JP5463991B2 - Biped walking robot - Google Patents

Description

  The present invention relates to a biped robot.

  Data for making the biped walking robot walk is called gait data. Gait data is usually represented by time-series data of a center of gravity (or trunk) target position and a foot target position. A target angle of each joint of the leg is calculated from the data at each time point of the gait data, and each joint is controlled so as to follow the target angle. In this specification, “time-series data” may be referred to as “trajectory”. The gait data is represented by the center of gravity and the target trajectory of the foot.

  A biped robot would fall over without proper gait data. Whether or not the robot falls is a matter that can be confirmed only after the robot actually walks. Therefore, in this specification, for the sake of convenience, it can be predicted that the robot can walk without falling down. ". Further, gait data that can realize a theoretically stable walking is referred to as “stable gait data” for convenience. In the following, for the sake of simplicity, the “biped walking robot” is simply referred to as “robot”.

  One method for generating stable gait data is to generate a center of gravity trajectory (time-series data) of the robot based on the ZMP equation and obtain a target angle for each leg joint from the center of gravity trajectory. Whether or not the robot falls is because the center of gravity trajectory of the robot is the dominant factor. Many biped walking robots generate gait data based on the ZMP equation.

  Each foot should just take off and land as planned in the gait data, but may not actually move as planned in the gait data due to disturbance or the like. Here, the “disturbance” is, for example, a contact force received from another object when the biped walking robot comes into contact with another object, or an inclination of the entire biped walking robot caused by unexpected undulations on the road surface. is there.

  Legs may take off due to disturbance. In such a case, the other part (for example, heel) usually protrudes while the part (for example, toe) of the foot is grounded. That is, the foot tilts with respect to the road surface unexpectedly. When the foot is inclined with respect to the road surface, it becomes impossible to apply a moment due to the floor reaction force to the trunk through the foot, and the possibility of falling is increased. Therefore, it is necessary to avoid the foot tilting unscheduled. For example, Patent Literature 1 discloses a technique for correcting the inclination of the foot with respect to the road surface. Hereinafter, the inclination of the foot with respect to the road surface is simply referred to as the inclination of the foot.

JP 2008-188684 A

  When the foot starts to tilt, the technology of Patent Document 1 is to control the trunk to return the tilt of the trunk (referred to as inverted control in Patent Document 1), and the tilted foot again faces the road surface. The control to make contact (referred to as copying control in Patent Document 1) is skillfully combined. Here, the inversion control acts in a direction in which the inclination of the foot becomes larger, and the copying control acts to reduce the inclination of the foot. That is, the inversion control and the copying control are in a trade-off relationship. The technique of Patent Document 1 is a technique for skillfully adjusting this trade-off.

  However, since the technique of Patent Document 1 is a technique for adjusting a trade-off, neither the inversion control nor the copying control is optimal. The present invention is based on a technical idea that is completely different from the technical idea disclosed in Patent Document 1, and does not require a concept of trade-off, and a technique for restoring the inclination of the foot while ensuring walking stability. provide.

  The inversion control in Patent Document 1 is based on a technical idea of returning the trunk inclination to the original target angle. The novel technique disclosed in the present specification does not return the trunk inclination to the target angle, but the acceleration of the center of gravity in the direction in which the trunk tilts (that is, the direction in which the inclination of the foot increases) By taking in the target angle of the joint, the inclination of the foot is restored by the reaction force of the inertial force that accelerates the trunk. Walking stability is ensured by recreating the center of gravity trajectory using the ZMP equation so that walking can be stably performed at the acceleration. In other words, this method suppresses the inclination of the foot by controlling the joints of the legs so as to accelerate the trunk in the direction in which the trunk tilts during the most recent period (for example, for several tens of milliseconds). At the same time, the leg joints are controlled so that walking is stable over a long-term span (for example, until the current free leg lands). As described above, the present invention determines the target joint angle of the leg so as to suppress the inclination of the foot in the short-term in the time axis direction and stabilize the walking in the long-term. With such a technical idea, both the suppression of foot inclination and walking stability are achieved without introducing the concept of trade-off. The present invention does not explicitly separate the period for suppressing the inclination of the foot and the period for stabilizing the walking on the time axis. In the ZMP equation that defines the trajectory of the center of gravity, there are two parameters: the parameter that governs the trajectory of the most recent center of gravity (acceleration of the center of gravity at the current time) and the parameter that governs the entire center of gravity trajectory after a predetermined time (the center of gravity speed after a predetermined time). By paying attention, the center-of-gravity trajectory is generated so that the walking stability of the entire trajectory is maintained over the long term so that the inclination of the foot is suppressed in the short term.

