KR101599142B1 - Method and Apparatus for Continuous Control of Electric Wheelchair using EMG - Google Patents

Abstract

An electric wheelchair control method and apparatus using EMG are presented. An electric wheelchair control method using the proposed EMG includes receiving a plurality of EMG signals using an EMG sensor attached to a wheelchair user, extracting a user's intention from the received plurality of EMG signals, And controlling the direction and the velocity of the wheelchair by using the generated control signal. Wherein the step of extracting a user's intention from the received plurality of EMG signals and performing a signal processing and a correction for using the plurality of EMG signals in combination to generate a control signal includes generating an RMS signal Acquiring an envelope using the EMG signal, normalizing the RMS signal to match the levels of the inputted plurality of EMG signals, performing SPW signal processing and correction to use the plurality of EMG signals in combination, And generating a control signal for continuously controlling the wheelchair according to the intention of the user using the RMS signal and the SPW signal processing

Description

TECHNICAL FIELD [0001] The present invention relates to a method and apparatus for continuous control of an electric wheelchair using EMG,

The present invention relates to a method and an apparatus for controlling an electric wheelchair of a patient suffering from limb paralysis or a patient who can not use the hand

Electric wheelchairs are being manipulated using joysticks, keyboards, voice recognition interfaces, and interfaces using mouth or tongue. Some of the patients who require an electric wheelchair can not operate the wheelchair in the conventional way due to limb paralysis, lack of strength, and the like. With the development of sensing technology, interfaces for manipulating medical aids by using muscle signals and brain signals have been developed and studied.

However, such a conventional interface has discontinuously grasped the intention of the user by analyzing the pattern of the signals rather than the magnitude of the self-motional movement of the muscle signals. To use this method as an interface in various environments or complex conditions, the interface becomes complicated because it requires various patterns. Also, the same input is repeated several times.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and an apparatus for continuously controlling an electric wheelchair by an arbitrary movement of a facial muscle by a limb paralysis patient or a patient who can not use the hand. Accordingly, it is an object of the present invention to provide an electric wheelchair continuous control method and apparatus using movable EMG in various residential environments.

In one aspect, an electric wheelchair control method using electromyography (EMG) proposed in the present invention includes receiving a plurality of EMG signals using an EMG sensor attached to a wheelchair user, And generating a control signal by performing signal processing and correction for using the plurality of electromyogram signals in combination, and controlling the direction and speed of the wheelchair to be continuously controlled using the generated control signal .

Wherein the step of extracting a user's intention from the received plurality of EMG signals and performing a signal processing and correction for using the plurality of EMG signals in combination to generate a control signal comprises: Obtaining an envelope using a root mean squaring (RMSK) signal, normalizing the RMS signal to match the levels of the received plurality of EMG signals, and using the plurality of EMG signals in combination Performing a Sum of Product Window (SPW) signal processing and correction for the wheelchair, and generating a control signal for continuously controlling the wheelchair according to the intention of the user using the RMS signal and the SPW signal processing .

The step of receiving a plurality of EMG signals using the EMG sensor attached to the wheelchair user may receive the user's voluntary exercise size as the plurality of EMG signals to extract the intention of the user.

Performing the SPW signal processing and correction to use the plurality of electromyogram signals in combination may remove an unwanted portion of the control signal by using a Zero Searching Window (ZSW).

The RMS signal and the SPW signal processing may be applied to the control signal by superimposing the plurality of EMG signals when the plurality of EMG signals occur at the same time.

By using the magnitude intervals of the plurality of superimposed EMG signals separately, they can be used as various inputs.

The step of generating a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal may perform filtering to remove noise of the control signal.

According to an aspect of the present invention, there is provided an apparatus for controlling an electric wheelchair using an EMG, the apparatus comprising: an input unit for receiving a plurality of EMG signals using an EMG sensor attached to a wheelchair user; An EGM signal processing unit for extracting a plurality of EMG signals, performing signal processing and correction for using the plurality of EMG signals in combination, and generating a control signal, and a control unit for continuously controlling the direction and speed of the wheelchair using the generated control signal .

The EGM signal processing unit may include an RMS signal normalization unit for obtaining an envelope using the RMS signals of the received plurality of EMG signals and for normalizing the RMS signals so as to match the levels of the received plurality of EMG signals, A SPW signal processing unit for performing SPW signal processing and correction to use the RMS signal and the SPW signal processing, a control signal generating unit for generating a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal processing Section.

The input unit may receive the user's voluntary exercise size as the plurality of EMG signals to extract the intention of the user.

The SPW signal processing unit can remove an undesired portion of the control signal by using the ZSW.

