CN105242677A - Quadruped robot biped support phase force hybrid force control method - Google Patents

Abstract

The invention provides a quadruped robot biped support phase force hybrid force control method. The method comprises the steps of (S1) projecting the overall movement of a robot to a radial plane and a normal plane, (S2) establishing a control model, simplifying the movement of the robot along the radial plane as a plane seven-connection-rod model, and simplifying the movement along the normal plane as a linear inverted pendulum model, and simplifying the plane seven-connection-rod model as a plane virtual telescoping leg model, wherein the control target of the plane virtual telescoping leg model is a mass center height, a body pitch angle and a horizontal displacement, (S3) carrying out hybrid controlling according to a control model, establishing the kinetic equation of the plane virtual telescoping leg model and establishing the kinetic equation through a Newton-Euler method, controlling the mass center height and the body pitch angle through a position servo method, and using the double-ring control method of outer ring position and inner ring foot end force for the horizontal displacement of a virtual telescoping leg plane model. The method has the advantages of good control effect and the improvement of robot adaptability.

Description

Hybrid control method for two-foot support phase force position of four-foot robot

Technical Field

The invention mainly relates to the technical field of robot motion control, in particular to a biped support phase force position hybrid control method suitable for a quadruped robot.

Background

The four-footed bionic robot can reach places which can not be reached by human beings or cannot be reached due to environmental danger, is a footed type robot imitating the motion mode of four-footed mammals, can carry out tasks such as material transportation, patrol and the like in a non-structural environment due to the strong adaptability of the four-footed bionic robot to complex terrains such as rocks, steep slopes and the like, and can replace human beings to carry out dangerous operation, so that the four-footed bionic robot has great research value.

Quadruped robots generally consist of four biomimetic legs and a body, each leg comprising a lateral joint and at least two anterior joints. By simulating and learning the motion mode of the quadruped robot in nature, the motion mode of the quadruped robot at present mainly has three gaits: TROT gait (diagonal sprint gait), BOUND gait (running gait) and WALK gait (crawling gait).

Control of a quadruped robot is a typical floating base control problem, with the control target generally being its body pose, which is primarily controlled by the support legs. Since the quadruped robot has a large number of degrees of freedom, it is very complicated to realize the overall control of the robot, and therefore, it is necessary to appropriately simplify the model.

When the quadruped robot is modeled, a multi-link plane model and a SLIP model are currently common and effective simplified models, both the models can reflect the kinematics and the dynamic characteristics of a trot gait, and a controller is designed according to the kinematics and the dynamic equation of a model-based control method.

The control method based on kinematics adopts a servo control rate, calculates the position information of each joint according to sensor information and a control target, and realizes the tracking of an expected track through a position servo law.

A control method based on dynamics generally adopts a multi-loop control method such as a moment calculation method and the like, calculates the foot end expected force through the pose information of the robot body, and controls the foot end expected force through calculating the joint driving moment, thereby achieving the purpose of controlling the pose of the robot body. The control method has good flexibility and strong adaptability to the unstructured environment. However, there is a disadvantage in that an internal force may be generated between the two support legs, the internal force is instantaneously released at the time when the support legs are switched to the swing leg mode, an unexpected impact is generated on the posture of the swing leg and the body, and the rigidity of the actuator is largely influenced by the overall mass with respect to the position control. Therefore, the force control method needs to design a reasonable force distribution method, and the movement capability of the robot can be enhanced while the pose of the robot is controlled.

Disclosure of Invention

The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a quadruped robot biped support phase force position hybrid control method which is good in control effect and capable of improving the adaptability of the robot.

In order to solve the technical problems, the invention adopts the following technical scheme:

a method for hybrid control of two-foot support phase force positions of a four-foot robot comprises the following steps:

s1: projecting the overall motion of the robot onto a radial plane and a normal plane; the radial plane is a plane which supports the feet and is vertical to the horizontal plane; the normal plane is a plane passing through the center of mass of the body and perpendicular to the radial plane;

s2: establishing a control model; the motion of the robot along the radial plane is simplified into a plane seven-connecting-rod model, and the motion on the normal plane is simplified into a linear inverted pendulum model; then simplifying the plane seven-connecting-rod model into a plane virtual telescopic leg model; the control targets of the plane virtual telescopic leg model are the height of a mass center, a pitch angle of the body and horizontal displacement;

s3: performing mixing control according to a control model; establishing a kinematic equation of a plane virtual telescopic leg model and a dynamic equation thereof by a Newton-Euler method, and adopting a force control mode for the horizontal displacement of the virtual telescopic leg plane model by a position servo control method for the height of a mass center and a body pitch angle.