The novel biped robot disclosed in this specification includes a plurality of distance sensors on each leg. The distance sensors are arranged at least two places before and after the sole, and measure the distance to the road surface. The controller calculates time series data of the target angles of the joints of the legs from the target trajectory of the legs and controls the joints of the legs so as to follow the calculated time series data. When the controller detects an unplanned inclination of the foot, the controller recalculates time series data of the target joint angle. The recalculation procedure is as follows. Center-of-gravity acceleration calculation step: When the distance sensor detects that the foot has tilted with respect to the road surface outside the schedule (other than the scheduled takeoff timing on the foot trajectory), the acceleration of the center of gravity at that time is calculated. calculate. Center of gravity trajectory determination step: The center of gravity is obtained by solving the ZMP equation using the velocity of the center of gravity realized by the calculated acceleration of the center of gravity as an initial value and having the velocity of the center of gravity determined in advance at the end of a predetermined period. The trajectory of is calculated. Joint target angle calculation step: recalculate time-series data of target angles of the leg joints from the calculated trajectory of the center of gravity. Then, the controller controls each joint of the leg so as to follow the time series data of the target angle recalculated in the joint target angle calculation step.

  Here, “when the foot tilts with respect to the road surface unexpectedly” means that the foot tilts with respect to the road surface in addition to the scheduled takeoff timing on the gait data (target foot trajectory). It means "when". When taking off as planned on the gait data, that is, when the foot tilts as planned, each joint of the leg is controlled according to the original gait data without any special processing. It is because it is good.

  The feature of the process executed by the controller is particularly the center-of-gravity trajectory determination step. The trajectory of the center of gravity for realizing stable walking is given by the ZMP equation. The ZMP equation is an equation that gives the acceleration of the center of gravity position. As will be described in the section of the embodiment of the invention, the ZMP equation can uniquely determine the center of gravity trajectory when the initial speed and the terminal speed of the center of gravity trajectory are given as boundary conditions for the current period of interest. In the center-of-gravity trajectory determination step, the current center-of-gravity speed determined by the center-of-gravity acceleration when the foot is tilted is given as an initial value. As a result, the acceleration of the center of gravity that was not scheduled is taken into a new center of gravity trajectory, and the center of gravity trajectory that realizes the acceleration is determined. By controlling each joint of the leg so as to realize a new acceleration, the inclination of the foot is suppressed by the reaction force of the force that accelerates the center of gravity. By giving the initial value as described above, the center of gravity trajectory that accelerates the center of gravity in the direction in which the trunk tilts is obtained in the most recent period (for example, for several tens of milliseconds), and the leg joint is based on such center of gravity trajectory. The inclination of the foot can be suppressed by controlling. On the other hand, at the end of the center of gravity trajectory, an appropriate end speed that does not become excessive is given as a boundary condition. This terminal speed condition ensures that the speed does not become excessive at the end of the center of gravity trajectory. It is avoided that the center of gravity trajectory obtained under such a terminal speed condition does not increase in speed at the terminal end, that is, it becomes excessive until the center of gravity speed increases and becomes uncontrollable. That is, walking stability is ensured.

  The biped walking robot of the present invention can suppress the inclination of the foot while ensuring walking stability even when the foot is inclined relative to the road surface unexpectedly.

The side view of a biped walking robot is shown. The flowchart of a gravity center track | orbit determination process is shown. An example in which the biped walking robot is tilted is shown. It is a figure explaining the usage of a distance sensor in the case of providing three distance sensors.

  FIG. 1 shows a side view of a biped robot 10 during walking. Hereinafter, the biped walking robot 10 is simply referred to as a robot 10. Each joint of the leg has an actuator for controlling the joint angle and an encoder for detecting the joint angle. The robot 10 includes a controller 20, and the controller 20 controls each joint of the leg based on the gait data. Here, “control each joint” means that the actuator is controlled to follow the target angle of each joint obtained from the gait data. The controller 20 normally calculates time-series data of target angles of each joint from gait data (target trajectory of foot and target trajectory of center of gravity), and controls each joint so as to follow the calculated target angle. When an unscheduled foot inclination is detected in the gait data, the controller 20 recalculates the center of gravity trajectory based on the acceleration of the center of gravity regardless of the center of gravity trajectory described in the gait data. Since a gait data generation method and a method for generating a joint target angle from gait data may be performed using a conventionally known method, detailed description thereof will be omitted.

  The robot 10 includes two distance sensors 16a and 16b on the sole, and detects the inclination of the foot using these distance sensors. The distance sensor 16a is disposed at a position corresponding to the toe, and the distance sensor 16b is disposed at a position corresponding to the heel. The distance sensors 16a and 16b measure the distance between the road surface and the sole of the foot at each position. The inclination angle of the foot with respect to the road surface can be obtained from the sensor data of the two distance sensors.