The RMS signal normalization unit and the SPW signal processing unit may superimpose the plurality of EMG signals to the control signal when the plurality of EMG signals occur at the same time.

The control signal generator may include a filter for removing noise of the control signal.

According to the embodiments of the present invention, the complexity of the system is reduced by proposing a control interface reflecting the size of the signal itself rather than a pattern generated by the signal, and the size of the motion of the muscles is adjusted without inputting the input signal repeatedly. Can be controlled.

FIG. 1 is a flowchart for explaining an electric wheelchair continuous control method using an EMG according to an embodiment of the present invention.
2 is a flowchart for explaining a step of extracting a user's intention from a plurality of input EMG signals according to an embodiment of the present invention and performing a signal processing and a correction to generate a control signal.
FIG. 3 is a diagram for explaining an electric wheelchair continuous control process using an EMG according to an embodiment of the present invention.
4 is a diagram illustrating an RMS signal and a normalized RMS signal according to an embodiment of the present invention.
5 is a diagram illustrating acceleration and deceleration profiles from an EMG signal according to an embodiment of the present invention.
6 is a view showing a configuration of an electric wheel chair continuous control apparatus using an EMG according to an embodiment of the present invention.
FIG. 7 is a diagram for explaining an experimental procedure using an electric wheelchair continuous control apparatus using an EMG according to an embodiment of the present invention.
8 is a diagram for explaining an EMG signal processing process for extracting a user's intention determination vector during an experiment process according to an embodiment of the present invention.
FIG. 9 is a graph showing required angular velocities of right and left wheels according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a flowchart for explaining an electric wheelchair continuous control method using an EMG according to an embodiment of the present invention.

An electric wheelchair continuous control method using an EMG includes receiving (110) a plurality of EMG signals using an EMG sensor attached to a wheelchair user, extracting a user's intention from the received plurality of EMG signals, (Step 120) of performing a signal processing and correction for using the electromyogram signal in combination to generate a control signal, and continuously controlling the direction and speed of the wheelchair using the generated control signal .

In step 110, a plurality of EMG signals may be received using the EMG sensor attached to the wheelchair user. For example, an electromyogram signal can be input by attaching an electrode to both muscles of a user's face. The EMG signal received from both muscles of the user's face may include the intention of the user. In other words, the user can arbitrarily input the magnitude of motion of the user into the plurality of electromyogram signals to extract the user's intention. It is possible to measure the self-contraction force of the muscles to enable continuous control of the electric wheelchair.

In step 120, a control signal may be generated by extracting a user's intention from the input plurality of electromyogram signals, and performing signal processing and correction for using the plurality of electromyogram signals in combination. The generating of the control signal may include acquiring an envelope using the RMS signals of the received plurality of EMG signals, normalizing the RMS signal to match the levels of the received plurality of EMG signals, The method comprising the steps of: performing a SPW signal processing and correction to use a signal in combination, and generating a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal processing . Will be described in more detail with reference to FIG.

2 is a flowchart for explaining a step of extracting a user's intention from a plurality of input EMG signals according to an embodiment of the present invention and performing a signal processing and a correction to generate a control signal. The step of extracting the intention of the user from the received plurality of EMG signals and performing the signal processing and the correction to generate the control signal includes a step 210 of acquiring an envelope using the RMS signals of the inputted plurality of EMG signals, (220) normalizing the RMS signal to match the levels of the received plurality of EMG signals, performing (230) SPW signal processing and correction to use the plurality of EMG signals in combination, And generating (240) a control signal for continuously controlling the wheelchair according to the intention of the user using the SPW signal processing.

In step 210, an envelope may be obtained using the RMS signals of the plurality of EMG signals.

In step 220, the RMS signal may be normalized to match the levels of the input EMG signals. Normalization of the RMS signal can be performed since the levels of the muscle signals of both sides of the measured user face are different.

In step 230, SPW signal processing and correction may be performed to use the plurality of EMG signals in combination. At this time, the SPW signal occurs when two signals are superimposed on each other, and the magnitude of the superimposed signal can also be obtained. This size can be reflected in the linear speed increase of the electric wheelchair. And if it increases above a certain value, the linear velocity of the wheelchair can be decelerated. In addition, a Zero-Searching Window (ZSW) can be used to eliminate the noise value generated when the signal size interval is discriminated.

In step 240, the RMS signal and the SPW signal processing may be used to generate a control signal for continuously controlling the wheelchair according to the intention of the user. In other words, the RMS signal and the SPW signal processing can be applied to the control signal by overlapping the plurality of EMG signals when the plurality of EMG signals occur at the same time. By using the magnitude sections of the plurality of superimposed electromyogram signals separately, they can be used as various inputs. In this way, the linear velocity and the rotational velocity of the electric wheelchair can be generated by using the electromyogram signal and the SPW signal processed signal obtained from both muscles of the RMS-processed user's face, thereby controlling the electric wheelchair continuously.

