阅读说明：本技术 骨骼机器人负载模型建模建立方法 (Skeletal robot load model modeling and establishing method ) 是由 杨晓冬 张向刚 于 2019-08-27 设计创作，主要内容包括：本发明涉及一种外骨骼机器人负载模型建模方法,电枢绕组的电感为L<Sub>a</Sub>,电枢绕组的电阻为R<Sub>a</Sub>,电枢电压为u<Sub>a</Sub>,流过电枢的电流为i<Sub>a</Sub>,电机轴的角速度为Ω,电机轴转角为θ；对电枢回路,根据基尔霍夫电压定律进行计算：当以电机轴转角为θ作为输出量时,带入关系<Image he="140" wi="233" file="DDA0002181384110000011.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>可以得到以转过角度作为输出时的微分方程为：<Image he="73" wi="700" file="DDA0002181384110000012.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>本方案提供的负载模型建,更加贴合人体实际运用。(The invention relates to a modeling method of an exoskeleton robot load model, wherein the inductance of an armature winding is L a The resistance of the armature winding is R a Armature voltage of u a The current flowing through the armature is i a The angular speed of the motor shaft is omega, and the rotational angle of the motor shaft is theta; and (3) calculating an armature loop according to kirchhoff voltage law: when the motor shaft rotation angle is theta as the output quantity, the relationship is brought The differential equation for a turn angle as output can be found as: the load model that this scheme provided is built, human actual application of laminating more.)

1. An exoskeleton robot load model modeling method is characterized by comprising the following steps:

s100: setting the inductance of the armature winding to be L_aThe resistance of the armature winding is R_aArmature voltage of u_aThe current flowing through the armature is i_aThe angular speed of the motor shaft is omega, and the rotational angle of the motor shaft is theta;

s200: calculating parameters;

for the armature loop, the following can be obtained according to kirchhoff's voltage law:

in the formula

the motor has an electromagnetic torque equation of

M＝K_mi_a

Wherein K_mThe motor torque coefficient is, and M is the armature current to generate the electromagnetic torque;

equation of torque balance on motor shaft

Wherein:

j is equivalent moment of inertia converted to the shaft of the motor, and if the moment of inertia of the shaft of the motor is J_MThe moment of inertia converted from load to motor shaft is J_MLWhen J is equal to J_M+J_ML；

B is the viscous friction coefficient converted to the motor shaft, and if the rotating shaft of the motor rotates, the viscous friction coefficient is B_MThe viscous friction coefficient of the load converted to the motor rotating shaft is B_MLIf B is equal to B_M+B_ML；

M_LIs the torque on the motor shaft, i.e. the load moment;

eliminating intermediate variable M, i by the motor motion equation_a,E_aIt is possible to obtain:

note the book

Then

The above equation is the differential equation of the armature control DC motor, and the input quantity is the armature voltage u_aAnd load torque M_LThe output is the motor shaft rotation speed omega, u_aFor control input, M_LIs a disturbance input;

when the motor shaft rotation angle is theta as the output quantity, the relationship is brought

2. a method for modeling a load model for an exoskeleton robot as claimed in claim 1 further comprising a step of inputting a moving load prior to modeling the load model, the moving load input comprising software modeling calculations and mathematical power model inversion calculations;

software modeling calculation, after the mechanical structure of the exoskeleton robot is designed, joint driving input including angles, angular velocities, angular accelerations and the like is added, and joint load torques can be obtained after the structure moves; the calculation method needs to establish a mechanical structure simulation model in advance and can meet the requirements of several action modes;

the mathematical power model is subjected to inverse calculation, a multi-link mechanical mathematical model is established through Larlang day, and model parameters at the moment are necessarily set according to the design requirement of an actual mechanical structure; the lagrangian dynamics model for a mechanical system is:

wherein q is [ q ]₁q₂q₃]^TH (q) is an inertia matrix,

H₁₁(q)＝I_t+I_ub+m_sL_Gs ²+m_tL_s ²+m_tL_Gt ²+m_ubL_s ²+m_ubL_t ²

+m_ubL_Gub ²+2m_tL_GtL_scos(q₂)+2m_ubL_tL_scos(q₂)

+2m_ubL_GubL_tcos(q₃)+2m_ubL_GubL_scos(q₂+q₃)

