral limb and lower limb according to the cut site. Femoral prosthesis is the leg used by sub-knee cuttings, and lower leg prosthesis is the prosthesis used by sub-elevation. Due to this cutting position, users of lower limb prostheses do not have much inconvenience in walking with assistive devices. However, in the case of thigh-foot wearers, adapting to the foot does not feel much difficulty in walking on the flat, but complains of inconvenience in walking in a walking environment such as a high slope or stairs.

In order to solve this problem, the current femoral leg research trend is being developed from passive femoral leg to intelligent femoral leg. Passive femoral limbs are flexion and extension of the knee joint by mechanical continuous motion, and intelligent femoral limbs control the actuator to reproduce the flexion and extension of the knee joint at the desired joint angle.

  In order to change the joint angle of intelligent limb according to the walking environment, the mode was changed with the remote controller in the past. However, the recent research trend is the type of sensor that is equipped with a prosthetic foot and actively estimates the walking environment, and moves the knee joint to the walking trajectory corresponding to the walking environment. However, the drawback of this type of research is that it is slow to respond to the pedestrian environment because it is possible to estimate the pedestrian environment only when the thigh prosthesis enters the pedestrian environment.

The present invention can provide a method and apparatus for intelligently changing the walking mode autonomously by actively grasping the flat walking environment and the ramp walking environment.

The present invention can use the data generated from the strain gauge in the inertial sensor module mounted on or attached to the foot and the fastening portion for fastening the artificial foot with the thigh foot. The present invention classifies the ground repulsion force estimated by the strain gauge according to the level and the slope, and classifies it based on a predetermined threshold value, and subdivides the walking step based on the angle value of the knee joint according to the progression of walking with the inertial sensor. The walking environment can be detected based on the characteristics of the walking step derived from the different characteristics.

According to one or more exemplary embodiments, a method of estimating a walking state may include obtaining an angle value of a knee joint of a user through a sensor attached to a foot or leg of a user; Determining a pole change pattern for the knee joint based on the angle value of the knee joint; And estimating the walking step of the user using the pole change pattern and the state machine.

In the method for estimating walking state according to an embodiment, estimating a walking step of the user may include: a characteristic in which four poles exist in one walking; A characteristic in which the pole change pattern has periodicity; And a walking step of the user, based on at least one property of the current step of the walking step being associated with at least one of a previous step and a subsequent step.

In the walking state estimating method according to an embodiment, the walking step may be divided based on the maximum bending and the maximum extension included in the pole change pattern.

The walking state estimation method may further include estimating a walking environment of the user based on the walking step.

The estimating of the walking environment in the walking state estimating method according to an embodiment may estimate the walking environment as one of flat, ramp up, and ramp down based on the characteristics of the pole generated in the estimated walking step. have.

In the method for estimating a walking state according to an embodiment, the pole change pattern may have a characteristic according to at least one of a walking step and a walking environment of the user.

In the walking state estimating method according to an embodiment, the walking step may include an initial folding device, an intermediate standing device, a late standing device, and an intermediate embossing device.

In one embodiment, an apparatus for estimating walking state may include a processor; And a memory including at least one instruction executable by the processor, wherein when the at least one instruction is executed in the processor, the processor uses a sensor attached to a foot or leg of the user to monitor the knee joint of the user. Obtain an angular value of, determine a pole change pattern for the knee joint based on the angular value of the knee joint, and estimate the gait step of the user using the pole change pattern and a state machine.

According to an embodiment, the intelligent femoral limb may provide a method and apparatus for autonomously changing a walking mode by actively identifying a flat walking environment and a ramp walking environment.

According to an embodiment, the data generated from the strain gauge may be used for the inertial sensor module mounted on or attached to the foot and the fastening part for fastening the artificial foot with the thigh foot. According to an embodiment, the ground repulsion force estimated by the strain gauge is classified according to a flat surface and a slope, and classified based on a predetermined threshold value, and the walking step is subdivided based on the angle value of the knee joint according to the progress of walking with the inertial sensor. The walking environment can be detected based on the characteristics of the walking step, which are derived from the different characteristics of the and ramps.

1 is a diagram illustrating a walking state estimating method according to at least one example embodiment.
2 is a view for explaining a walking step according to the angle of the knee joint according to an embodiment.
3 is a diagram illustrating a flat walking trajectory according to an exemplary embodiment.
4 is a diagram illustrating an uphill walking trajectory according to an exemplary embodiment.
5 is a diagram illustrating a downhill walking trajectory according to an exemplary embodiment.
6 is a diagram for describing a pole, according to an exemplary embodiment.
7 is a diagram for describing a walking step, according to an exemplary embodiment.
8 is a diagram illustrating a walking state estimation system, according to an exemplary embodiment.
9 is a diagram illustrating a walking state estimation system according to another exemplary embodiment.
10 is a diagram illustrating a walking state estimation apparatus, according to an exemplary embodiment.

