A stable motion planning method for bipedal robots based on motion state classification

Abstract

The present invention provides a method for stable motion planning of a bipedal robot based on motion state classification, the method comprising collecting data through the robot body sensor to determine the position and speed of the body center of mass; calculating the limit speed and stability coefficient according to the motion state of the body center of mass; performing motion state classification and stability analysis according to the robot walking speed and the state of the combined external force; constructing a foothold gait planning to determine the parameters of the next swing leg step length and motion time, so as to plan the stability coefficient within a stable range, and then repeating the above steps in a cycle to achieve the purpose of stable motion. The method classifies the motion state of the bipedal robot by constructing a stability coefficient, can consider the influence of external interference forces on the motion law of the robot, and plans gait parameters to achieve stable motion, thereby improving the stability ability of the motion planning of the bipedal robot.

Description

Technical Field

[0001] The present invention relates to the technical field of legged robots, and more particularly to a stable motion planning method for a bipedal robot based on motion state classification. Background Art

[0002] Compared with wheeled and tracked ground mobile robots, legged robots have extremely strong movement flexibility and terrain adaptability due to their unique discrete foot movement rules. However, due to the small number of supporting feet, the movement stability of bipedal robots is often difficult to determine in practical applications. Movement stability is to study the impact of disturbances on the movement of the system, so as to establish criteria for judging whether the movement state is stable. The movement stability criteria of legged robots are classified into static and dynamic movement stability criteria based on the walking state. The dynamic stability criteria of bipedal robots are generally based on the construction of ground points with physical significance, called ground reference points, to define the stability margin and calculate the value. The more commonly used ones are zero moment point ZMP, center of pressure COP, foot end rotation index FRI, center of mass moment pivot CMP, etc., while the ground reference points constructed based on the force conditions only consider the reaction force of the ground, and do not consider the robot's movement state and the interference of impact forces.

[0003] In summary, the stable motion planning of legged robots currently has poor practicality and universality, and does not take into account the robot's motion state and the interference of impact forces. Summary of the invention

[0004] Aiming at the problems that the existing stable motion planning methods for bipedal robots have poor practicality and universality in dynamic motion conditions and do not consider the robot's motion state and the interference of impact forces, a stable motion planning method for bipedal robots based on motion state classification is proposed.

[0005] The present invention proposes a method for stable motion planning of a biped robot based on motion state classification, comprising the following steps:

[0006] Step S1, establishing a simplified model of a biped robot;

[0007] Step S2, obtaining the center of mass motion state of the robot at the current time t0, wherein the center of mass motion state includes the position P(t0) and speed of the center of mass

[0008] Step S3, based on the simplified model of the biped robot and the motion state of the center of mass, construct the dynamic differential equation of the robot, and calculate the velocity change of the center of mass of the robot under the influence of external force; further calculate the instantaneous capture point position according to the velocity change of the center of mass, and take the distance between the instantaneous capture point position and the position of the end of the supporting leg of the robot in the current stepping motion as the stability coefficient ΔP(t0) at the current time t0, and calculate the limit speed of the biped robot under stable motion

[0009] Step S4, according to the stability coefficient ΔP(t0) and the limit speed The size of the motion state is classified and the stability is analyzed;

[0010] Step S5, planning the next gait parameters according to the motion state of the robot, wherein the gait parameters include the step length l and the step time Δt;

[0011] Step S6, according to the planned gait parameters, control the robot to take a step, and repeat steps S2-S6 after landing to achieve stable movement.

[0012] Furthermore, the simplified model of the bipedal robot is a linear inverted pendulum model.

[0013] Furthermore, in step S2, the center of mass motion state of the robot at the current moment is collected by the body sensor of the robot.

[0014] Furthermore, in step S3, the dynamic differential equation of the robot is constructed as follows:

[0015]

[0016] in, is the acceleration of the robot's center of mass at time t, g is the acceleration due to gravity, z is the height of the center of mass, P(t) is the position of the robot's center of mass at time t, m is the weight of the robot, and F(t) is the resultant force of the generalized force acting on the robot at time t;

[0017] The speed change of the center of mass of the computing robot under the influence of external force is specifically:

[0018] Solve the constructed dynamic differential equation, P(t), The initial values ​​are the center of mass position P(t0) and center of mass velocity P(t0) at the current time t0. That is, the change of the center of mass motion speed of the robot under the influence of external force is obtained:

[0019]

[0020] Among them, t0 represents the current time, P(t0), are the position and velocity of the center of mass at the current moment obtained in step S1, is a fixed parameter of the robot, and m is the total mass of the robot.