As a further improvement of the invention: in step S3, the height of the center of mass and the pitch angle of the body are controlled by the leg length, and the horizontal displacement of the body is controlled by controlling the horizontal foot end force, so as to realize approximate decoupling control of the posture of the body.

As a further improvement of the invention: a classical PID controller is used as a force outer ring controller, so that the contact force of the foot end of the supporting leg accurately tracks the expected force.

As a further improvement of the invention: and a position servo controller is adopted to control the knee joint and the ankle joint so as to realize the tracking of the joint angle.

Compared with the prior art, the invention has the advantages that:

1. when the quadruped robot Trot moves in a gait mode, the invention can realize accurate tracking control of the body pose and has stronger adaptability to discontinuous expected tracks.

2. The force distribution strategy based on the positive pressure at the foot end can effectively avoid the phenomenon that the foot end of the supporting leg slides in the walking process of the robot, and enables the robot to have stronger acceleration performance.

3. The invention can improve the adaptability of the quadruped robot to the unstructured environment and realize the walking of the robot on uneven ground.

4. The invention can improve the load capacity of the robot by increasing the rigidity of the position controller.

5. The invention has clear structure and distinct hierarchy, and has better theoretical value and engineering significance.

Drawings

Fig. 1 is a schematic structural diagram of a quadruped robot platform in a specific application example of the invention.

FIG. 2 is a schematic view of a radial plane projection in a specific application example of the present invention.

Fig. 3 is a decoupling control block diagram of a planar virtual telescopic leg structure in a specific application example of the present invention.

FIG. 4 is a schematic flow diagram of the method of the present invention.

Detailed Description

The invention will be described in further detail below with reference to the drawings and specific examples.

The method of the invention is mainly suitable for quadruped robots. As shown in fig. 1, the system structure of the quadruped robot is composed of a body and four legs, wherein each leg comprises a hip lateral rotating joint and three forward rotating joints (a hip anterior joint, a knee joint and an ankle joint). Each joint is driven by a hydraulic actuator and incorporates a displacement sensor and a force sensor for sensing the length of the actuator and the driving force. And a three-dimensional force sensor is arranged at the foot end of each leg and is used for detecting contact force information of the foot end of the robot and the environment and detecting pose information of the robot under an inertial system through the IMU. The quadruped robot has the following mechanical structure characteristics: (1) the gravity center of the robot is close to the geometric center of the body; (2) the mass of the robot body is much greater than that of the legs.

In the present invention, a force and position hybrid control method for biped robot biped support will be described by taking a TROT gait biped support phase as an example. The invention discloses a biped support phase force position hybrid control method for a quadruped robot, which adopts a force position hybrid control method to carry out decoupling control on the pose of a robot body so as to enhance the adaptability of the robot to the non-structural ground and load changes. The invention provides a force distribution method based on the positive pressure of the foot end, which reduces the possibility of the foot end of a supporting leg sliding and improves the acceleration performance of a robot.

As shown in fig. 4, the method for hybrid controlling the phase force position of the biped support of the quadruped robot comprises the following steps:

s1: projecting the overall motion of the robot onto a radial plane and a normal plane; the radial plane is a plane which supports the feet and is vertical to the horizontal plane; the normal plane is a plane passing through the center of mass of the body and perpendicular to the radial plane;

because the four-footed robot degree of freedom is various, it is comparatively complicated to establish the holistic kinematics of robot and kinetic equation, consequently need rationally simplify the four-footed robot: the trot gait is a repeated switching process of two foot support states, so the present invention projects the overall motion of the robot to a radial plane supported by the pair of feet and perpendicular to the horizontal plane, and a normal plane passing through the center of mass and perpendicular to the radial plane.

S2: establishing a control model;

in a specific application example, the motion of the robot along the radial plane can be simplified into a plane seven-link model, and the motion along the normal plane can be simplified into a linear inverted pendulum model.