Specific structural or functional descriptions of the embodiments are disclosed for purposes of illustration only, and may be practiced in various forms. Accordingly, the embodiments are not limited to the specific disclosure, and the scope of the present specification includes changes, equivalents, or substitutes included in the technical idea.

Terms such as first or second may be used to describe various components, but such terms should be interpreted only for the purpose of distinguishing one component from another. For example, the first component may be referred to as a second component, and similarly, the second component may also be referred to as a first component.

When a component is referred to as being "connected" to another component, it should be understood that there may be a direct connection or connection to that other component, but there may be other components in between.

Singular expressions include plural expressions unless the context clearly indicates otherwise. As used herein, the terms "comprise" or "have" are intended to designate that the described feature, number, step, operation, component, part, or combination thereof exists, but includes one or more other features or numbers, It should be understood that it does not exclude in advance the possibility of the presence or addition of steps, actions, components, parts or combinations thereof.

Unless defined otherwise, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art. Terms such as those defined in the commonly used dictionaries should be construed as having meanings consistent with the meanings in the context of the related art and, unless expressly defined herein, are not construed in ideal or excessively formal meanings. Do not.

Embodiments to be described below can be used to control the voltage. Hereinafter, exemplary embodiments will be described in detail with reference to the accompanying drawings. Like reference numerals in the drawings denote like elements.

1 is a diagram illustrating a walking state estimating method according to at least one example embodiment.

According to an embodiment, the gait stage may be estimated based on the change of the knee joint, and the walking environment of the intelligent thigh leg may be estimated based on the estimated gait stage. An inertial sensor that can be easily attached to the body may be used for estimating knee change. There are four poles in the knee joint change in gait. These pole change patterns show reproducibility in the same walking environment, and have different patterns in walking environments of different situations. This pole change pattern may appear with periodicity in walking. When the pole change pattern is subdivided in stages, the current stage has a characteristic of being correlated with the previous and subsequent stages. In order to estimate the walking environment using the characteristics of the pole change pattern, the walking step may be estimated first. In other words, the gait step may be estimated using a pole change pattern and a state machine according to the angle change of the knee joint, and the walking environment may also be estimated based on this.

According to an embodiment of the present disclosure, a method and an apparatus for estimating a walking state that can more flexibly support walking by estimating the walking environment before entering the changed walking environment using the healthy side information are described. Various sensors, such as body pressure insoles and inertial sensors, may be used without limitation to obtain walking information on the healthy side.

Human walking generally has the characteristic that the walking trajectory of the knee joint varies according to the walking environment. However, even though the walking trajectory has a different walking trajectory, the walking trajectory according to the walking environment is constantly reproduced. This may be the point of change of the joint angle resulting from bending and extension of the knee joint. This point may mean that the angle change sign is changed from negative to positive and positive to negative. This changing point may be referred to as a pole. In general gait, the ideal knee joint trajectory may have four poles in one gait, and the gait step may be estimated based on this characteristic. Each pole may be associated with a change in the before / after trajectory. The walking step can be estimated based on state machines and poles based on these characteristics.

Referring to FIG. 1, a walking state estimation method performed by a processor of a walking state estimation apparatus according to an embodiment is illustrated. Here, the walking state may mean a walking step or a walking environment.

In operation 110, the walking state estimating apparatus obtains an angle value of the knee joint of the user through a sensor attached to the foot or leg of the user.

For example, the sensor may comprise an inertial sensor. The inertial sensor can be composed of three-axis accelerometer and three-axis gyro sensor, and it can measure linear acceleration and angular velocity of each axis. Using this, position, velocity, posture, etc. can be estimated. The inertial sensor is attached to the thigh about 5cm above the knee and about 5cm below the knee, and the knee plane is estimated using the difference between the two inertial sensor angles. The knee angle appearing in Fig. 1 can be measured. In some embodiments, a low frequency filter may be applied to the inertial sensor data and the differential data to reduce malfunction.

In operation 120, the walking state estimating apparatus determines a pole change pattern for the knee joint based on the angle value of the knee joint. The pole change pattern according to an embodiment may have characteristics according to at least one of a walking step and a walking environment of the user.

In step 130, the walking state estimating apparatus estimates the walking step of the user by using the pole change pattern and the state machine. According to an exemplary embodiment, an apparatus for estimating a walking state includes a characteristic in which four poles exist in one walk, a characteristic in which the pole change pattern is periodic, and a current step of the walking step is associated with at least one of a previous step and a subsequent step. The gait step of the user may be estimated based on at least one of the characteristics having a. The walking step may be classified based on the maximum bending and the maximum extension included in the pole change pattern. The walking stage may be classified into an initial folding stage, an intermediate standing stage, a late standing stage, and an intermediate stage stage.