[0021] Furthermore, in step S3, the instantaneous capture point position is further calculated according to the velocity change of the center of mass as follows:

[0022]

[0023] Among them, P icp (t0) is the instantaneous capture point position at the current time t0, P gcom (t0) is the ground projection position of the center of mass at the current time t0, is the velocity of the robot’s center of mass at the current time t0, and ω is a fixed parameter of the robot.

[0024] Furthermore, in step S3, the calculation of the limit speed of the bipedal robot under stable motion is specifically:

[0025] The limit speed refers to the maximum speed of the robot's stable movement without external force to reduce the speed. If the limit speed is exceeded, the movement will evolve into the robot overturning. The limit speed is calculated based on the step length l0 and the step time Δt0 of the previous step.

[0026]

[0027] in, are the fixed parameters of the robot, g is the acceleration of gravity, z is the height of the center of mass, Δt0 is the step time of the previous step, l0 is the step length of the previous step, and P(t0) is the position of the robot's center of mass at the current time t0.

[0028] Furthermore, in step S4, the motion state classification and stability analysis are specifically as follows:

[0029] The stability coefficient ΔP(t0) at the current time t0 and the maximum step length l max , Limit speed By comparison, the maximum step length l max It is a fixed parameter determined by the mechanical parameters of the robot. According to the comparison results, the robot's motion state is divided into the following four categories and the stability is analyzed:

[0030] 1) If ΔP(t0) = 0, the bipedal robot is in a state of standing still or stationary, which is a stable state;

[0031] 2) If 0<ΔP(t0)≤l max, the biped robot is in the first motion state, in which it can enter a stationary state or maintain the first motion state by changing the gait parameters of the next step, and will enter the second motion state if the gait parameters of the next step are not changed;

[0032] 3) If Where ω is a fixed parameter of the robot, P gcom (t0) is the ground projection position of the center of mass at the current time t0, then the bipedal robot is in the second motion state, in which the bipedal robot needs to continue walking to maintain stability. If the gait parameters of the next step are not changed, it will enter an unstable state;

[0033] 4) If The bipedal robot is then in an unstable state, in which stability can only be restored by external forces, and stable motion planning cannot change this state.

[0034] Furthermore, in step S5, the gait parameters of the next step are planned according to the motion state of the robot:

[0035] If the bipedal robot is in a standing or static state and needs to maintain the original state, the step length is selected as l=0, and the step time is selected as any value; if it needs to enter the first motion state from the standing or static state, the step length is selected as l<l max , the step time selection Where Δt min The shortest support time that the robot can achieve when taking one step, which is determined by the robot's own mechanical properties. is the velocity of the center of mass under the influence of external forces;

[0036] If the biped robot is in the first motion state and needs to enter a static state, the step length is selected as l = || P icp (t0+Δt)-P foot (t0+Δt)||, where (t0+Δt) represents the moment when the next swing leg lands, P foot is the position of the end of the supporting leg and is a constant value during the current step. Δt is the step time. Select the shortest support time Δt that can be achieved. min , ||.|| represents the distance between two position points; if the current first motion state needs to be maintained, the step length l is selected in the range of l<||P icp (t0+Δt)-P foot (t0+Δt)||, where Δt is the step time, select

[0037] If the biped robot is in the second motion state, the step length is selected as the maximum step length l max; The step time selection robot step l max The shortest support time Δt that can be achieved min , try to avoid entering an unstable state.

[0038] Furthermore, in step S6, the stable motion means that the stability coefficient ΔP does not exceed Where ω is a fixed parameter of the robot, P gcom (t0) is the ground projection position of the center of mass at the current time t0.

[0039] Furthermore, the step length is the distance between the end of the swing leg of the robot and the end of the supporting leg.

[0040] The beneficial effects of the present invention are:

[0041] The present invention proposes a stable motion planning method for a biped robot based on motion state classification, which can take into account the motion state of the robot and the interference of impact force, and provide motion state component basis and stable motion planning method for biped robots with different parameter specifications.