H₁₂(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_t ²+m_ubL_Gub ²+2m_ubL_GubL_tcos(q₃)+m_ubL_tcos(q₂)+m_tL_GtL_scos(q₂)+m_ubL_GubL_scos(q₂+q₃)

H₁₃(q)＝I_u+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)+m_ubL_GubL_scos(q₂+q₃)

H₂₁(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_Gub ²+m_ubL_t ²+m_tL_GtL_scos(q₂)

+m_ubL_sL_tcos(q₂)+2m_ubL_GubL_tcos(q₃)

+m_ubL_sL_Gubcos(q₂+q₃)

H₂₂(q)＝I_t+I_ub+m_ubL_t ²+m_tL_Gt ²+m_ubL_Gub ²

+2m_ubL_GubL_tcos(q₃)

H₂₃(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₁(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

+m_ubL_GubL_scos(q₂+q₃)

H₃₂(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₃(q)＝I_ub+m_ubL_Gub ²

G₁(q)＝-m_ubgL_ssin(q₁)-m_ubgL_tsin(q₁+q₂)

-m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_ssin(q₁)

-m_tgL_Gtsin(q₁+q₂)-m_sgL_Gssin(q₁)

G₂(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_Gtsin(q₁+q₂)

-m_ubgL_tsin(q₁+q₂)

G₃(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)

The load torque of each joint can be approximately obtained by inputting gait motion in various action modes and substituting joint pose, angular velocity and angular acceleration serving as known quantities into a system Lagrange model.

3. A method for modeling a load model for an exoskeleton robot as claimed in claim 2 further comprising a reducer model building method;

after the reducer is added, the energy conservation theorem obtains that the torque of the motor can be improved by mu times, the rotating speed is reduced by mu times, and mu is the reduction ratio of the reducer, so the torque balance equation on the motor shaft is modified as follows:

in addition, need to convert into rotational inertia on the reduction gear output shaft with rotational inertia in the motor shaft, the conversion mode is:

J′_re＝J_re+μ²J_m

B′_re＝B_re+μ²B_m

wherein Jre is the rotational inertia of the speed reducer around the self axis, and Jm is the rotational inertia of the motor around the self axis, J'_reThe moment of inertia of the rotating shaft of the motor is converted into the moment of inertia of the output shaft of the speed reducer, Bre is the damping ratio of the speed reducer around the self shaft, Bm is the damping ratio of the motor around the self shaft, B'_reThe damping ratio of the rotational inertia on the rotating shaft of the motor to the output shaft of the speed reducer is reduced.

4. A method for modeling an exoskeleton robot load model as claimed in claim 3 further comprising a controller design, said controller design being:

U_d＝k_p(θ-θ_exp)+k_i∫(θ-θ_exp)+k_dω_d。

5. the method of modeling an exoskeleton robot load model as recited in claim 4 further comprising a motor reducer module design;

the input of the motor module is the voltage control quantity u (t) output by the controller, and the output is the torque M;

designing a model of the motor reducer according to the working principle of the motor and the reducer;

motor armature loop equation:

wherein E is_d＝K_eω_d

Since the inductance Ld is small and negligible, the armature current is expressed as:

the electromagnetic torque equation of the motor is as follows:

M_d＝K_mi_d

the torque output by the motor is T, and T is Md

After the reduction ratio u is added, the torque T' output by the reducer is equivalent to:

T′＝uT＝u(T_M-J_Mα-B_Mω)。

6. an exoskeleton robot load model modelling method as claimed in claim 5 wherein the knee joint motor parameter amplitude limits are as follows:

max(T′)＝91N·m，

max(ω_d)＝2.86rad/s，

max(i_d)＝12.9A，

max(T)＝0.6N·m。

Technical Field

The invention relates to the field of exoskeleton robots, in particular to a skeletal robot load model modeling and establishing method.

Background

The exoskeleton robot integrates the technologies of sensing, controlling, information fusion, mobile computing and the like, and provides a comprehensive electromechanical structure for enhancing the load bearing capacity and reducing the energy consumption of a wearer. The exoskeleton robot is a robot worn outside a human body, and is also called a wearable robot. The exoskeleton robot has the core function of assisting human bodies, so that heavier loads can be loaded, the load model calculation is particularly important, and the load of the existing exoskeleton robot has many unreasonable places.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a skeletal robot load model modeling and establishing method which is used for realizing load modeling of an exoskeleton robot.