는 현재 슬관절 각도,

는 입각기와 유각기를 구분하기 위한 기준각,

는 S_i 보행단계에서 S_j로 변한 시점의 슬관절 각도를 나타낼 수 있다.Table 1 according to an embodiment may be a walking step estimation condition. S1 may represent an initial stator in the initial grounder, S2 may represent a terminal stancer in the intermediate stance, S3 may be a center stir in the late stance, and S4 may represent an initial ground in the middle stancer.

Is the current knee angle,

Is the reference angle for distinguishing the standing and staging periods,

May represent the knee angle at the point of time changed to S _j in the walking step S _i .

In operation 140, the walking state estimating apparatus may estimate the walking environment of the user based on the walking step. According to an exemplary embodiment, an apparatus for estimating walking state may estimate the walking environment as one of flat, ramp up, and ramp down based on the characteristics of the pole generated in the estimated walking step.

According to an exemplary embodiment, an apparatus for estimating walking state uses characteristics in which four poles exist in one step, characteristic points according to a walking step and a walking environment, and each pole has an associated characteristic. Can be. In other words, the walking state estimating apparatus may estimate the walking step by using the pole of the knee angle, the knee angle, the knee angle at the previous pole, the state machine, and further estimate the walking environment.

2 is a view for explaining a walking step according to the angle of the knee joint according to an embodiment.

According to an embodiment, the gait step may be divided into four steps according to flexion and extension, which are characteristics of the knee joint.

The human walking described above may be composed of one organic motion, not different motions. The walking foot can reach the ground, transfer weight, move the weight to the next foot to move forward, and take off the opposite foot and bring it back to the front. This can be seen as a sequential logic in which past inputs and outputs continue to affect the output, a feature similar to a finite state machine. Through this, the walking step according to the change of the knee joint can be estimated based on the finite state machine.

According to an exemplary embodiment, a walking cycle may be largely divided into a stance and a swing stage. For example, in one walking cycle, the staging and staging vessels may have a ratio of 60% to 40%. . The stance represents the total time the foot is in contact with the ground: initial contact (IC), load response (LR), mid stance (M.St), mid stance (T.St) , Terminal stance), and pre-swing (PS) can be divided into five stages. Standing is the time when the foot is floating in the air for the body to move forward. The initial stinger (I.Sw, Initial Swing), the middle stinger (M.Sw, Mid Swing), and the final stinger (T.Sw, Terminal) Swing) can be divided into three stages.

초기접지기,

중간입각기,

말기입각기,

중간유각기로 구분이 될 수 있다. 일실시예에 따른 보행 주기에서 입각기는 초기접지기, 중간입각기, 말기입각기의 3단계로 구성되고, 유각기는 중간입각기로 1단계로 구성될 수 있다.During the walking cycle, the knee angle (ie, knee joint angle) occurs twice in flexion and extension, as shown in FIG. 2. Up to 5 degrees, flexion can occur at 40-70% to 65 degrees, and at 70-97% to 2 degrees of extension. The walking steps at the maximum extension and the maximum bending of FIG. 2 are respectively

Early Folding,

Middle Position,

Horse,

It can be classified as a middle basin. In the walking cycle according to an embodiment, the standing unit may be composed of three stages of initial folding, middle standing, and final standing, and the rising stage may be configured in one step as an intermediate standing.

According to an embodiment, the pole may mean a point where the derivative is zero in the range of possible derivatives. The pole can be detected when the change in slope changes from increasing to decreasing or from decreasing to increasing. In the walking phase, the maximum flexion and the maximum temple may be effective elements for distinguishing the walking phase or the walking environment. Maximum flexion and maximum extension can be effectively detected through the pole detector.

A finite state machine according to one embodiment represents a machine with a finite number of states that can exist, and can be a computational model consisting of finite states and transformations between these states. Finite state machines can be used to define the state of a machine to a particular state. The finite state machine can determine the state based on the data of the current state and the previous state when it is difficult to distinguish the states by the current data value alone. In the case of walking, it is difficult to grasp the walking stage only by the current knee angle, and since the sequential movement occurs in the case of normal walking, the walking stage can be effectively detected by using a finite state machine.