[0042] The present invention derives the limiting speed of the bipedal robot through a simplified dynamic model of the legs and the kinematics of the center of mass of the bipedal robot, and summarizes the dynamic motion conditions into several motion states through the motion state of the robot's center of mass and analyzes the stability. The current motion state of the robot can be clearly and intuitively judged, and its corresponding motion stability is supported by this theory, so as to plan the subsequent corresponding motion gait parameters and prevent the robot from being in a state of imminent overturning, that is, an unstable motion state. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 The figure is a flow chart of a stable motion planning method for a biped robot based on motion state classification.

[0044] Figure 2 Schematic diagram of a simplified model of a bipedal robot.

[0045] Figure 3 Schematic diagram of the motion state classification of a bipedal robot.

[0046] Figure 4 Schematic diagram of the stable motion planning method for a bipedal robot. DETAILED DESCRIPTION

[0047] The present invention will be further described below in conjunction with the accompanying drawings:

[0048] Figure 1 The steps and the relationship between the components in the embodiment are described.

[0049] Figure 2The simplified model of the bipedal robot shown in the figure simplifies the bipedal robot into a simplified model of a spherical center of mass and two massless retractable legs. The characteristics and laws of its foot-type walking motion are as follows: the foot end is in point contact with the ground, and the two retractable legs always keep the center of mass on the same horizontal plane during the motion, so that the three-dimensional motion is decoupled into the two-dimensional plane motion, so that the directions of all derived vector physical quantities are in the horizontal plane direction, and the external interference force and impact force are taken into account. This method is established on this widely used simplified model. If it is extended to the actual robot motion, this motion law can also be achieved through joint motion.

[0050] Figure 3 The specific meaning of the motion state classification shown in the figure is that according to the stability coefficient and the capture point position analysis, the four motion states are: ΔP≈0, the bipedal robot is in the state of marking time; ΔP≤l max , the biped robot is in the shortest support time Δt min Walk a maximum step length movement distance l max It is capable of stable movement; The bipedal robot can be stable when in a state of continuous walking motion; A bipedal robot can walk stably for a short period of time, but will eventually fall if it is not subject to restoring external forces.

[0051] The stability coefficient at the current moment is determined by the swing leg step length l and the swing leg step time Δt parameters, while the maximum stability coefficient is determined by the maximum value l that the swing leg step length l can achieve. max , the minimum value Δt that the swing leg step time can achieve min The maximum values ​​of the two are determined by the specific specifications and motion performance of the bipedal robot in specific implementation.

[0052] Figure 4 The specific meaning of the stable motion planning method shown in the figure is that through the above stability coefficient calculation and motion state classification, the motion state quantity and dynamic analysis quantity of the robot's center of mass at time t0 can be obtained: the center of mass position P(t0), the center of mass speed Center of mass position P(t0), current foot position P foot (t0), the generalized force condition F(t), perform capture point position analysis and motion state classification, and plan the gait parameters of the next movement, the swing leg step length l and the swing leg step time Δt, and obtain the motion state quantity and dynamic analysis quantity of the robot's center of mass at t0+Δt after the next movement, and repeat this cycle.

[0053] It should be noted that P foot (t0+Δt) is the position where the foot end instantly falls at the moment t0+Δt, and its specific robot implementation is planned by the specific motion trajectory.

[0054] The specific implementation steps of the present invention are as follows:

[0055] A method for planning stable motion of a bipedal robot based on motion state classification, wherein the bipedal robot performs stability analysis through motion state classification and dynamically adjusts gait parameters to achieve stable motion, and the planning method comprises:

[0056] Step 1: Collect data through the robot body sensor to determine the body center of mass position P and speed and acceleration The sensors include a six-dimensional IMU sensor, a six-dimensional force sensor and a lidar sensor.

[0057] Step 2: Set the instantaneous capture point position P icp The position of the end of the supporting leg P foot The distance between them is ΔP=||P icp -P foot || is used as the stability coefficient of the biped robot during movement, and the maximum movement speed of the biped robot under stable movement conditions is derived.

[0058] Step 3: According to the robot's walking speed The motion state classification and stability analysis are carried out based on the state F(t) of the resultant external force.

[0059] Among them, walking speed is the instantaneous velocity of the center of mass of the bipedal robot, and F(t) is the resultant of the generalized forces acting on the center of mass of the legged robot, which can be calculated from the center of mass acceleration collected by the sensor.

[0060] Step 4: Classify the legs of the footed robot into swing legs and support legs, determine the step length l of the swing leg, and the support time of the support leg, that is, the time Δt required for the swing leg to take a step.