The purpose of the invention is realized by the following technical scheme:

a skeletal robot load model modeling and establishing method comprises the following steps:

s100: setting the inductance of the armature winding to be L_aThe resistance of the armature winding is R_aArmature voltage of u_aThe current flowing through the armature is i_aThe angular speed of the motor shaft is omega, and the rotational angle of the motor shaft is theta;

s200: calculating parameters;

for the armature loop, the following can be obtained according to kirchhoff's voltage law:

in the formula

Is armature back electromotive force with its magnitude proportional to the field magnetic flux and the rotation speed and its direction proportional to the armature voltage u_aIn contrast, K_eIs the back electromotive force coefficient;

the motor has an electromagnetic torque equation of

M＝K_mi_a

Wherein K_mThe motor torque coefficient is, and M is the armature current to generate the electromagnetic torque;

equation of torque balance on motor shaft

Wherein:

j is equivalent moment of inertia converted to the shaft of the motor, if the shaft of the motor rotates by itselfAmount is J_MThe moment of inertia converted from load to motor shaft is J_MLWhen J is equal to J_M+J_ML；

B is the viscous friction coefficient converted to the motor shaft, and if the rotating shaft of the motor rotates, the viscous friction coefficient is B_MThe viscous friction coefficient of the load converted to the motor rotating shaft is B_MLIf B is equal to B_M+B_ML；

M_LIs the torque on the motor shaft, i.e. the load moment;

eliminating intermediate variable M, i by the motor motion equation_a,E_aIt is possible to obtain:

note the book

Then

The above equation is the differential equation of the armature control DC motor, and the input quantity is the armature voltage u_aAnd load torque M_LThe output is the motor shaft rotation speed omega, u_aFor control input, M_LIs a disturbance input;

when the motor shaft rotation angle is theta as the output quantity, the relationship is brought

The differential equation for a turn angle as output can be found as:

further, before modeling the load model, a motion load input step is also included, wherein the motion load input step comprises software modeling calculation and mathematical power model inversion calculation;

software modeling calculation, after the mechanical structure of the exoskeleton robot is designed, joint driving input including angles, angular velocities, angular accelerations and the like is added, and joint load torques can be obtained after the structure moves; the calculation method needs to establish a mechanical structure simulation model in advance and can meet the requirements of several action modes;

the mathematical power model is subjected to inverse calculation, a multi-link mechanical mathematical model is established through Larlang day, and model parameters at the moment are necessarily set according to the design requirement of an actual mechanical structure; the lagrangian dynamics model for a mechanical system is:

wherein q is [ q ]₁q₂q₃]^TH (q) is an inertia matrix,

is a Coriolis term, g (q) is a gravity term, T ═ T₁T₂T₃]Representing the resultant external moment, T, acting on the skeleton suit₁Indicating moment of ankle joint, T₂Representing knee joint moment, T₃Representing a hip joint moment; h, H (q),

the specific form of G (q) is as follows:

H₁₁(q)＝I_t+I_ub+m_sL_Gs ²+m_tL_s ²+m_tL_Gt ²+m_ubL_s ²+m_ubL_t ²+m_ubL_Gub ²+2m_tL_GtL_scos(q₂)+2m_ubL_tL_scos(q₂) +2m_ubL_GubL_tcos(q₃)+2m_ubL_GubL_scos(q₂+q₃)

H₁₂(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_t ²+m_ubL_Gub ²+2m_ubL_GubL_tcos(q₃) +m_ubL_tcos(q₂)+m_tL_GtL_scos(q₂)+m_ubL_GubL_scos(q₂+q₃)

H₁₃(q)＝I_u+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)+m_ubL_GubL_scos(q₂+q₃)

H₂₁(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_Gub ²+m_ubL_t ²+m_tL_GtL_scos(q₂) +m_ubL_sL_tcos(q₂)+2m_ubL_GubL_tcos(q₃) +m_ubL_sL_Gubcos(q₂+q₃)

H₂₂(q)＝I_t+I_ub+m_ubL_t ²+m_tL_Gt ²+m_ubL_Gub ²+2m_ubL_GubL_tcos(q₃)

H₂₃(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₁(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃) +m_ubL_GubL_scos(q₂+q₃)

H₃₂(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₃(q)＝I_ub+m_ubL_Gub ²

G₁(q)＝-m_ubgL_ssin(q₁)-m_ubgL_tsin(q₁+q₂) -m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_ssin(q₁) -m_tgL_Gtsin(q₁+q₂)-m_sgL_Gssin(q₁)

G₂(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_Gtsin(q₁+q₂) -m_ubgL_tsin(q₁+q₂)

G₃(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)

the load torque of each joint can be approximately obtained by inputting gait motion in various action modes and substituting joint pose, angular velocity and angular acceleration serving as known quantities into a system Lagrange model.