The single leg of the plane seven-connecting-rod model is provided with three forward rotating joints, and the kinematics and the dynamics characteristics of the single leg can be equivalent to a telescopic leg structure model, so that the plane seven-connecting-rod model is simplified into a plane virtual model with a telescopic leg structure. The control targets of the plane virtual telescopic leg model are the height of a mass center, the pitch angle of the body and horizontal displacement.

S3: performing mixing control according to a control model; establishing a kinematic equation of a plane virtual telescopic leg model and a dynamic equation thereof by a Newton-Euler method, and adopting a force control mode for the horizontal displacement of the virtual telescopic leg plane model by a position servo control method for the height of a mass center and a body pitch angle. Namely:

first, a kinematic equation of the model is established and a kinetic equation thereof is established by a newton-euler method. The position control method has high rigidity, so that the loading capacity of the robot is improved by the position servo control method for the height of the mass center and the pitch angle of the body.

Because the plane virtual telescopic leg model contains 4 driving joints and the overall control target is only three, the overall control of the model can be realized only by adding a constraint condition.

Because the horizontal foot end force of the supporting leg is naturally constrained by the friction cone, the whole structure is also constrained by the dynamics of the ZMP point, and in order to prevent the foot end of the virtual telescopic leg model from sliding, a constraint equation needs to be added to the horizontal foot end force, namely the horizontal force distribution problem. Therefore, a force control method is adopted for the horizontal displacement of the virtual telescopic leg plane model.

In a specific application example, the projection of the quadruped robot in the radial plane can be simplified into a seven-link structure as shown in fig. 2 (the left front leg and the right rear leg are used as supporting legs). Wherein, (x, z) is the position of the mass center of the body under an inertial coordinate system, m is the mass of the body, and L₀Is the length of the robot body, d is the width of the body, l₁,l₂,l₃The length of ankle joint, knee joint and hip joint of leg respectively, theta₁,θ₂,θ₃Is the ankle joint angle, knee joint angle, hip joint angle of the left front leg theta₄,θ₅,θ₆The ankle joint angle, the knee joint angle and the hip joint angle of the right hind leg,the pitch angle of the body is O, A, B, C, D being the origin of the world system (i.e. the connection line of the foot ends of the two supporting legs)Midpoint), right rear leg hip and body connection point, left front leg hip and body connection point, right rear leg foot endpoint, left front leg foot endpoint, F_x1,F_z1,F_x2,F_z2Respectively, the contact force received at points C and D. Leg mass and inertia are ignored in modeling because the leg mass is small relative to the body mass.

From the left diagram in fig. 2, it can be obtained that the leg lengths AC and BD are respectively:

$\left|\right|AC\left|\right|=\sqrt{{\left(l_1+l_{2}cos\left(\mathrm{\π}-{\mathrm{\θ}}_4\right)+l_3\cos \left({\mathrm{\θ}}_5-{\mathrm{\θ}}_4\right)\right)}^2+{(l_{2}sin(\mathrm{\π}-{\mathrm{\θ}}_4)+l_{3}sin({\mathrm{\θ}}_5-{\mathrm{\θ}}_4))}^2}---\left(1\right)$

$\left|\right|BD\left|\right|=\sqrt{{\left(l_1+l_{2}cos\left(\mathrm{\π}-{\mathrm{\θ}}_1\right)+l_3\cos \left({\mathrm{\θ}}_2-{\mathrm{\θ}}_1\right)\right)}^2+{(l_{2}sin(\mathrm{\π}-{\mathrm{\θ}}_1)+l_{3}sin({\mathrm{\θ}}_2-{\mathrm{\θ}}_1))}^2}---\left(2\right)$

as can be seen from the expressions (1) to (2), the leg lengths | | | AC | | and | | | BD | | | and the joint angle θ, respectively₄,θ₅And theta₁,θ₂Accordingly, the articulated leg illustrated in the left diagram of FIG. 2 can be reduced kinematically to the virtual telescoping leg model illustrated in the right diagram, wherein α₃,α₄Respectively the front and rear leg incident angles.