Before explaining the processing when an unplanned foot inclination is detected on the gait data, the center-of-gravity trajectory that ensures walking stability will be explained. In FIG. 1, reference numeral 10A indicates the posture of the robot at the current time T0, and reference numeral 10B indicates the planned posture of the robot at the time T1. The symbol x _g (T0) indicates the x coordinate of the centroid G at the current time T0, and the symbol z _g (T0) indicates the z coordinate of the centroid G at the current time T0. The code code q _x (T0) indicates the x coordinate of the ZMP at the current time T0. x _p (T0) indicates the x coordinate of the foot position at the current time T0. Here, the foot position _xp is represented by the position of a point fixed to the toe of the foot.

Hereinafter, the x component x _g (i) of the barycentric trajectory (the symbol “i” represents the time in the time discrete time system) will be described. The following discussion also holds for the y component. ZMP is defined as a point at which a concentrated reaction force equivalent to a pressure (reaction force) distributed on the sole intersects the floor surface. When the moment around the pitch axis (y-axis) acting on the center of gravity G is expressed by r _y (i), the balance equation around the pitch axis (y-axis) around the center of gravity G at time (i) is Given in 1).

Here, m represents the mass of the entire robot, and g represents the acceleration of the center of gravity. z _g (i) represents the z coordinate of the center of gravity G. The superscript (2) represents twice the time differentiation. That is, x _g ⁽²⁾ (i) represents the x component of the acceleration of the center of gravity G. (Equation 1) corresponds to the ZMP equation in the x-axis direction. (Equation 1) represents the ZMP position q _x (i) at time i, the gravity center positions x _g (i), z _g (i) and their accelerations x _g ⁽²⁾ (i), z _g ⁽²⁾ ( The relationship of i) is shown.

The z component of the acceleration of the center of gravity G, that is, z _g ⁽²⁾ (i) is given in advance. z _g ⁽²⁾ (i) substantially corresponds to the vertical movement of the waist during walking. The x component of the acceleration of the center of gravity G, that is, x _g ⁽²⁾ (i) and the x component x _g ⁽¹⁾ (i) of the velocity of the center of gravity G are based on Euler approximation using a minute time interval Δt. It is expressed by the following (Equation 2).

From (Equation 2), the x component x _g ⁽²⁾ (i) of the acceleration of the center of gravity G is the position of three consecutive times of the center of gravity G, x _g (i), x _g (i + 1), x _g (i + 2) Can be approximated by the following (Equation 3).

Substituting (Equation 3) into (Equation 1) and setting r _y (i) = 0, the following (Equation 4) is obtained.

(Expression 4) is a variable of the centroid position (x component) at three consecutive times of the centroid acceleration x _g ⁽²⁾ (i) in the x-axis direction in the ZMP equation of (Expression 1), that is, x _g ( It is a ternary equation approximated by i + 1), x _g (i), and x _g (i−1). (Expression 4) is obtained by _calculating the three consecutive points x _g (i + 1), x _g (i), and x _g (i−1) of the x component of the centroid position in time by the position q _x (i) of the ZMP This indicates that the positional relationship is determined.

  When (Equation 4) is expressed in matrix for i = 1 to n, the following (Equation 5) is obtained. Time (n) corresponds to scheduled time T1 when the free leg (right leg) lands in FIG. The walking speed is given in advance, and therefore the scheduled landing time T1 is fixed. Since the current time T0 can also be detected, the time span of (Equation 5) is the time TS from the current time T0 to the scheduled landing time T1. The order n of (Expression 5) is determined by a value obtained by dividing the time span TS by a predetermined sampling time Δt.

Note that q _x (1) and q _x (n) in (Equation 5) have different meanings from other q _x (i), that is, the position of ZMP. The first line of (Equation 5) corresponds to the initial boundary condition that guarantees that the initial speed matches the center of gravity speed at the current time. This corresponds to a terminal boundary condition that guarantees that the measured speed is met. From the definition of (Expression 2), q _x (1) = v (T0) Δt and q _x (n) = v (T1) Δt. As shown in FIG. 1, v (T0) and v (T1) are x components of the velocity of the center of gravity G at the current time T0 and the free leg landing scheduled time T1. From (Equation 5), the x component [Xg] (that is, x _g (i), i = 1 to n) of the gravity center trajectory can be obtained by the following (Equation 6).