Further, a speed reducer model building method is also included;

after the reducer is added, the energy conservation theorem obtains that the torque of the motor can be improved by mu times, the rotating speed is reduced by mu times, and mu is the reduction ratio of the reducer, so the torque balance equation on the motor shaft is modified as follows:

in addition, need to convert into rotational inertia on the reduction gear output shaft with rotational inertia in the motor shaft, the conversion mode is:

J′_re＝J_re+μ²J_m

B′_re＝B_re+μ²B_m

wherein Jre is the rotational inertia of the speed reducer around the self axis, and Jm is the rotational inertia of the motor around the self axis, J'_reFor a motor to rotateThe rotational inertia on the shaft is converted into the rotational inertia on the output shaft of the speed reducer, Bre is the damping ratio of the speed reducer around the self shaft, Bm is the damping ratio of the motor around the self shaft, and B'_reThe damping ratio of the rotational inertia on the rotating shaft of the motor to the output shaft of the speed reducer is reduced.

Further, the system comprises a controller design, wherein the controller design is as follows:

U_d＝k_p(θ-θ_exp)+k_i∫(θ-θ_exp)+k_dω_d。

furthermore, the device also comprises a motor reducer module design;

the input of the motor module is the voltage control quantity u (t) output by the controller, and the output is the torque M;

designing a model of the motor reducer according to the working principle of the motor and the reducer;

motor armature loop equation:

wherein E is_d＝K_eω_d

Since the inductance Ld is small and negligible, the armature current is expressed as:

the electromagnetic torque equation of the motor is as follows:

M_d＝K_mi_d

the torque output by the motor is T, and T is Md

After the reduction ratio u is added, the torque T' output by the reducer is equivalent to:

T′＝uT＝u(T_M-J_Mα-B_Mω)。

further, the amplitude of the motor parameters of the knee joint is limited as follows:

max(T′)＝91N·m，

max(ω_d)＝2.86rad/s，

max(i_d)＝12.9A，

max(T)＝0.6N·m

the invention has the beneficial effects that: according to the scheme, the modeling is respectively carried out on the motor, the speed reducer, the controller and the load input, so that the obtained motor load model is more fit for practical application, the rationality of load control is improved, and the fitting degree of the exoskeleton robot and the human body is greatly improved.

Drawings

FIG. 1 is a process of model selection for a motor and a reducer;

FIG. 2 is a motor load model;

FIG. 3 is a schematic diagram of a driving circuit;

FIG. 4 is a diagram of a person carrying a squat load condition;

FIG. 5 is a graph of angular velocity change of a joint;

FIG. 6 is a diagram of a model-based control system simulation framework;

FIG. 7 is a knee joint desired curve generation module during walking;

FIG. 8 is a model of a controller;

FIG. 9 is a knee motor and reducer module;

FIG. 10 is a first kinetic module;

FIG. 11 is a kinetic module two;

fig. 12 shows a motor control method.

Detailed Description

The technical solution of the present invention is further described in detail with reference to the following specific examples, but the scope of the present invention is not limited to the following.

A building method of a skeletal robot load model comprises the steps of firstly carrying out motion load input calculation before modeling.

In the motion process, under various action modes, joint loads are dynamically changed, so that important consideration needs to be given to the selection of a servo motor and a speed reducer, and the maximum torque required to be output by the motor can meet the requirement of a normal action mode particularly under the condition of paying attention to different action modes. There are two control methods for determining the moving load input.