For example, referring to FIG. 2, the change in the knee angle according to walking becomes the maximum extension at the initial folding, and the knee angle may be about 5 degrees. After the initial folding of paper, a small bend occurs in the middle standing, and the knee angle may be about 18 degrees. In the midposition, the end position is the maximum extension again, the knee angle is about 5 degrees. Thereafter, the maximum extension occurs during the walking period from the late standing angle to the middle stipple, and the knee angle may be about 65 degrees. Afterwards, it may return to the initial grounding and the maximum extension may occur. Such walking characteristics may be an important factor for distinguishing walking steps.

It is difficult to detect using the commonly used threshold values because the initial grounding and the late standing angle occur in the 5 degree range, and the maximum extension and the maximum bending value vary according to the pedestrian or walking environment. Therefore, in the present specification, a method of using a finite state machine in which pole detection is performed at a maximum extension point and a maximum bending point and a pole detected point is configured by a state transformation calculation model may be used.

3 is a diagram illustrating a flat walking trajectory according to an exemplary embodiment.

Referring to FIG. 3, a walking trajectory on a flat surface is shown according to an embodiment.

The graph shown in FIG. 3 may be expressed as a percentage of the period of walking by normalizing the time-based walking data. In other words, the x-axis represents the trajectory of the knee joint change during the standing and walking phases of the gait as the ratio of the entire section. The y-axis represents the angle of flexion and extension of the knee joint in degrees. 4 and 5, the graph may be shown on the same basis.

내지

구간은 슬관절의 최대 굴곡 및 최대 신전을 의미하며, 일실시예에 따른 보행 단계를 8단계로 구분하면,

~

구간은 초기 접지기부터 부하반응기,

~

구간은 중간입각기부터 말기입각기 초기까지,

~

구간은 말기입각기 중기부터 초기유각기까지,

~

구간은 중간유각기부터 말기유각기까지를 나타낼 수 있다. 해당 구간이 슬관절의 운동 방향의 전환 시점이기 때문에, 보행 궤적에서 굴곡으로부터 신전으로 변화 구간이 정의될 수 있다.Shown in FIG. 3

To

The interval means the maximum flexion and the maximum extension of the knee joint, and if the walking step according to one embodiment divided into eight steps,

To

The interval is from the initial grounder to the load reactor,

To

The interval is from the middle stand to the beginning of the late stand,

To

The interval is from the late stages to the early stages,

To

The interval may represent the middle basin to the final basin. Since the corresponding section is a transition point of the direction of movement of the knee joint, a section of change from bending to temple in the walking trajectory may be defined.

4 is a diagram illustrating an uphill walking trajectory according to an exemplary embodiment.

Referring to FIG. 4, a walking trajectory on an uphill road is illustrated, according to an exemplary embodiment.

4 is based on the difference in the height of the forefoot and the back foot relative to the center of the body according to the angle of the ground, the forefoot may be based on the high and the rear foot relative to the center of the body. In the case of an uphill climb, the front ground is at a higher position than the flat ground during the initial folding, so that the foot touches the ground at a higher position than the flat ground. In the late stage, the rear ground is lower than in the case of flat, so it needs to be the maximum temple to transmit the force forward. Due to these characteristics, the walking trajectory of FIG. 4 may appear.

번 위치인 말기유각기로부터 초기접지기까지 신전이 작게 발생했으며

번 위치인 중간입각기에서 굴곡이 더 높게 발생할 수 있다. 그리고,

번과

번 위치에서는 평지의 보행 궤적과 큰 차이점이 나타나지 않는다. According to FIG. 4, in the case of a change in the knee joint occurring in an uphill walk in comparison with the flat walk,

From the late stage basin to the initial fold,

Higher bends may occur in the middle position, in position. And,

Burn

In position 1, there is no significant difference with the walking trajectory of the plain.

As described above, the road surface in the direction of walking is located higher than the previous road surface, because the angle of the hip joint is increased to ground the foot, and the knee joint is not fully extended in the stilts to match the inclination angle of the foot. Features may appear.

5 is a diagram illustrating a downhill walking trajectory according to an exemplary embodiment.

Referring to FIG. 5, the walking trajectory at the downhill according to an embodiment is illustrated.

The feature points shown in FIG. 5 may be based on the difference in the height of the forefoot and the back foot with respect to the center of the body according to the angle of the ground as in the case of the uphill, but the specific aspects may be different. For example, in the case of downhill, the downhill may be based on the characteristics of the lower paw and the higher hind foot based on the center of the body. During the initial folding, the front ground is located at a lower position than the flat ground, so the maximum extension is achieved, and the foot touches the ground at a lower position than the flat ground. In late standing, the back ground is at a higher position relative to the flat land. Because it is downhill, it does not need to be fully extended to transmit the force forward, and the knee angle can be increased compared to the flat to prevent it from falling forward by becoming the maximum extension. Due to these characteristics, the walking trajectory of FIG. 5 may appear.