Referring again to FIG. 1, in step 130, the direction and speed of the wheelchair may be continuously controlled using the generated control signal. At this time, filtering may be performed to remove noise of the control signal.

In other words, in order to solve the problems of the prior art, the present invention is applied to the control of a method of proposing an autonomous motion magnitude of an EMG signal obtained from both muscles of a user's face. At this time, when the signals occur at the same time, the signals are superimposed and this size is applied to the control signal. In addition, by using the size interval of the superimposed signal differently, various inputs are made possible. Specifically, the rotational speed control of the electric wheelchair was generated by the magnitude of the EMG signal measured on the right or left face. When facial muscle signals occur at the same time, they are reflected in the linear velocity increase of the wheelchair when the signal is below a certain value. On the other hand, if the signal of the facial muscles is generated at the same time, it is reflected in the decrease of the linear velocity of the wheelchair when the signal is above a certain value. Also, the left rotation speed is determined according to the size of the left facial muscle signal, and the right rotation speed is determined by the size of the right facial muscle signal. In conclusion, it is possible to reduce the complexity of the system by generating four control signals (acceleration, deceleration, right turn, left turn) with only two facial muscle signals. 3, the process of continuously controlling the electric wheelchair using the EMG according to an embodiment of the present invention will be described in detail.

FIG. 3 is a diagram for explaining an electric wheelchair continuous control process using an EMG according to an embodiment of the present invention.

First, an electrode is attached (311) to a user 350 using an electric wheelchair to receive an EMG signal including an intention of the user through the EMG sensor 312. In other words, the EMG signal can be measured from the right and left two muscle signals (Right of Zygomaticus Major Muscle) and LZM (Left of Zygomaticus Major Muscle) used in the facial expression of the user. The magnitude of the EMG signal measured from such RZM and LZM may increase or decrease due to spontaneous movement of the user's facial muscles. With the RZM and LZM, the independent speed and direction of the two wheels of the electric wheelchair is adjustable. The EMG signal processing 320 for controlling the electric wheelchair can be performed using the received EMG signal.

The EMG signal processing 320 may first generate an RMS signal (step 321). The measured EMG signal (Raw EMG signal) is a set of SMUAP (Single Motor Unit Action Potential) generated from each neuron-dominated muscle fiber by motor neurons, and thus includes the intention of the user. However, since the measured EMG signal (Raw EMG signal) has a large amount of noise, we must extract the average value using the root mean squares (RMS) signal processing. At this time, since the EMG signal is measured from the human skin, several points to be considered are as follows. The EMG signal can vary depending on the physical conditions of the person (e.g., sweat and fatigue), the location of the sensor attachment, and environmental conditions such as ambient temperature. This will be described in more detail with reference to FIG.

4 is a diagram illustrating an RMS signal and a normalized RMS signal according to an embodiment of the present invention.

As described above, the measurement value of the EMG signal may vary depending on the environmental conditions. For this reason, as shown in FIG. 4A, the EMG signal 410a of the RZM and the EMG signal 420a of the LZM may have different magnitude levels during muscle contraction. Therefore, a normalization process is required to match the magnitude level. The normalization process can use Equation (1).

수학식(1)

Equation (1)

Here, x _{rms, i} (t) is an RMS signal with a window size of 300 samples, where one sample represents a sampling time T = 0.001 [s]. n _i (t) represents the normalized signal. L _{rms, i} and U _{rms, i} represent the lower and upper bounds of RZM for i = r and RZM for LZM when i = 1.

In fact, there are two representative standardization methods for biomedical signal processing such as Maximum Voluntary Isomeric Contraction (MVIC) and Reference Voluntary Contraction (RVC). MVIC can be used when the muscle is at maximum contractile, and RVC can be used when the muscle is moderately contracted to improve the biomedical signal agility. In other words, RVC can be used for normal muscle contraction. RVC is preferred over MVIC for wheelchair control because maximal muscle contraction can make muscles tired easily. The upper limit value of the RVC can be selected in half of the MVIC using Equation (2).

수학식(2)

Equation (2)

Here, MVICi represents the required RMS signal when the muscle is maximally contracted for i = r, l (right, left). On the other hand, the lower limit value in the stable condition of the muscles can be calculated as Equation (3).

수학식(3)

Equation (3)

Where M _{L, i} represents the window size, where M _{L, i} = 10,000 means 10 [s] for i = r, l (right, left). For example, as shown in FIG. 4 (a), the upper and lower limits of the RMS signal of the LZM represent a scaled voltage unit obtained from a 12-bit analog-to-digital converter at [V · sc], and sc represents a scale constant have. The upper limit value lower limit values shown in the formulas (2) and (3) can all be determined in advance through the preprocessing process.