The method comprises the following steps: modeling calculation through professional mechanical software

In the process of designing the mechanical structure of the skeleton garment system, the strength of each joint and each limb needs to be considered, static analysis and dynamic analysis are needed, and the motion analysis process is generally analyzed through software such as Adamas, ProE, Solid Edge and Solid works. Therefore, the analysis through professional mechanical software is more accurate and reliable, and is the first method for calculating the joint motion load. The calculation idea is as follows: after the mechanical structure is well designed, joint driving input including angles, angular velocities, angular accelerations and the like is added, and joint load torques can be obtained after the structure moves. The calculation method needs to establish a mechanical structure simulation model in advance and can meet the requirements of several action modes.

The second method comprises the following steps: by inversion calculation of mathematical dynamic model

If the joint load torque can not be normally obtained through mechanical structure simulation software, the calculation can only be carried out through the inversion of a mathematical dynamics model. The calculation idea is as follows: a multi-link mechanical mathematical model is established through Larlang, and model parameters at the moment are necessarily set according to the design requirements of an actual mechanical structure. The lagrangian dynamics model for a mechanical system is:

wherein q is [ q ]₁q₂q₃]^TH (q) is an inertia matrix,

is a Coriolis term, g (q) is a gravity term, T ═ T₁T₂T₃]Representing the resultant external moment, T, acting on the skeleton suit₁Indicating moment of ankle joint, T₂Representing knee joint moment, T₃Representing the hip joint moment. H, H (q),

specific form of G (q)Has the following formula

H₁₁(q)＝I_t+I_ub+m_sL_Gs ²+m_tL_s ²+m_tL_Gt ²+m_ubL_s ²+m_ubL_t ²+m_ubL_Gub ²+2m_tL_GtL_scos(q₂)+2m_ubL_tL_scos(q₂) +2m_ubL_GubL_tcos(q₃)+2m_ubL_GubL_scos(q₂+q₃)

H₁₂(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_t ²+m_ubL_Gub ²+2m_ubL_GubL_tcos(q₃) +m_ubL_tcos(q₂)+m_tL_GtL_scos(q₂)+m_ubL_GubL_scos(q₂+q₃)

H₁₃(q)＝I_u+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)+m_ubL_GubL_scos(q₂+q₃)

H₂₁(q)＝I_t+I_ub+m_tL_Gt ²+m_ubL_Gub ²+m_ubL_t ²+m_tL_GtL_scos(q₂) +m_ubL_sL_tcos(q₂)+2m_ubL_GubL_tcos(q₃) +m_ubL_sL_Gubcos(q₂+q₃)

H₂₂(q)＝I_t+I_ub+m_ubL_t ²+m_tL_Gt ²+m_ubL_Gub ²+2m_ubL_GubL_tcos(q₃)

H₂₃(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₁(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃) +m_ubL_GubL_scos(q₂+q₃)

H₃₂(q)＝I_ub+m_ubL_Gub ²+m_ubL_GubL_tcos(q₃)

H₃₃(q)＝I_ub+m_ubL_Gub ²

G₁(q)＝-m_ubgL_ssin(q₁)-m_ubgL_tsin(q₁+q₂) -m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_ssin(q₁) -m_tgL_Gtsin(q₁+q₂)-m_sgL_Gssin(q₁)

G₂(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)-m_tgL_Gtsin(q₁+q₂) -m_ubgL_tsin(q₁+q₂)

G₃(q)＝-m_ubgL_Gubsin(q₁+q₂+q₃)

The load torque of each joint can be approximately obtained by inputting gait motion in various action modes and substituting joint pose, angular velocity and angular acceleration serving as known quantities into a system Lagrange model.

As a preferable scheme of this embodiment, the method further includes a method for selecting the type of the motor and the reducer, and the type selection of the motor and the reducer needs to be considered in a compromise manner. The structural size of the motor and the reducer is the first major constraint because the exoskeleton requires that the structural size of each component is as small as possible. And secondly, the motor has insufficient output torque, and a reducer is required to be configured to meet the requirement of large load. During model selection, according to the load torque determined by each joint, firstly determining a torque parameter, and then determining the requirement of a rotating speed index. The method is characterized in that the configuration type of the low-torque motor and the high-reduction-ratio speed reducer is determined according to load torque, and the minimum size of the motor is guaranteed. If the joint rotating speed can not meet the maximum rotating speed requirement in the action mode, the joint rotating speed can be properly adjusted; if the maximum torque and the maximum rotating speed required by the action mode are not at the same point, maintaining the model selection parameter unchanged; if the joint torque meets the requirement but the rotating speed cannot meet the requirement, the original output torque of the motor needs to be increased, and the rotating speed ratio of the speed reducer needs to be reduced. The motor and reducer model selection process is shown in fig. 1.