번 위치에서는 큰 차이가 없으나,

번 위치인 중간 입각기에서 굴곡이 더 높은 위치에서 발생되며,

번 위치에서는 신전이 작게 발생되며,

번 위치에서는 평지 보행 궤적보다 슬관절의 굴곡이 일부 높게 발생될 수 있다.According to FIG. 5, the ramp downhill walking is compared with flat walking

There is no big difference in position 1,

In the middle standing position, the bend occurs at the higher position,

In the second position, the temple is small,

In position 1, the knee flexion may occur higher than the flat walking trajectory.

As mentioned earlier, when the temple is placed on an initial fold, such as a plain, on a sloped downhill walk, the temple is moved from the middle position to move the body's weight back to the back of the ground to prevent the center of the body from shifting down the slope. This characteristic may appear as the value of M is reduced. In addition, there is a feature that the knee angle is partially increased in the midglow. This may be because the foot is higher than the center of the body and the height of the leg is adjusted through the flexion of the knee so that the foot does not touch the ground when the gait is in swing.

3 to 5, the flat, the slope down, and the case of the slope up as a whole, it may be the same point that the bending and extension is repeated twice in one cycle of walking in each walking environment. Thus, walking in each walking situation may have a continuous sequence. However, the shape of flexion and temple are different in each walking environment.

,

,

,

의 구간에서 평지를 기준으로 경사로 상행, 경사로 하행에서 유의미한 변화가 나타날 수 있으며, 이를 이용하여 보행 단계뿐만 아니라 보행 환경이 추정될 수 있다.Therefore, the point of change of the knee joint in each walking environment

,

,

,

Significant changes may occur in ascending and descending slopes on the basis of flats in the interval of, and the walking environment as well as the walking phase may be estimated using this.

6 is a diagram for describing a pole, according to an exemplary embodiment.

Referring to FIG. 6, a maximum point and a minimum point are shown according to an embodiment.

According to one embodiment, the bending point and the change point of the extension of the knee joint may correspond to poles in the second plane. Here, the pole means a point where the derivative coefficient becomes '0' in the range where the derivative is possible, and also means a place where the slope of the graph becomes zero in the second plane. In the case of a digital system, it may not be easy to detect the point where the derivative value becomes '0', so that the pole can be detected by detecting a section in which the sign of the derivative value changes from positive to negative or from negative to positive. Can be.

The pole can be divided into the maximum and the minimum in detail. In the case of the maximum, the sign of the derivative value changes from positive to negative. The minima point indicates that the sign of the derivative value changes from negative to weak. In other words, the maximum point may be a point that changes from a flexion to a temple, and the minimum point may be a point that changes from a temple to a flexion. Therefore, the walking step can be detected based on the bending and extension of the walking.

Such flexion and extension may be detected based on the threshold. However, in the case of a method of detecting a walking step using a threshold value, it is not possible to detect the occurrence of a specific range of meaningful other walking steps. There is also a method of using multiple thresholds, but it may not be possible to detect a particular step if the thresholds overlap. Therefore, in the case of walking, since different steps are made in a specific range, the walking step needs to be performed based on pole detection.

7 is a diagram for describing a walking step, according to an exemplary embodiment.

Referring to FIG. 7, four walking steps are shown, according to one embodiment.

The normal gait according to an embodiment has a certain pattern, and if it deviates from the regular pattern, it may be regarded as an abnormal gait or a pathological gait. Therefore, walking can occur in order to be seen as normal walking.

Based on this characteristic, the feature points used for the gait step detection may correspond to the maximum and minimum points of the knee angle. In other words, the walking step can be detected based on the maximum and minimum points. During one walking cycle, the minimum and maximum points for the knee angle may occur twice. The first minima and the second minima are divided into different walking stages, and the first and second minima may also be classified into different walking stages. Thus, the walking steps can be divided based on the state transition diagram of the finite state machine. The state transition diagram may be configured as shown in FIG. 7 based on the feature points derived through the gait analysis.

8 is a diagram illustrating a walking state estimation system, according to an exemplary embodiment.

Referring to FIG. 8, a walking state estimation system for estimating a walking step according to an embodiment is shown.

According to an exemplary embodiment, the walking state estimation system may estimate a walking step by detecting a singular point using a knee angle and a pole (maximum point, minimum point) as parameters. In this case, an exponentially weighted filter and a state machine may be used to remove noise depending on the performance or frequency characteristics of the sensor for high reliability.

로 정의될 수 있다. The knee angle in the intelligent prosthesis is measured by the inertial sensor, and the current knee angle measured by the inertial sensor is

It can be defined as.