From fig. 2, a kinematic model of a planar virtual telescopic leg structure can be established as follows:

$y=\frac{L_{1}sin\left({\mathrm{\α}}_3\right)+L_{2}sin\left({\mathrm{\α}}_4\right)}{2}---\left(4\right)$

$x=\frac{L_{1}cos\left({\mathrm{\α}}_3\right)+L_{2}cos\left({\mathrm{\α}}_4\right)}{2}---\left(5\right)$

the kinetic equation is also available as follows:

$f_{x1}+f_{x2}=m\stackrel{\·\·}{x}---\left(6\right)$

$f_{z1}+f_{z2}=mg+m\stackrel{\·\·}{z}---\left(7\right)$

the moment for the hip joint is as follows:

τ₁＝L₁sin(α₃)f_x1-L₁cos(α₃)f_z1(9)

τ₂＝L₂sin(α₄)f_x1-L₂cos(α₄)f_z1(10)

for the kinetic equations (6) to (10), the telescoping leg model shown in the right diagram in fig. 2 and the joint leg model shown in the left diagram in fig. 2 are also equivalent. Thus, the planar seven-link model shown in the left diagram of fig. 2 can be simplified from both kinematic and dynamic aspects to the virtual telescoping leg model shown in the right diagram of fig. 2.

Based on the method, in a specific application example, the pose needs to be decoupled and controlled;

for incident angle α₃,α₄The method comprises the following steps:

0<α₃<π,0<α₄<π(11)

thus, there are:

0<sin(α₃)≤1,0<sin(α₄)≤1(12)

-1<cos(α₃)<1,-1<cos(α₄)<1(13)

for forward displacement, it can be seen from equation (5) that there is a singular point α₃＝α₄Pi/2, it is therefore not suitable to control the forward displacement with leg length. However, there are no singular points for equations (3), (4), and thus, the leg length l can be passed₁,l₂To control the pitch angle and the height of the mass center of the body, and for the sake of simplicity, the controlIntroduction control bookBody pitch angleMethod of (1), and

writing (3) and (4) in matrix form is:

$\left[\begin{array}{c}{S}_y\\ y\end{array}\right]=J_1\left[\begin{array}{c}{L}_1\\ L_2\end{array}\right]---\left(14\right)$

wherein, $J_1=\left[\begin{array}{cc}\frac{sin\left({\mathrm{\&alpha;}}_3\right)}{2L_0}& -\frac{sin\left({\mathrm{\&alpha;}}_4\right)}{2L_0}\\ \frac{sin\left({\mathrm{\&alpha;}}_3\right)}{2}& \frac{\sin \left({\mathrm{\&alpha;}}_4\right)}{2}\end{array}\right].$

from the formula (3), J₁Nonsingular, and forward displacement versus body pitch angle and massThe effect of heart height may be determined by the Jacobian matrix J₁Elimination, and therefore can be considered as approximately decoupled kinematically.

Because the control target of the plane telescopic leg structure has horizontal displacement, a mass center height and a body pitch angle, wherein the mass center height and the body pitch angle are controlled by the leg length, and the forward displacement can be realized by the control of the foot end force. In the dynamic equation, only the equation (6) related to the forward position needs to be added with a horizontal force distribution equation to realize the calculation of the end force of the foot, and further realize the control of the horizontal displacement. To minimize the possibility of foot slip, a force distribution method is used as described below, and a dual loop control method is used for control.

The probability of slip at the foot end is defined as SCI (slip constraint index):

$SCI=max\left(\frac{\left|F_x\right|}{\left|F_z\right|}\right)---\left(15\right)$

wherein F_x，F_zThe ground reaction force received by the robot during walking. The smaller the SCI value, the less likely slippage at the foot end will occur. To keep the SCI as small as possible, the force distribution can be:

f_x1d/f_x2d＝f_z1/f_z2(16)

the expected foot end force obtained by combining (6) and (16) is as follows:

$\left[\begin{array}{c}{f}_{x1d}\\ f_{x2d}\end{array}\right]=\left[\begin{array}{c}{f}_{z1r}/(f_{z1r}+f_{z2r})\\ f_{z2r}/(f_{z1r}+f_{z2r})\end{array}\right]m\stackrel{\&CenterDot;\&CenterDot;}{x}---\left(17\right)$

the horizontal foot end desired force calculated by equation (17) is completely independent of the centroid height and body pitch angle, so the outer loop, i.e., horizontal foot end force and the other two control objectives, can be considered to be kinematically decoupled.