[0061] Step 5: Construct the foothold gait planning and determine the next swing leg foothold point P foot The two parameters of movement time Δt are used to plan the stability coefficient ΔP within the stable range, and the above steps are repeated in a cycle to achieve the purpose of stable movement.

[0062] It should be noted that the bipedal robot ignores the leg mass and structure and simplifies it. The center of mass motion is derived based on the linear inverted pendulum model. The linear inverted pendulum stabilizes the center of mass motion on the same horizontal plane without changing the center of mass height, thereby decoupling the center of mass motion of the bipedal robot. The maximum motion speed According to its inference, its kinetic equation is:

[0063]

[0064] Where g is the acceleration due to gravity and z is the height of the center of mass. Solving its dynamic differential equation, we can get the law of center of mass motion:

[0065]

[0066]

[0067] in P(t), P(t0), They are respectively the center of mass displacement, the initial value of the center of mass displacement, the center of mass velocity and the initial value of the center of mass velocity, all of which are horizontal vectors.

[0068] Define dimensionless energy and set it to 0:

[0069]

[0070] This means that if the robot can be stable and stationary under this motion law, the relationship between the center of mass velocity and the center of mass displacement can be solved by the above formula to obtain the instantaneous capture point position P icp :

[0071]

[0072] Among them, the capture point position P icp Indicates that the bipedal robot can reach a stable static state after only one step, and the landing point of the step needs to coincide with this point at the current moment. On the one hand, the capture point position P icp As time changes, on the other hand, the law of leg motion of the biped robot is to continuously determine the discrete footfall points. The characteristic of stable motion is that the discrete leg motion needs to chase the capture point position P icp .

[0073] Capture point position P icp The motion law of the instantaneous capture point can be derived by derivation to obtain the motion speed and law of the instantaneous capture point:

[0074]

[0075] Between adjacent walking steps, the instantaneous capture point movement speed and its law can be solved by solving the above differential equation to obtain:

[0076]

[0077] Let the instantaneous capture point position P icp and the ground projection position P of the centroid gcom The distance between them is ΔP=||P icp -P gcom || As the stability coefficient of the biped robot during movement, let the maximum value of the stability coefficient ΔP be d max .

[0078] The maximum stability coefficient d can be derived from the above formula max , assuming d max The initial time is t0, and the d at the previous step is max-1 The time is t0+Δt, d max to d max-1 The process is a step time Δt min The maximum step length l max , there is the following relationship:

[0079] P icp (t0+Δt)=P icp (t0)e ωΔt ≤d max-1 +l max

[0080] According to the above formula, we can get d max The recursive relationship is:

[0081] P icp (t0)e -ωΔt =P icp (t0)≤(d max-1 +l max ) -ωΔt =d max

[0082] Taking the difference, we can get the geometric progression relationship:

[0083] d max -d max-1 =e -ωΔt (d max-1 -d max-2 )

[0084] d max-1 -d max-2 =e -ωΔt (d max-2 -d max-3 )

[0085] …

[0086] d2-d1=e -ωΔt (d1-d0)

[0087] Sum the above geometric series and get the maximum stability coefficient d according to the summation formula: max :

[0088]

[0089]

[0090] It can be seen that the stability coefficient is determined by the walking motion's step time Δt and the corresponding step length l.

[0091] Furthermore, the maximum speed in step 2 is derived from the maximum value of the stability coefficient ΔP. The concept is that for each legged robot with certain parameters, there is an exact limit speed. If the limit speed is exceeded and there is no external force to reduce the speed, the robot will eventually overturn after the motion evolution. According to the instantaneous capture point speed and the maximum step length l of the swing leg max , the shortest support time of the supporting leg Δt min , the maximum speed that can make the fuselage stationary through stable motion planning can be iteratively derived

[0092]

[0093] Furthermore, in step 3, the motion state classification and instantaneous capture point position P icp and the ground projection position P of the centroid gcom It changes with the movement process, according to the instantaneous capture point position P icp and the ground projection position P of the centroid gcom The distance between them is the stability coefficient ΔP and the maximum step length l of the swinging leg max The comparison is divided into:

[0094]

[0095] It should be noted that the four motion states are:

[0096] ΔP≈0, the biped robot is in a state of standing still or stationary, which is a stable state; the stable state means that the position of the instantaneous capture point always falls within the support surface surrounded by the foot ends of the biped robot;

[0097] ΔP≤l max , the biped robot is in the first motion state, in which, if the gait parameters are not adjusted through stable motion planning, it will evolve into the second motion state;

[0098] The biped robot is in the second motion state and needs to perform continuous walking motion through stable motion planning to maintain stability, otherwise it will evolve into an unstable state;

[0099] The bipedal robot is in an unstable state. It can move stably when walking for a short period of time, but it will eventually fall if it is not subject to restoring external forces.