Normalization (322) of the RMS may be performed using this RMS signal. Normalization of the RMS is performed using the RMS signal shown in FIG. 4 (a), and the normalized RZM signal 410b and the normalized LZM signal 420b are signals having similar sizes as shown in FIG. 4 (b) Can be obtained.

In the next step, a sum of product window (SPW) signal processing 323 may be performed using equation (4).

수학식(4)

Equation (4)

Where M _{? Represents} the corresponding window size, where M _? = 50, nl and nr denote the normalized LZM signal and normalized RZM signals, respectively.

The SPW signals can then be distinguished (324). The SPW operation can be used to find out whether RZM and LZM are simultaneously deflated or not. SPW < RTI ID = 0.0 > zeta < / RTI > (t) is not activated nearly as long as either RZM or LZM is constricted, while conversely, RZM and LZM are largely activated when both are contracted at the same time. Therefore, SPW operation can be used to indicate the simultaneous operation of two muscles. This can be used to generate the linear acceleration and deceleration of the required wheelchair. In order to reduce the burden of the calculation amount, the equation (4) can be transformed into the equation (5).

수학식(5)

Equation (5)

In equation (5), only two multiplications and additions are required. The command to accelerate and decelerate the electric wheelchair can be generated after applying the upper limit value and the lower limit value for the SPW signal as suggested in FIG. This will be described in more detail with reference to FIG.

5 is a diagram illustrating acceleration and deceleration profiles from an EMG signal according to an embodiment of the present invention. The acceleration and deceleration command profile can be expressed using Equations (6) and (7).

수학식(6)

Equation (6)

수학식(7)

Equation (7)

Where a (t) and d (t) denote the linear acceleration and deceleration of the wheelchair, respectively, and U _T and L _T denote the upper and lower limits, respectively. In FIG. 5A, the SPW signal 510a between the lower limit value and the upper limit value can be utilized to generate the acceleration signal 510b as shown in FIG. 5 (b). 5 (a), the SPW signal higher than the upper limit value may be utilized to generate the deceleration signal 510c as shown in FIG. 5 (c). However, the acceleration signal 510b may include a cut part due to the threshold as shown in FIG. 5 (b). These parts must be removed in order not to affect the acceleration. In other words, the acceleration signal correction 325 can be performed using the upper limit value and the lower limit value for the SPW signal. To do so, a zero searching window (ZSW) can be used as in equation (8).

수학식(8)

Equation (8)

와 같이 나타낼 수 있다. M_a는 해당 윈도우 사이즈를 나타내고, 여기에서 M_a=150이다. ZSW를 사용함으로써, 도 5(c)에 보여진 것과 같이 가속의 원하지 않는 부분을 제거할 수 있다. 위 과정을 통해, EMG 신호로부터 요구되는 가속 및 감속 명령 프로파일 모두를 찾을 수 있다. 그리고, 더 활성화된 SPW 신호들이 전동 휠체어의 감속을 위해 사용될 수 있다.Where a _c (t) represents the corrected acceleration, ζ (t) represents the shape of the ZSW,

As shown in Fig. M _a represents the corresponding window size, where M _a = 150. By using ZSW, unwanted portions of the acceleration can be removed as shown in Fig. 5 (c). Through the above procedure, both the acceleration and deceleration command profiles required from the EMG signal can be found. Further activated SPW signals can be used for deceleration of the electric wheelchair.

으로 나타낸 결정 벡터는 수학식(1)의 표준화된 RZM RMS 신호 및 표준화된 LZM RMS 신호를 수집함으로써 얻어질 수 있고, 수학식(8)의 수정된 가속 신호 및 수학식(7)의 감속 신호는 수학식(9)과 같이 나타낼 수 있다.
In the next step, a user's intention such as a right turn, a left turn, an acceleration, and a deceleration may be extracted (331). At this time, a user's intention determination matrix and a decision vector may be used.

Can be obtained by collecting the normalized RZM RMS signal and the normalized LZM RMS signal of Equation (1), and the modified acceleration signal of Equation (8) and the deceleration signal of Equation (7) (9). &Quot; (9) "

수학식(9)

Equation (9)

Wherein the first and second elements may represent activations of the RZM and LZM respectively and the third and fourth elements may respectively represent an acceleration signal and a deceleration signal according to simultaneous activation levels have. Then, the user's intention determination matrix denoted by H can be expressed by Equation (10).

수학식(10)

Equation (10)

은 우회전 의사를 추출하기 위해 선택된 기저 벡터(basis vector)이고, 이것은 오직 RZM 활성화만을 반영한다.