(1) Establishing a motor load model:

the operating principle of the dc servomotor is shown in fig. 2. The operation of the direct current motor is that input electric energy is converted into mechanical energy, for the direct current motor shown in fig. 2, input armature voltage generates armature current in an armature loop, and a closed coil flowing through the armature current interacts with a magnetic field to generate electromagnetic torque to drive a load to rotate. The entire armature circuit can be simplified as shown in fig. 3 when the dc servo motor drives the load.

Let the inductance of the armature winding be L_aThe resistance of the armature winding is R_aArmature voltage of u_aThe current flowing through the armature is i_aThe angular velocity of the motor shaft is Ω, and the motor shaft angle is θ.

For the armature loop, the following can be obtained according to kirchhoff's voltage law:

in the formula

Is armature back electromotive force with its magnitude proportional to the field magnetic flux and the rotation speed and its direction proportional to the armature voltage u_aIn contrast, K_eIs the back electromotive force coefficient.

The motor has an electromagnetic torque equation of

M＝K_mi_a

Wherein K_mFor the motor torque coefficient, M is the armature current producing electromagnetic torque.

Equation of torque balance on motor shaft

Wherein:

j is the equivalent moment of inertia translated to the motor shaft. If the self rotational inertia of the motor rotating shaft is J_MThe moment of inertia converted from load to motor shaft is J_MLWhen J is equal to J_M+J_ML。

And B is the coefficient of viscous friction translated to the motor shaft. If the self-rotation viscosity friction coefficient of the motor rotating shaft is B_MThe viscous friction coefficient of the load converted to the motor rotating shaft is B_MLIf B is equal to B_M+B_ML。

M_LThe torque on the motor shaft, i.e., the load torque (disturbance torque).

Eliminating intermediate variable M, i by the motor motion equation_a,E_aIt is possible to obtain:

note the book

Then

The above equation is the differential equation of the armature control DC motor, and the input quantity is the armature voltage u_aAnd load torque M_LAnd the output is the motor shaft rotating speed omega. u. of_aFor control input, M_LIs the perturbation input.

When the motor shaft rotation angle is theta as the output quantity, the relationship is brought

The differential equation for a turn angle as output can be found as:

(2) the method comprises the following steps of (1) establishing a reducer model:

after the reducer is added, the torque of the motor can be improved by mu times and the rotating speed is reduced by mu times by the energy conservation theorem. μ is the reduction ratio of the reducer. The torque balance equation on the motor shaft is thus modified to:

in addition, need to convert into rotational inertia on the reduction gear output shaft with rotational inertia in the motor shaft, the conversion mode is:

J′_re＝J_re+μ²J_m

B′_re＝B_re+μ²B_m

jre is the moment of inertia of the speed reducer around its own axis, and Jm is the moment of inertia of the motor around its own axis. J'_reThe rotational inertia of the motor rotating shaft is converted into the rotational inertia of the output shaft of the speed reducer. Bre is the damping ratio of the speed reducer around the self shaft, Bm is the damping ratio of the motor around the self shaft. B'_reThe damping ratio of the rotational inertia on the rotating shaft of the motor to the output shaft of the speed reducer is reduced.

And (3) control simulation of lower limb joint movement:

aiming at the movement of hip joints and knee joints of lower limbs, respective control laws are designed, the movement model simulation is carried out, and a mathematical model is built and is shown in a Simulink file. The important parameters such as angle, angular speed, motor voltage control quantity, load torque and the like can be observed. The simulation model can be used as a reference basis for preliminary model selection of the motor.

Fig. 6 is a simulation framework diagram of an exoskeleton robot control system built by Matlab. Wherein the SubThetastexp submodule is an expected angle curve generation module; the SubgstRotate is a control object module, and the object module is constructed according to the kinematic dynamics analysis of the model; subcobtrl is the controller module. This section sequentially carries out detailed design on each module.

1) Desired angle generating module design

According to a function expression fitting expected input curves of the knee joint and the hip joint in the squatting and rising walking mode. The expected input angle generation module is a module built by using the function expression obtained by fitting. The function of this module is to output the desired input profile of the control system. FIG. 7 is a block diagram of a knee joint desired curve generation module during walking.