[0100] Furthermore, the motion state analysis in step 3 takes into account the change in the velocity of the center of mass caused by the combined force of the external disturbance and the impact generalized force. Its dynamic differential equation can be expressed as:

[0101]

[0102] The speed analysis results are:

[0103]

[0104] Here m is the total mass of the robot. Substitute this function into the above center of mass motion equation for analysis.

[0105] Furthermore, in step 4, the swing leg step length l and the support leg support time, i.e., the swing leg step time Δt, are determined by the stability coefficient and the motion state component, and have maximum and minimum values, respectively, i.e., the swing leg step length l has a maximum value l max The support time of the supporting leg, i.e. the stepping time of the swinging leg, has a minimum value Δt min The specific parameters are determined according to the specific specifications and dynamic performance of the robot.

[0106] The selection of the swing leg step length l and the swing leg step time Δt are as follows:

[0107] If the bipedal robot is in a standing or static state and needs to maintain the original state, the step length is selected as l=0, and the step time is selected as any value; if it needs to enter the first motion state from the standing or static state, the step length is selected as l<l max , the step time Δt is selected Where Δt min The shortest support time that the robot can achieve when taking one step, which is determined by the robot's own mechanical properties. is the velocity of the center of mass under the influence of external forces;

[0108] If the biped robot is in the first motion state and needs to enter a static state, the step length is selected as l = ΔP(t0+Δt) = ||P icp (t0+Δt)-P foot (t0+Δt)||, where (t0+Δt) represents the moment when the next swing leg lands, P foot is the position of the end of the supporting leg and is a constant value during the current step. Δt is the step time. Select the shortest support time Δt that can be achieved. min , ||.|| represents the distance between two position points; if the current first motion state needs to be maintained, the step length l is selected in the range of l<||P icp (t0+Δt)-P foot (t0+Δt)||, where Δt is the step time, select

[0109] If the biped robot is in the second motion state, the step length is selected as the maximum step length l max; The step time selection robot step l max The shortest support time Δt that can be achieved min .

[0110] Furthermore, the stability range in step 5 refers to the support area enclosed by the foot of the bipedal robot. Due to the influence of the body specifications of the legged robot, the stability ranges of different legged robots are different. If the bipedal robot is in surface contact with the ground, P icp If the biped robot is in point contact with the ground, P icp Need to be with P foot Coincidence is a stable state;

[0111] Further, repeating the above steps in step 5 means determining the speed analysis result according to the motion state classification and stability analysis, dynamically adjusting the swing leg step length l and the swing leg step time Δt parameters, and l = P foot (t+Δt)-P foot (t).

[0112] The above embodiments are only preferred embodiments of the present invention and are not limitations of the technical solutions of the present invention. Any simple modifications, supplementary modifications and equivalent substitutions made within the scope of the principles of the present invention shall be deemed to fall within the scope of protection of the patent of the present invention.