은 좌회전 의사를 추출하기 위해 선택된 기저 벡터이고, 오직 LZM 활성화만을 반영한다. From here,

Is the basis vector chosen to extract the right-handed physician, which reflects only RZM activation.

Is the base vector selected to extract the left-turn physician, and reflects only LZM activation.

은 휠체어의 선형 가속 의사를 추출하기 위해 선택된 기저 벡터이고, 그러므로 이것은 RZM 및 LZM 활성화뿐만 아니라, 가속도를 반영할 수 있다.

은 휠체어의 선형 감속의사를 추출하기 위해 선택된 기저 벡터이고, 이것은 RZM 및 LZM 활성화뿐만 아니라, 감속도 반영할 수 있다.

Is the base vector selected to extract the linear acceleration intent of the wheelchair, and thus can reflect acceleration as well as RZM and LZM activation.

Is a base vector selected to extract a linear deceleration impression of a wheelchair, which can reflect RZM and LZM activation as well as deceleration.

는 수학식(11)과 같이 예상되는 벡터

를 얻기 위해 결정 행렬 H의 기저벡터로 투사하여 얻어질 수 있다.
The decision vector measured in real time

(11): < EMI ID =

Lt; RTI ID = 0.0 > H < / RTI >

수학식(11)

Equation (11)

은

로 나타낼 수 있는 모션 벡터이다. 모션 벡터의 각각의 요소는 전동 휠체어의 요구되는 모션 발생을 위해 이동 의사를 반영할 수 있다. 따라서, 상기 예상된 벡터는 종종 생체 의학 간섭으로부터 영향을 받을 수 있고, 예상되는 벡터의 고주파 요소를 제거하기 위해 LPF(Low Pass Filter, 저주차통과필터)가 적용될 수 있다. 그러면 전동 휠체어 장치를 위해 요구되는 선형속도 및 회전속도의 연속적 생성을 위해 LPF를 갖는 모션 벡터가 사용될 수 있다.From here

silver

Which is a motion vector. Each element of the motion vector may reflect movement intention for the required motion generation of the electric wheelchair. Thus, the predicted vector can often be affected by biomedical interference, and a low pass filter (LPF) can be applied to remove the high frequency components of the expected vector. The motion vector with the LPF can then be used for the continuous generation of the linear velocity and the rotational velocity required for an electric wheelchair apparatus.

의 요소들의 선형 조합을 사용함으로써 생성될 수 있다. 예를 들어, 수학식(12) 및 수학식(13)과 같이 선형 속도는 가속과 감속 사이의 차이를 적분함으로써 결정될 수 있고, 회전 속도는 RZM 및 LZM 활성화 사이의 차이로부터 직접적으로 결정될 수 있다.
The linear velocity and rotational speed required for driving an electric wheelchair are determined by the motion vector

Lt; RTI ID = 0.0 > of < / RTI > For example, linear velocities such as (12) and (13) can be determined by integrating the difference between acceleration and deceleration, and the rotation rate can be determined directly from the difference between RZM and LZM activation.

수학식(12)

Equation (12)

수학식(13)

Equation (13)

은 각각 요구되는 전동 휠체어의 선형속도 및 회전속도를 나타내고, v_d(t - T)은 이전 샘플에서 요구되는 속도를 나타낸다. T는 샘플링 타임을 나타내고, 여기에서 T=0.001[s]이고, K_a, K_d 및 K_q는 각각 가속, 감속 및 회전속도를 위한 이득(Gain, 게인)을 나타내고, K_a=5, K_d=10, K_q=5을 사용할 수 있다. 그리고, σ_i(t)와 같은 형태의 한계값 동작은 작은 이동 의사를 무시할 뿐만 아니라, 이동 의사의 잦은 변화를 방지하기 위해 적용될 수 있다.
Where v _d (t) and

Respectively represent the required linear speed and rotational speed of the electric wheelchair and v _d (t - T) represents the speed required in the previous sample. T represents the sampling time, where T = 0.001 [s], K _a , K _d and K _q represent gains for acceleration, deceleration and rotation speed, respectively, and K _a = 5, K _d = 10, and K _q = 5 can be used. And, the threshold operation like the form σ _i (t) can be applied not only to ignore small movements but also to prevent frequent changes of movement intention.

및

이 RZM 및 LZM의 활성화를 포함하기 때문에 선형속도 및 회전속도를 동시에 갖는 복잡한 모션이 가능할 수 있다.Here, the elements of the motion vector can affect the movement of the wheelchair only when it is greater than the average of the elements, otherwise it can be ignored. Also, for safety reasons, the gain of deceleration must be greater than the gain of acceleration. Basis vector of linear acceleration and deceleration

And

This involves the activation of RZM and LZM, so complex motions with both linear velocity and rotational velocity can be possible.