Gait repeatedly touches and falls off the ground, which can act as noise to the inertial sensors. Since such noise is a momentary impact, it is determined to be a high frequency noise and a low pass filter may be required to remove the noise. The low pass filter may be applied with various filters such as an exponentially weighted moving average (EWMA).

The exponentially weighted moving average filter is a first-order low-frequency filter, which is expressed by Equation 1 below.

는 가중치이며 0<

<1인 상수이다. 이는 평균필터와 유사하지만 평균필터에서는 가중치를 지정할 수 없고, 데이터 개수에 따라 자동으로 결정되는 단점이 있기 때문에 다를 수 있다. 또한, 최신 데이터에 더 높은 가중치가 부여되기 때문에 실제 데이터의 민감도가 낮을 수 있다.here

Is a weight and 0 <

Is a constant of <1. This is similar to the average filter, but the average filter can not be assigned a weight, it may be different because there is a disadvantage that is automatically determined according to the number of data. In addition, since the higher weight is given to the latest data, the sensitivity of the actual data may be low.

에 수학식 1의 1차 저주파 통과 필터를 거친 신호를

로 정한다.therefore,

The signal passed through the first order low pass filter

Decide on

라고 정한다. 여기서

은 관성데이터로 데이터를 취득하는 시간이다.In order to detect the pole, the walking angle must be differentiated, and the derivative is performed as shown in Equation 2, and the differential walking angle is

It is decided. here

Is the time for acquiring data as inertia data.

It may be necessary to determine whether the detected pole is a local maximum or local minimum. The pole is a case where the derivative value is '0', but in the case of a digital system, such a situation rarely appears, and thus it may be determined by the discriminant equation (3).

는

를 미분한 미분 값이다.Here, LocalMax is the maximum point and the maximum point to bend in the temple, LocalMin is the minimum point and the minimum point to the temple in the bend.

Is

Is the derivative value of the derivative.

값으로 판별되며, 이 값은 이전 보행주기에서 입각기 동안의 극대값으로 결정될 수 있다. 정상적인 보행은 순차적인 보행 패턴을 가지기 때문에, 이와 같은 검출조건이 설정될 수 있다.Therefore, the minimum value and the maximum value can be detected from the walking trajectory of the knee joint based on Equation 3, and when applied to the state machine of FIG. This can be derived. S1 may be a minimum point appearing after the initial folding, S2 may be a maximum point appearing after the intermediate stance, S3 may be a minimum point appearing after the late stance, and S4 may be a maximum point appearing after the intermediate stance. Therefore, the detection condition of S1 is that the current walking trajectory is the minimum point of the standing state, and the previous walking step may be S4. The detection condition of S2 is that the current walking trajectory is the maximum point of the standing state, and the previous walking step may be S1. The detection condition of S3 is that the current walking trajectory is the minimum point of the standing state, and the previous walking step may be S2. The detection condition of S4 is that the current walking trajectory is the maximum point of the stipple phase, and the previous walking step may be S3. In the current walking trajectory,

Determined by the value, this value may be determined as the maximum value during the standing period in the previous walking cycle. Since the normal walking has a sequential walking pattern, such a detection condition can be set.

Table 2 below shows pole-based walking step detection conditions.

The S1 phase begins at the minimum and ends at the maximum, and knee angles can occur mainly below 40 degrees. By continuity with the previous walking step, the previous step may be S4. Therefore, the detection equation is as follows.

는 현재 무릎 각도를 의미한다.

는 극대, 극소를 판별하는 조건식일 수 있다.Here, State (n-1) means the previous step,

Means current knee angle.

May be a conditional expression for determining the maximum and the minimum.

The S2 phase begins at the maximum and ends at the minimum and knee angles can occur below 40 degrees on average. By the continuity with the previous walking step, the previous step may be S1. Therefore, the detection equation is as follows.

The S3 phase starts at the minimum and ends at the maximum, which can occur at an average of 40 degrees or less for the knee angle. By continuity with the previous walking step, the previous step is S2. Therefore, the detection equation is as follows.

The S4 phase begins at the maximum and ends at the minimum, and in the case of knee angles, it can begin at an average of 40 degrees or more. By the continuity with the previous walking step, the previous step may be S3. Therefore, the detection equation is as follows.

In an actual walking step determination system, the noise factor is a non-negligible factor that causes system malfunction and detection. Noise components generated in the system include noise generated by the acceleration sensor and gyro sensor, noise introduced into the inertial measurement device due to the impact generated from the initial grounder, which is the starting point of the stance during walking, the inertial measurement device case and the internal module. There is a shaking due to the separation of high frequency noise. Such high frequency noise may cause a decrease in the reliability of the walking step determination.

According to an embodiment, the walking state estimation system may improve accuracy by applying a low pass filter to the knee angle estimating step and the derivative filter.