The horizontal foot end force is controlled by joint moment, and the following can be obtained from (9) and (10):

$\left[\begin{array}{c}{\mathrm{\&tau;}}_1\\ {\mathrm{\&tau;}}_2\end{array}\right]=J_2\left[\begin{array}{c}{f}_{x1}\\ f_{x2}\end{array}\right]+G---\left(18\right)$

wherein, $J_2=\left[\begin{array}{cc}{L}_{1}sin\left({\mathrm{\&alpha;}}_3\right)& 0\\ 0& L_2\sin \left({\mathrm{\&alpha;}}_4\right)\end{array}\right],G=\left[\begin{array}{c}-L_{1}cos\left({\mathrm{\&alpha;}}_3\right)f_{z1}\\ -L_{2}cos\left({\mathrm{\&alpha;}}_4\right)f_{z2}\end{array}\right].$

from the formula (3), J₂Is nonsingular and passes through J₂And the compensation term G eliminates the effect of the centroid height and pitch angle on the joint moments, so the inner loop can be considered to be approximately kinetically decoupled.

By adopting the method, the height of the mass center and the pitch angle of the body are controlled through the leg length, the horizontal displacement of the body is controlled through controlling the end force of the horizontal foot, and the approximate decoupling control of the pose of the body can be realized.

Based on the method of the invention, in a specific application example, controller design is required;

1) designing a force controller;

based on the formula (17), a classical PID controller is designed as a force outer ring controller, and the specific form is as follows:

$\left[\begin{array}{c}{f^{\&prime;}}_{x1d}\\ {f^{\&prime;}}_{x2d}\end{array}\right]=m\left[\begin{array}{c}{f}_{z1r}/(f_{z1r}+f_{z2r})\\ f_{z2r}/(f_{z1r}+f_{z2r})\end{array}\right]\&lsqb;{\mathrm{Kp}}_x(x_d-x)+{\mathrm{Kd}}_x({\stackrel{\&CenterDot;}{x}}_d-\stackrel{\&CenterDot;}{x})+{\mathrm{Ki}}_x{\&Integral;}_0^t(x_d-x)d\mathrm{\&tau;}\&rsqb;---\left(19\right)$

wherein, f'_x1d,f'_x2dPeriod of time ofContact force of the foot end, x_dAt a desired x-direction position, f_z1r,f_z2rIs the vertical foot end force.

Because the robot needs to keep balance in motion and is restrained by the ZMP points, for the virtual telescopic leg structure, in order to maintain the support state of the two feet of the robot, the ZMP points need to be limited on the line segment of the connecting line of the two feet. The constraint, without regard to centroid height and pitch dynamics, is of the form:

$k_1\frac{x_1-x}{z}g\&le;\frac{{\mathrm{Kp}}_x(x_d-x)+{\mathrm{Kd}}_x({\stackrel{\&CenterDot;}{x}}_d-\stackrel{\&CenterDot;}{x})+{\mathrm{Ki}}_x{\&Integral;}_0^t(x_d-x)d\mathrm{\&tau;}}{m}\&le;k_1\frac{x-x_2}{z}g---\left(20\right)$

wherein, 0<k₁<1，x₁,x₂The horizontal positions of the points D and C are respectively.

In addition, the foot end is subject to the natural constraints of the friction cone: f'_x1d<μf_z1r、f'_x1d<μf_z1dμ is the coefficient of friction, the controller becomes:

$f_{x1d}=\left\{\begin{array}{cc}{f^{\&prime;}}_{x1d}& -f_{x1\_max}<{f^{\&prime;}}_{x1d}<f_{x1\_max},f_{x1\_max}=k_2{\mathrm{\&mu;f}}_{z1r}\\ -f_{x1\_max}& {f^{\&prime;}}_{x1d}\&le;-f_{x1\_max},f_{x1\_max}=k_2{\mathrm{\&mu;f}}_{z1r}\\ f_{x1\_max}& {f^{\&prime;}}_{x1d}\&GreaterEqual;f_{x1\_max},f_{x1\_max}=k_2{\mathrm{\&mu;f}}_{z1r}\end{array}\right.---\left(21\right)$

$f_{x2d}=\left\{\begin{array}{cc}{f^{\&prime;}}_{x2d}& -f_{x2\_max}<{f^{\&prime;}}_{x2d}<f_{x2\_max},f_{x2\_max}=k_2{\mathrm{\&mu;f}}_{z2r}\\ -f_{x2\_max}& {f^{\&prime;}}_{x2d}\&le;-f_{x2\_max},f_{x2\_max}=k_2{\mathrm{\&mu;f}}_{z2r}\\ f_{x2\_max}& {f^{\&prime;}}_{x2d}\&GreaterEqual;f_{x2\_max},f_{x2\_max}=k_2{\mathrm{\&mu;f}}_{z2r}\end{array}\right.---\left(22\right)$

by adjusting the parameter Kp_x,Kd_x,Ki_xThe forward displacement can be controlled, and the phenomenon that the robot turns over or the foot end slides in the motion process is avoided.