The ratio relation between the linear / rotational speed of the electric wheelchair and the speed of the two wheels according to the proposed method can be expressed as Equation (14).

수학식(14)

Equation (14)

Here, j _omega can represent Jacobian as follows.

는 각각 휠체어의 선형속도 및 회전속도를 나타낸다. ω_r 및 ω_l 은 각각 오른쪽 및 왼쪽 바퀴의 각속도를 나타낸다.R represents the radius of the wheel, where R = 170 mm. W represents the distance between the center of the right wheel and the center of the left wheel, where W = 695 [mm]. v and

Respectively represent the linear velocity and the rotational velocity of the wheelchair. ? _r and? _l represent angular velocities of the right and left wheels, respectively.

In the next step, the control unit 300 can control two wheels of the electric wheel chair (332). The required angular velocity of the two wheels can be obtained from Equations (12) and (13) by using Equation (15).

수학식(15)

Equation (15)

이고, ω_d,r 및 ω_d,l 은 각각 전동 휠체어의 오른쪽 및 왼쪽 바퀴의 각 속도를 나타낸다. 두 바퀴의 요구되는 각속도는 전동 휠체어(340)의 DC 모터 증폭기의 구동(341)을 위해 PWM(pulse width modulation) 레벨로 변환될 수 있다. 그리고, 기계적 부분(342)과 증분형 인코더(343)를 이용하여 전동 휠체어(340)를 제어할 수 있다. 이를 위해, 모션 컨트롤러는 수학식(16)과 같이 나타낼 수 있다.
From here,

Ω _{d, r} and ω _{d, l} denote angular velocities of the right and left wheels of the electric wheelchair, respectively. The required angular velocity of the two wheels can be converted to a pulse width modulation (PWM) level for driving 341 the DC motor amplifier of the electric wheelchair 340. The mechanical wheel part 342 and the incremental encoder 343 can be used to control the electric wheelchair 340. For this purpose, the motion controller can be expressed by the following equation (16).

수학식(16)

Equation (16)

의 대각선 요소를 갖는 PWM 변환 행렬을 나타낸다. 여기에서 i=r,l, (오른쪽 및 왼쪽의) K_v에 대하여 K_p,i=49.95이고, 대각선 이득 행렬은 단위 값보다 조금 더 큰 요소를 가질 수 있고, K_v,i=1.1이다. 그리고, ψ는 대각선 보상 행렬은 아래와 같이 바퀴에서 각각 요구되는 각속도가 아래에 해당하는 바퀴의 실제 각속도보다 작을 때만 활성화 된다. Where K _p is

Lt; RTI ID = 0.0 > of diagonal < / RTI > Where K _{p, i} = 49.95 for i = r, l, K _v (right and left), and the diagonal gain matrix can have an element slightly larger than the unit value and K _{v, i} = 1.1. And, ψ is activated only when the diagonal compensation matrix is smaller than the actual angular velocity of the corresponding wheel below, as shown below.

Here, the above elements can realize the immediate deceleration of the wheel.

6 is a view showing a configuration of an electric wheel chair continuous control apparatus using an EMG according to an embodiment of the present invention.

The electric wheelchair continuous control apparatus using EMG may include an input unit 610, an EGM signal processing unit 620, and a control unit 630.

The input unit 610 can receive a plurality of EMG signals using the EMG sensor attached to the wheelchair user. For example, an electromyogram signal can be input by attaching an electrode to both muscles of a user's face. The EMG signal received from both muscles of the user's face may include the intention of the user. In other words, the user can arbitrarily input the magnitude of motion of the user into the plurality of electromyogram signals to extract the user's intention. It is possible to measure the self-contraction force of the muscles to enable continuous control of the electric wheelchair.

The EMG signal processing unit 620 may extract a user's intention from the received plurality of EMG signals, and may generate a control signal by performing signal processing and correction for using the plurality of EMG signals in combination.

The EMG signal processing unit 620 may include an RMS signal normalization unit 621, an SPW signal processing unit 622, and a control signal generation unit 623.

The RMS signal normalization unit 621 may normalize the RMS signal to obtain an envelope using the received RMS signals of the plurality of EMG signals and adjust the level of the received plurality of EMG signals.

The SPW signal processing unit 622 can perform SPW signal processing and correction to use a plurality of electromyogram signals in combination. Further, the SPW signal processing unit 622 can remove unwanted portions of the control signal by using the ZSW.

The control signal generator 623 may generate a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal processing.