Claims

1. A stable motion planning method for a biped robot based on motion state classification, characterized in that: The following steps are involved: Step S1, establishing a simplified model of a biped robot; Step S2, obtaining the center of mass motion state of the robot at the current time t0, wherein the center of mass motion state includes the position P(t0) of the center of mass, the speed and the total external force F(t) acting on the robot; Step S3, based on the simplified model of the biped robot and the motion state of the center of mass, construct the robot's dynamic differential equation, and calculate the velocity change of the center of mass of the robot under the influence of external force; further calculate the instantaneous capture point position according to the velocity change of the center of mass, and take the distance between the instantaneous capture point position and the position of the end of the supporting leg of the robot in the current stepping motion as the stability coefficient P(t0) at the current time t0, and calculate the limit speed of the biped robot under stable motion Step S4, according to the stability coefficient ΔP(t0) and the limit speed The size of the motion state is classified and the stability is analyzed; Step S5, planning the next gait parameters according to the motion state of the robot, wherein the gait parameters include the step length l and the step time Δt; Step S6, controlling the robot to take steps according to the planned gait parameters, and repeating steps S2-S6 after landing to achieve stable movement; In step S3, the calculation of the limit speed of the bipedal robot under stable motion is specifically: The limit speed refers to the maximum speed of the robot's stable movement without external force to reduce the speed. If the limit speed is exceeded, the movement will evolve into the robot overturning. The limit speed is calculated based on the step length l0 and the step time Δt0 of the previous step. in, are the fixed parameters of the robot, g is the acceleration of gravity, z is the height of the center of mass, Δt0 is the step time of the previous step, l0 is the step length of the previous step, and P(t0) is the position of the center of mass of the robot at the current time t0; In step S4, the motion state classification and stability analysis are specifically as follows: The stability coefficient ΔP(t0) at the current time t0 and the maximum step length l max , Limit speed By comparison, the maximum step length l max It is a fixed parameter determined by the mechanical parameters of the robot. According to the comparison results, the robot's motion state is divided into the following four categories and the stability is analyzed: 1) If ΔP(t0) = 0, the bipedal robot is in a state of standing still or stationary, which is a stable state; 2) If 0<ΔP(t0)≤l max , the biped robot is in the first motion state, in which it can enter a stationary state or maintain the first motion state by changing the gait parameters of the next step, and will enter the second motion state if the gait parameters of the next step are not changed; 3) If Where ω is a fixed parameter of the robot, P gcom (t0) is the ground projection position of the center of mass at the current time t0, then the bipedal robot is in the second motion state, in which the bipedal robot needs to continue walking to maintain stability. If the gait parameters of the next step are not changed, it will enter an unstable state; 4) If Then the biped robot is in an unstable state, in which it can only be restored to stability by external forces, and stable motion planning cannot change this state; In step S5, the gait parameters for the next step are specifically planned according to the motion state of the robot: If the bipedal robot is in a standing or static state and needs to maintain the original state, the step length is selected as l=0, and the step time is selected as any value; if it needs to enter the first motion state from the standing or static state, the step length is selected as l<l max , the step time selection Where Δt min The shortest support time that the robot can achieve when taking one step, which is determined by the robot's own mechanical properties. is the velocity of the center of mass under the influence of external forces; If the biped robot is in the first motion state and needs to enter a static state, the step length is selected as l = || P icp (t0+Δt)-P foot (t0+Δt)||, where (t0+Δt) represents the moment when the next swing leg lands, P foot is the position of the end of the supporting leg and is a constant value during the current step. Δt is the step time. Select the shortest support time Δt that can be achieved. min , ||·|| represents the distance between two points; If the current first motion state needs to be maintained, the step length l is selected within the range of l<||P icp (t0+Δt)-P foot (t0+Δt)||, where Δt is the step time, select If the biped robot is in the second motion state, the step length is selected as the maximum step length l max ; The step time selection robot step l max The shortest support time Δt that can be achieved min , to avoid entering an unstable state.

2. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: The simplified model of the biped robot is a linear inverted pendulum model.

3. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: In step S2, the center of mass motion state of the robot at the current moment is collected by the body sensor of the robot.

4. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: In step S3, the dynamic differential equation of the robot is constructed as follows: in, is the acceleration of the robot's center of mass at time t, g is the acceleration due to gravity, z is the height of the center of mass, P(t) is the position of the robot's center of mass at time t, m is the weight of the robot, and F(t) is the resultant external force on the robot at time t, where the resultant external force refers to the generalized force resultant; The speed change of the center of mass of the computing robot under the influence of external force is specifically: Solve the constructed dynamic differential equation, P(t), The initial values ​​are the center of mass position P(t0) and center of mass velocity P(t0) at the current time t0. That is, the change of the center of mass motion speed of the robot under the influence of external force is obtained: Among them, t0 represents the current time, P(t0), are the position and velocity of the center of mass at the current moment obtained in step S1, is a fixed parameter of the robot, and m is the total mass of the robot.

5. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: In step S3, the instantaneous capture point position is further calculated according to the velocity change of the center of mass: Among them, P icp (t0) is the instantaneous capture point position at the current time t0, P gcom (t0) is the ground projection position of the center of mass at the current time t0, is the velocity of the robot’s center of mass at the current time t0, and ω is a fixed parameter of the robot.

6. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: In step S6, the stable motion means that the stability coefficient ΔP does not exceed Where ω is a fixed parameter of the robot, P gcom (t0) is the ground projection position of the center of mass at the current time t0.

7. The method for stable motion planning of a biped robot based on motion state classification according to claim 1, characterized in that: The step length is the distance between the end of the swing leg of the robot and the end of the supporting leg.

Images