Based on the formula (18), the contact force of the foot end of the supporting leg can accurately track the expected force by adopting a classical PID controller.

2) Designing a position controller;

based on equation (14), the inverse kinematics equation can be derived:

$\left[\begin{array}{c}{L}_1\\ L_2\end{array}\right]=J_1^{-1}\left[\begin{array}{c}{S}_y\\ y\end{array}\right]---\left(23\right)$

adopting a position servo controller:

$f={\mathrm{Kp}}_L(L_d-L)+{\mathrm{Kv}}_L({\stackrel{\&CenterDot;}{L}}_d-\stackrel{\&CenterDot;}{L})---\left(24\right)$

wherein f is the thrust on the actuator. By adjusting the controller parameter Kp_L,Kv_LThe leg length can be controlled and the position controller can have great rigidity. However, since the discontinuity of position estimation often occurs during leg switching or irregular ground operation in Trot gait, the robot can be kept running in balance by limiting the driving force for small deviation between the actual value and the expected value, but the difference between the actual leg length and the expected leg length is very large easily in Trot gait, and the expected leg length needs to be continuously led by adding a proper transition process. When the transition process is added, the foot end point is required to be contacted with the ground, and under the condition that the pitch angle of the body is not changed, the constraint conditions for the change of the leg length can be obtained as follows:

from formula (7):

$m\stackrel{\&CenterDot;\&CenterDot;}{z}=f_{z1}+f_{z2}-mg>-mg---\left(25\right)$

the following steps are provided:

$\begin{array}{c}{a}_{L1}=\stackrel{\&CenterDot;\&CenterDot;}{z}/\sin \left({\mathrm{\&alpha;}}_3\right)\\ a_{L2}=\stackrel{\&CenterDot;\&CenterDot;}{z}/\sin \left({\mathrm{\&alpha;}}_4\right)\end{array}---\left(26\right)$

therefore, the approximate constraint conditions for the leg length acceleration can be obtained as follows:

$\begin{array}{c}{a}_{L1}>-g/\sin \left({\mathrm{\&alpha;}}_3\right)\\ a_{L2}>-g/\sin \left({\mathrm{\&alpha;}}_4\right)\end{array}---\left(27\right)$

thus, an acceleration-limited second-order transition can be added to the leg length as follows:

$\begin{array}{c}{a^{\&prime;}}_{L1}={\mathrm{Kp}}_{a1}\left(L_1-L_{1r}\right)+{\mathrm{Kd}}_{a1}{\stackrel{\&CenterDot;}{L}}_{1r}\\ {a^{\&prime;}}_{L2}={\mathrm{Kp}}_{a2}\left(L_2-L_{2r}\right)+{\mathrm{Kd}}_{a2}{\stackrel{\&CenterDot;}{L}}_{2r}\end{array}---\left(28\right)$

equation (29) becomes after adding a slice to the acceleration:

$\begin{array}{c}{a}_{L1}=\left\{\begin{array}{cc}{a^{\&prime;}}_{L1}& {a^{\&prime;}}_{L1}>a_{L1\_\min },a_{L1\_\min }=-k_{2}g/\sin \left({\mathrm{\&alpha;}}_3\right)\\ a_{L1\_\min }& {a^{\&prime;}}_{L1}\&le;a_{L1\_\min },a_{L1\_\min }=-k_{2}g/\sin \left({\mathrm{\&alpha;}}_3\right)\end{array}\right.\\ a_{L2}=\left\{\begin{array}{cc}{a^{\&prime;}}_{L2}& {a^{\&prime;}}_{L2}>a_{L2\_\min },a_{L2\_\min }=-k_{2}g/\sin \left({\mathrm{\&alpha;}}_4\right)\\ a_{L2\_\min }& {a^{\&prime;}}_{L2}>a_{L2\_\min },a_{L2\_\min }=-k_{2}g/\sin \left({\mathrm{\&alpha;}}_4\right)\end{array}\right.\end{array}---\left(29\right)$

the leg length is desirably given by:

$\begin{array}{c}{L}_{1d}=L_{10}+\underset{0~t}{\&Integral;\&Integral;}{a}_{L1}d\mathrm{\&tau;}d\mathrm{\&tau;}\\ L_{2d}=L_{20}+\underset{0~t}{\&Integral;\&Integral;}{a}_{L2}d\mathrm{\&tau;}d\mathrm{\&tau;}\end{array}---\left(30\right)$

wherein L is₁₀,L₂₀The initial leg lengths of the two legs, respectively.