9 is a diagram illustrating a walking state estimation system according to another exemplary embodiment.

Referring to FIG. 9, a walking state estimation system for estimating a walking environment according to an embodiment is illustrated.

During walking, weight acceptance, single limb support, and limb advancement are performed, and the walking trajectory required at the walking stage may vary depending on the walking environment. have. Therefore, in order to detect a walking environment, it is necessary to use the singularity of the walking step which arises according to a walking environment.

In FIG. 8, the walking stage was estimated as one of four stages, namely, initial folding, middle standing, terminal standing, and middle recess based on the change in the joint angle. The characteristics of the walking environment for each walking step may be as follows.

In an early grounder, your feet are in front of you, and movement can occur the moment your feet touch the ground. At this time, the angle of the knee joint may vary depending on the inclination in the walking environment up the ramp. This may be because the minimum value of S1 occurs higher than the flat walking and the ramp down in order to walk up the slope.

The pentagram passes over the center of gravity, which was supported by the last period of the stand, to the lower limb, and the foot may be behind the body in preparation for the pentagram. At this time, the angle of the knee may vary depending on the inclination which is the environment after walking. The environment that shows a large difference depending on the walking environment at this stage may be a case of descending the ramp. In the case of descending the ramp, the degree of knee extension is small at S3, and the degree of bending at S4 may be increased. This may be because in a ramp downhill walking environment, the foot is higher when the foot is behind.

Therefore, as there is a characteristic according to the inclination of the walking environment, the walking environment may be estimated using the maximum and minimum points generated at each step based on the estimated walking step.

The change of poles at the initial grounding and the entire stator according to the slope described above can be used. The differences in pole changes in each walking environment are summarized as follows. In the case of ascending upward, the minimum point is higher than the knee joint angle between the flat and the downward slope at the initial folding period, and in the case of the downward slope, the maximum value is higher than the angle of the knee joint between the flat and the upward slope at all stages.

과 전유각기가 시작될 때 무릎 각

의 차이를 이용하며 수학식 8과 같이 표현될 수 있다.Therefore, the condition for detecting the walking environment is the knee angle when the initial folding is started.

Knee angle when and full breastplate starts

Using the difference of can be expressed as Equation 8.

값이 양의 방향으로 커지는 경향을 보이며, 내리막의 경우

값이 음의 방향으로 커지는 특징이 있다. 이러한 특징을 바탕으로 표 3와 같은 검출 조건이 결정될 수 있다. 여기서,

는 개인별 차이가 있을 수 있기 때문에 평지보행을 통하여 도출된 값이 적용될 수 있다. 일반적으로 평지 보행에서 첫 번째 신전 시작점은 평균 약 2~5도 정도 이내이며, 두 번째 신전 시작점은 평균 3~7도일 수 있다. 따라서 두 지점의 차이가 작게는 1도 크게는 5도 정도 차이를 보이므로, 일반적으로

=5도로 정의될 수 있다. 이 때, 건측의 운동정보를 추정하는 관성 센서의 측정 오차 및 부하반응기에서 개인별 특성을 더 고려하게 되면,

의 값은 7~10도 정도로 설정될 수 있으며, 이를 임계값으로 설정될 수 있다.The reason why Equation 8 is simple is derived from the walking characteristics in the ascending and descending flat and ramps.

Value tends to increase in the positive direction,

The characteristic is that the value increases in the negative direction. Based on these characteristics, detection conditions as shown in Table 3 may be determined. here,

The value derived through the flat walking can be applied because there may be individual differences. In general, the first temple starting point is about 2-5 degrees on average, and the second temple starting point may be 3-7 degrees on average. Therefore, the difference between the two points is as small as 1 degree and as large as 5 degrees.

Can be defined as = 5 degrees. At this time, when considering the measurement error of the inertial sensor for estimating the motion information of the healthy side and the individual characteristics in the load reactor,

The value of may be set to about 7 ~ 10 degrees, it may be set as a threshold.

Table 3 shows the walking environment detection conditions.

However, since the walking environment estimation may be different for each pedestrian, tuning of the threshold may be required. Therefore, in addition to the relevant methods, a condition-based pedestrian environment estimation method can be proposed. The additional method may be a linguistic classification that is not mathematically established as a discrimination method according to the walking pattern of the previous step and the next step based on the pole point. Table 4 shows detection conditions for each of these walking environments.

What is common in the detection conditions may be that S2 is larger than S1 in the walking steps of the flat, the ramp up and the ramp down. And the uphill and downhill environment can be distinguished based on the size of S1 and S3. In addition, by comparing the size of S1 with S4, the detection accuracy of each walking environment can be ensured. Based on these conditions, the rules shown in Table 5 can be derived.