The RMS signal normalization unit 621 and the SPW signal processing unit 622 may apply the plurality of EMG signals to the control signal when the plurality of EMG signals occur at the same time.

The control unit 630 can continuously control the direction and speed of the wheel chair using the generated control signal. The control unit 630 may include a filter for removing noise of the control signal.

The apparatus for continuously controlling the electric wheelchair using the EMG according to the embodiment of the present invention can operate according to the method for continuously controlling the electric wheelchair using the EMG described with reference to FIGS.

FIG. 7 is a diagram for explaining an experimental procedure using an electric wheelchair continuous control apparatus using an EMG according to an embodiment of the present invention.

In order to show the effect of the electric wheelchair continuous control system using the proposed EMG, obstacle avoidance was performed in an experimental environment as shown in FIG. 7, obstacles 731, 732, 734, 735, and 736 at five points are installed as shown in FIG. Then, the user controls the electric wheelchair from the starting point 710 to the arrival point 720 along the routes 711 and 721 by using the electric wheelchair continuous control apparatus using the proposed EMG. Here, the solid line 711 indicates the forward route, and the dotted line 721 indicates the return route. These experiments can control the linear speed and the rotational speed of the electric wheelchair by activating the RZM and LZM, respectively, as in the case of adjusting the wheels of a conventional vehicle and pressing the accelerator pedal.

8 is a view for explaining an EMG signal processing process for extracting a decision vector during an experiment according to an embodiment of the present invention.

Before performing the experiment described above, the upper and lower limits of the RMS signal should be predetermined, since the EMG may vary depending on the environment and physical conditions. For example, through the preprocessing of the EMG signal, the upper and lower limits of the RMS can be expressed as U _{rms, r} = 40, U _{rms, l} = 38, L _{rms, r} = 9, and L _{rms , and l} = 10. In addition, the lower limit value and the upper limit value can be utilized in obstacle avoidance in real time. In the EMG signal processing and control method, first, the standardized signals can be obtained from the RMS signal as shown in FIG. 8 (a). Figure 8 (a) shows the normalized RMS signal 820a from the RMS signal 810a and the LZM (left facial muscle signal) from the RZM (right facial muscle signal) during obstacle avoidance.

Next, the SPW signal can be obtained by the equation (4) as shown in Fig. 8 (b). Then, the acceleration and deceleration signals are applied to the SPW signal by applying the threshold operation of Equations (6) and (7) to the SPW signal and by eliminating an unnecessary portion of the acceleration using ZSW via Equation (8) c). An acceleration command profile 810c and a deceleration command profile 820c can be obtained as shown in Fig. 8 (c).

Next, the decision vector of the equation (9) can be obtained by using the signals of Figs. 8 (a) and 8 (c), and the motion vector can be calculated from the equation (11). At this time, the required linear velocity and rotational velocity can be calculated from equations (12) and (13), respectively.

Next, the required angular velocity of the two wheels can be obtained from equations (15) and (16). The results are shown in Fig.

9 is a graph showing the required angular velocity 910 of the right wheel and the required angular velocity 920 of the left wheel according to an embodiment of the present invention. The required angular velocity of the two wheels can be obtained from equations (15) and (16). In other words, FIG. 9 is a diagram showing how the required angular velocity of the left and right wheels is generated from the motion vector of equation (11). Experimental results are described at six points from T1 to T6 shown in Fig. 8 (a) and Fig.

The angular velocity of the right wheel may increase when RZM is activated at the time point of T1 shown in Figs. 8 (a) and 9. The angular velocity of the left wheel can increase when LZM is activated at the time of T2. The electric wheelchair can be stopped at the time of T3 when the SPW is activated larger than UT. RZM can be largely activated at time points T4 and T5, respectively, and the angular velocity of the right wheel can be greatly increased to avoid obstacles. As a result, the RZM can be largely activated for a long time (for about 5 seconds) at the time point T6 shown in FIG. 8 (a). From the experimental results, it can be seen that various continuous control commands such as left turn, right turn, straight run and stop of the electric wheelchair are generated. Furthermore, it can be confirmed that the control signal generated by the method shown in FIG. 9 is always continuous and flexible.

The apparatus described above may be implemented as a hardware component, a software component, and / or a combination of hardware components and software components. For example, the apparatus and components described in the embodiments may be implemented within a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), a PLU a programmable logic unit, 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 execution of the software. For ease of understanding, the processing apparatus may be described as being used singly, but those skilled in the art will recognize that the processing apparatus may have a plurality of processing elements and / As shown in FIG. For example, the processing unit may comprise a plurality of processors or one processor and one controller. Other processing configurations are also possible, such as a parallel processor.