The overall control block diagram of the planar virtual telescopic leg model can be obtained as described above, and is shown in fig. 3.

3) Designing a seven-connecting-rod model joint position servo controller;

the controller with the seven-connecting-rod structure is designed similarly to a virtual telescopic leg model controller, and a force control mode is required to be adopted for hip joints, and a position servo control mode is required to be adopted for knee joints and ankle joints. As the single leg comprises three joints in the front direction, one more degree of freedom is provided for leg length control, and the purpose of adding three degrees of freedom is mainly to consider the driving performance of the hydraulic cylinder, enhance the driving capability of the single leg or improve the system performance. Therefore, to simplify the computational process when performing inverse kinematics solution as follows, joint constraints are added as follows:

θ₄＝θ₅,θ₁＝θ₂(31)

this is easily obtained from the left diagram in FIG. 2 and equation (31):

${\mathrm{\&theta;}}_{4d}={\mathrm{\&theta;}}_{5d}=acos\left(\frac{l_2^2+4l_1^2-L_{2d}^2}{4l_1{l}_2}\right)---\left(32\right)$

${\mathrm{\&theta;}}_{1d}={\mathrm{\&theta;}}_{2d}=acos\left(\frac{l_2^2+4l_1^2-L_{1d}^2}{4l_1{l}_2}\right)---\left(33\right)$

the knee joint and ankle joint angles are controlled by a servo controller:

$\mathrm{\&tau;}={\mathrm{Kp}}_{\mathrm{\&theta;}}({\mathrm{\&theta;}}_d-\mathrm{\&theta;})+{\mathrm{Kv}}_{\mathrm{\&theta;}}({\stackrel{\&CenterDot;}{\mathrm{\&theta;}}}_d-\stackrel{\&CenterDot;}{\mathrm{\&theta;}})---\left(34\right)$

wherein τ is joint drive torque. By adjusting the controller parameter Kp_θ,Kv_θThe tracking control of the joint angle can be realized.

The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above-mentioned embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (4)

1. A method for hybrid control of the phase force position of the two-foot support of a four-foot robot is characterized by comprising the following steps:

s1: projecting the overall motion of the robot onto a radial plane and a normal plane; the radial plane is a plane which supports the feet and is vertical to the horizontal plane; the normal plane is a plane passing through the center of mass of the body and perpendicular to the radial plane;

s2: establishing a control model; the motion of the robot along the radial plane is simplified into a plane seven-connecting-rod model, and the motion on the normal plane is simplified into a linear inverted pendulum model; then simplifying the plane seven-connecting-rod model into a plane virtual telescopic leg model; the control targets of the plane virtual telescopic leg model are the height of a mass center, a pitch angle of the body and horizontal displacement;

s3: performing mixing control according to a control model; establishing a kinematic equation of a plane virtual telescopic leg model and a dynamic equation thereof by a Newton-Euler method, and adopting a force control mode for the horizontal displacement of the virtual telescopic leg plane model by a position servo control method for the height of a mass center and a body pitch angle.

2. The hybrid control method for biped support phase force and position of the quadruped robot as claimed in claim 1, wherein in step S3, the height of the center of mass and the pitch angle of the body are controlled by the leg length, and the horizontal displacement of the body is controlled by controlling the horizontal foot end force, so as to realize the approximate decoupling control of the posture of the body.

3. The biped robot biped support phase force position hybrid control method according to claim 2, characterized in that a classical PID controller is adopted as a force outer loop controller, so that the support leg foot end contact force accurately tracks the expected force.

4. The biped robot biped support phase force position hybrid control method according to claim 2, characterized in that the knee joint and the ankle joint are controlled by a position servo controller to realize the tracking of the joint angle.