Such a walking environment determination rule may be derived based on the walking cycle characteristics. This may be because walking is periodic and similar patterns are reproduced. In Table 4, the reason why the value of 20% of S4 is applied to the detection condition is to increase recognition rate by supplementing the feature that the angle difference is derived from the slight difference as shown in FIGS. . In addition, when the slope is uphill and downhill based on S1, the slope uphill may be a feature in which S1 has a higher angle than S3, and in the downhill slope, S1 has a lower angle based on S3. The reason why the percentage is applied to the value of S4 in the detection condition is that as shown in FIGS. 3 to 5, S4 shows a similar angle with no significant change in the flat and the slope uphill, while in the downhill slope, the angle value of S4 increases as the slope increases. It may be due to the point. Therefore, since the angle of S4 is not always a fixed angle, it is set as a section corresponding to 20% of S4, and 20, which is a percentage of percentage, satisfies all of the rule conditions without errors as shown in FIGS. 3 to 5. It may be due to the point. Therefore, the value may change according to the user in the future. If you do not use the value of S4 and limit the range to belonging values such as fuzzy, there may be a problem that you need to change the belonging value range as the inclination increases. This is because the detection condition value can be flexibly changed based on the value of.

10 is a diagram illustrating a walking state estimation apparatus, according to an exemplary embodiment.

Referring to FIG. 10, the walking state estimating apparatus 1000 according to an embodiment includes a memory 1010 and a processor 1020. The memory 1010 and the processor 1020 may communicate with each other via a bus 1030.

The memory 1010 may include instructions that can be read by a computer. The processor 1020 may perform the aforementioned operations as an instruction stored in the memory 1010 is executed in the processor 1020. The memory 1010 may be a volatile memory or a nonvolatile memory.

The processor 1020 may be a device that executes instructions or programs or controls the license plate recognition apparatus 1200. The processor 1020 may process the above-described operations, and thus a detailed description thereof will be omitted.

The embodiments described above may be implemented as hardware components, software components, and / or combinations of hardware components and software components. For example, the devices, methods, and components described in the embodiments may include, for example, processors, controllers, arithmetic logic units (ALUs), digital signal processors, microcomputers, field programmable gates (FPGAs). It may be implemented using one or more general purpose or special purpose computers, such as an array, a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications running on the operating system. The processing device may also access, store, manipulate, process, and generate data in response to the execution of the software. For convenience of explanation, one processing device may be described as being used, but one of ordinary skill in the art will appreciate that the processing device includes a plurality of processing elements and / or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as parallel processors.

The software may include a computer program, code, instructions, or a combination of one or more of the above, and may configure the processing device to operate as desired, or process independently or collectively. You can command the device. Software and / or data may be any type of machine, component, physical device, virtual equipment, computer storage medium or device in order to be interpreted by or to provide instructions or data to the processing device. Or may be embodied permanently or temporarily in a signal wave to be transmitted. The software may be distributed over networked computer systems so that they may be stored or executed in a distributed manner. The software and data may be stored on one or more computer readable recording media.

The method according to the embodiment may be embodied in the form of program instructions that can be executed by various computer means and recorded in a computer readable medium. The computer readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the media may be those specially designed and constructed for the purposes of the embodiments, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical media such as CD-ROMs, DVDs, and magnetic disks, such as floppy disks. Magneto-optical media, and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

Although the embodiments have been described with reference to the accompanying drawings, those skilled in the art may apply various technical modifications and variations based on the above. For example, the described techniques may be performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. may be combined or combined in a different form than the described method, or other components. Or, even if replaced or substituted by equivalents, an appropriate result can be achieved.

Claims (5)

Obtaining an angle value of the knee joint of the user through a sensor attached to the foot or leg of the user;
Determining a pole change pattern for the knee joint based on the angle value of the knee joint; And
Estimating the walking step of the user using the pole change pattern and the state machine
Including,
Estimating the walking step of the user
The presence of four poles in one walk;
A characteristic in which the pole change pattern has periodicity; And
A characteristic in which the current step of the walking step is associated with at least one of a previous step and a subsequent step
A walk state estimation method for estimating a walking step of the user based on at least one of the characteristics. delete The method of claim 1,
The gait step is divided based on the maximum bending and the maximum extension included in the pole change pattern, walking state estimation method. The method of claim 1,
Estimating a walking environment of the user based on the walking step
Further comprising a walking state estimation method. The method of claim 4, wherein
Estimating the walking environment
Based on the characteristics of the pole generated in the estimated walking step, the walking condition estimation method for estimating the walking environment to any one of flat, ramp up and ramp down.

Images