The software may include a computer program, code, instructions, or a combination of one or more of the foregoing, and may be configured to configure the processing device to operate as desired or to process it collectively or collectively Device can be commanded. The software and / or data may be in the form of any type of machine, component, physical device, virtual equipment, computer storage media, or device , Or may be permanently or temporarily embodied in a transmitted signal wave. The software may be distributed over a networked computer system and 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 an embodiment may be implemented in the form of a program command that can be executed through 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 to be recorded on the medium may be those specially designed and configured for the embodiments or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media 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 machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. For example, it is to be understood that the techniques described may be performed in a different order than the described methods, and / or that components of the described systems, structures, devices, circuits, Lt; / RTI > or equivalents, even if it is replaced or replaced.

Therefore, other implementations, other embodiments, and equivalents to the claims are also within the scope of the following claims.

Claims (13)

In an electric wheelchair control method using an EMG,
Receiving a plurality of EMG signals using an EMG sensor attached to the wheelchair user;
Extracting a user's intention from the input plurality of EMG signals, and performing a signal processing and a correction for using the plurality of EMG signals in combination to generate a control signal; And
Continuously controlling the direction and speed of the wheel chair using the generated control signal
Lt; / RTI >
Wherein the step of extracting a user's intention from the received plurality of EMG signals and performing a signal processing and a correction for using the plurality of EMG signals in combination to generate a control signal,
Performing SPW signal processing and correction to use the plurality of electromyogram signals in combination
Wherein the wheelchair is controlled by an electric motor. The method according to claim 1,
Wherein the step of extracting a user's intention from the received plurality of EMG signals and performing a signal processing and a correction for using the plurality of EMG signals in combination to generate a control signal,
Acquiring an envelope using the received RMS signals of the plurality of EMG signals;
Normalizing the RMS signal to match levels of the plurality of input EMG signals; And
Generating a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal processing
Further comprising: an electromagnetically coupled wheelchair (EMG). The method according to claim 1,
Wherein the step of receiving a plurality of EMG signals using the EMG attached to the wheelchair user comprises:
Wherein the user's intentional motion magnitude is input to the plurality of electromyogram signals to extract the intention of the user
An electric wheelchair control method using EMG. 3. The method of claim 2,
Performing SPW signal processing and correction to use the plurality of electromyogram signals in combination,
The use of ZSW eliminates unwanted parts of the control signal
An electric wheelchair control method using EMG. 3. The method of claim 2,
Wherein the RMS signal and the SPW signal processing are performed by superimposing the plurality of EMG signals on the control signal when the plurality of EMG signals are simultaneously generated
An electric wheelchair control method using EMG. 6. The method of claim 5,
By using the magnitude intervals of the plurality of superimposed EMG signals separately, it can be used as various inputs
An electric wheelchair control method using EMG. 3. The method of claim 2,
Wherein the step of generating a control signal for continuously controlling the wheel chair according to the intention of the user using the RMS signal and the SPW signal comprises:
And performs filtering to remove noise of the control signal
An electric wheelchair control method using EMG. An electric wheelchair control apparatus using an EMG,
An input unit for receiving a plurality of EMG signals using an EMG sensor attached to the wheelchair user;
An EGM signal processing unit for extracting a user's intention from the received plurality of EMG signals and performing signal processing and correction for using the plurality of EMG signals in combination to generate a control signal; And
A controller for continuously controlling the direction and speed of the wheel chair using the generated control signal,
Lt; / RTI >
The EGM signal processing unit,
A SPW signal processing unit for performing SPW signal processing and correction to use the plurality of electromyogram signals in combination,
And an electric wheelchair control device using the EMG. 9. The method of claim 8,
The EGM signal processing unit,
An RMS signal normalization unit that obtains an envelope using the RMS signals of the received plurality of EMG signals and normalizes the RMS signals to match levels of the received plurality of EMG signals; And
A control signal generator for generating a control signal for continuously controlling the wheelchair according to the user's intention using the RMS signal and the SPW signal processing,
And an electric wheelchair control device for controlling the wheelchair. 9. The method of claim 8,
Wherein the input unit comprises:
Wherein the user's intentional motion magnitude is input to the plurality of electromyogram signals to extract the intention of the user
An electric wheelchair control system using EMG. 10. The method of claim 9,
The SPW signal processing unit,
The use of ZSW eliminates unwanted parts of the control signal
An electric wheelchair control system using EMG. 10. The method of claim 9,
Wherein the RMS signal normalization unit and the SPW signal processing unit superimpose the plurality of EMG signals and apply the plurality of EMG signals to the control signal
An electric wheelchair control system using EMG. 10. The method of claim 9,
Wherein the control signal generator comprises:
And a filter for removing noise of the control signal
An electric wheelchair control system using EMG.

Images