2) Controller

The controller comprises a proportion link, a differentiation link and an integration link. The proportional link is used for proportionally reflecting the deviation of the system, and when the system has the deviation, the proportional link immediately plays a role in eliminating the deviation, and the disadvantage is that a steady-state error exists. The integration link has the function of eliminating the steady-state error of the system, the size of the integration function is determined by setting an integration constant, and the integration link has the defect that the stability of the system is poor. The differential link can reflect the variation trend of system errors and accelerate the response speed.

The controller is obtained by combining the characteristics of proportional control, integral control and differential control. The controller can eliminate the defects caused by a single controller, and the controller of the system is designed as follows:

U_d＝k_p(θ-θ_exp)+k_i∫(θ-θ_exp)+k_dω_d

the Simulink model of the controller is shown in fig. 8. The output value u (t) of the controller is the sum of proportional, integral and derivative of the polynomial in the formula. ω d is the angular velocity, which can be measured by a gyroscope. The ideal control effect of the system can be achieved by adjusting the coefficients Kp, Ki and Kd.

3) Motor reducer module design

The input of the motor module is the voltage control quantity u (t) output by the controller, and the output is the torque M. And designing a model of the motor reducer according to the working principle of the motor and the reducer.

Motor armature loop equation:

wherein E is_d＝K_eω_d

Since the inductance Ld is small and negligible, the armature current is expressed as:

the electromagnetic torque equation of the motor is as follows:

M_d＝K_mi_d

the torque output by the motor is T, and T is Md

After the reduction ratio u is added, the torque T' output by the reducer is equivalent to:

T′＝uT＝u(T_M-J_Mα-B_Mω)

and finally, limiting the amplitude of some important parameters, wherein the amplitude of the knee joint motor parameters is limited as follows:

max(T′)＝91N·m，

max(ω_d)＝2.86rad/s，

max(i_d)＝12.9A，

max(T)＝0.6N·m

after adding the actuator model and the above related constraints, the Simulink model for building the motor and reducer module is shown in fig. 9.

4) Control object model

And performing single-side lower limb simulation on the system by using the model of the system. The object model consists of two modules. One is a kinematic module, shown in FIG. 10, whose inputs are the component accelerations and the moments required for the joint, and outputs are the constructed angular velocities. One is a kinematic module, shown in fig. 11, whose inputs are the angular changes of the joints and outputs are the position, velocity and acceleration of the member. In summary, the input of the entire target module is torque, and the output is angle θ.

Personnel safety guarantee:

the personnel safety assurance strategies for exoskeletons are mainly divided into the following two categories:

1) active safety protection

The active safety protection is that in a control loop, safety control is at the highest priority of the control loop, the postures of all parts of a human body are settled in real time through detecting the joint angle and the inclination angle of the human body in real time and inverse kinematics, once the influence of exoskeleton action on the safety state of the human body is found, all other tasks in the control loop are immediately stopped, a safety protection program is started, a system is forced to return to be stable, the human body and the machine body are helped to be restored to a set safety initial state through set control steps, and then the execution of other tasks is allowed.

2) Passive safety protection

The passive safety protection is characterized in that an external safety button is arranged on the exoskeleton body, when a wearer finds that the exoskeleton is possibly unstable and the body does not perform corresponding protection actions, the safety button can be pressed down, at the moment, the exoskeleton can forcibly throw off the load to reduce the weight, and meanwhile, the exoskeleton body is powered off, at the moment, the load of the wearer is only the exoskeleton body, and the exoskeleton can move independently easily, so that the dangerous state is relieved. The exoskeleton state can be judged through a fault detection and initialization button, and the next action is determined.

The lower limb power driving mode is as follows:

the Hall sensor and the encoder of the motor can be directly connected with the controller, the built-in logic of the controller obtains data collected by the sensor, and closed-loop control is carried out according to expected data set by the core board, so that the control difficulty of the motor is greatly reduced. The whole motor control mode is shown in fig. 12.

The foregoing is illustrative of the preferred embodiments of this invention, and it is to be understood that the invention is not limited to the precise form disclosed herein and that various other combinations, modifications, and environments may be resorted to, falling within the scope of the concept as disclosed herein, either as described above or as apparent to those skilled in the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.