(Expression 6) represents that the center-of-gravity trajectory [Xg] is obtained when the ZMP trajectory [Qx] including the initial boundary condition and the terminal boundary condition is given. (Equation 6) gives the center of gravity trajectory [Xg] with respect to the ZMP trajectory [Qx]. That is, the obtained gravity center position x _g (i) at time (i) indicates a relative position with respect to the ZMP position. Therefore, (6) after it determined the relative center of gravity pathway [Xg] for ZMP trajectory [Qx] by time (0) of the center of gravity position _x g (0) is the center of gravity position _x g at time T0 in FIG. 1 ( The ZMP trajectory [Qx] and the gravity center trajectory [Xg] are shifted in parallel to the road surface (XY plane) so as to coincide with T0). By this process, the continuity between the center of gravity trajectory obtained by (Equation 6) and the center of gravity orbit up to that point (the position at the end is x _g (T0)) is maintained. Thus, the center of gravity trajectory from time T0 to T1 is obtained. Since ZMP is the center point of the floor reaction force received by the foot, the foot position is determined so that the ZMP position is in the foot plane.

  The fact that the center-of-gravity trajectory obtained by (Equation 6) guarantees walking stability will be described. The boundary conditions given to obtain the solution [Xg] of (Equation 6) are the centroid speed at the initial time and the centroid speed at the end time. Of these, the center-of-gravity speed at the initial time may be the actual center-of-gravity speed at the current time. This is to ensure the continuity of the past center of gravity trajectory up to the current time and the future center of gravity orbit from the current time. When an appropriate value is given as the center of gravity speed at the end time, for example, the speed V0 equal to the walking speed, the center of gravity trajectory [Xg] obtained by (Equation 6) is guaranteed to always have the speed V0 at the end. . The speed of the center of gravity at the end of the orbit (speed in the x-axis direction) increases as the distance from the ZMP position increases. This is presumed that the speed of the tip of the inverted pendulum becomes faster as the tip position is farther from the fulcrum. The tip of the inverted pendulum corresponds to the center of gravity, and the fulcrum of the inverted pendulum corresponds to the ZMP. The ZMP equation derives a position where the velocity is realized when an appropriate magnitude of velocity is given as a terminal boundary condition. A position where an appropriate speed can be realized is nothing but a position (position in the XY plane) that is not far from the ZMP. The fact that the position of the center of gravity is not so far away from the ZMP means that the robot itself does not tilt greatly. That is, if the ZMP equation gives a speed of an appropriate magnitude as the end boundary condition, it gives a center of gravity position where the robot does not tilt greatly as the end position. If the speed of the center of gravity is an appropriate size given at the end, and the position of the center of gravity is not far from the ZMP, it means that walking stability is secured at the end of the center of gravity trajectory. By setting the end time of the center-of-gravity trajectory to be a scheduled landing time of 2 steps ahead or 3 steps ahead, it is ensured that it is possible to walk stably up to 2 steps ahead or 3 steps ahead.

The legged robot 10 calculates the center-of-gravity trajectory in which walking stability is ensured up to two steps ahead by the above (Equation 6). However, here, the time interval from the current time to the scheduled landing time of the second step is adopted as the time period (period from i = 0 to n) considered in (Equation 6). The gait data of the robot 10 also includes a target foot trajectory (X _p (i) in FIG. 1). The target foot trajectory is generated by connecting the current landing position and the next landing position with a predetermined curve. The robot 10 performs the above-described center-of-gravity trajectory generation process in a predetermined period (for example, every 20 msec), and as a result, always controls each joint of the leg based on the center-of-gravity trajectory that ensures walking stability. The time-series data of the target joint angle is obtained by so-called reverse kinematics (center of gravity reverse kinematics) from the position of the center of gravity and the position of the foot at each time.

  As described above, the controller 20 of the legged robot 10 periodically generates a gravity center trajectory from the current time to two steps ahead. The controller 20 normally updates the center-of-gravity trajectory (gait data) regularly. However, when an inclination of a foot that is not scheduled in the gait data is detected, the controller 20 changes the center-of-gravity trajectory described in the gait data. Regardless, the center of gravity trajectory is recalculated based on the acceleration of the center of gravity. Processing of the controller 20 at this time will be described next.

  FIG. 2 shows a flowchart of processing executed by the controller 20. The controller 20 acquires sensor data from the two distance sensors 16a and 16b (S2), and determines whether the foot is inclined from the road surface unexpectedly (S4). The controller 20 compares the inclination angle of the foot with respect to the road surface obtained from the sensor data with a predetermined inclination angle of the foot determined by the gait data, and the inclination angle is zero and the distance is zero on the gait data (that is, the foot). If the sensor data indicates that the inclination angle of the foot is not zero, it is determined that an unscheduled inclination has occurred.

  When an unscheduled inclination does not occur in the foot (S4: NO), the controller 20 controls the leg joint based on the regularly updated gait data. That is, the controller 20 calculates the target angle of each joint from the center of gravity trajectory (and the target trajectory of the foot) described in the gait data by inverse kinematics (S10), and the joint target from which the actual joint angle is obtained. Each joint is controlled to follow the corner (S12).

  On the other hand, when an unscheduled inclination occurs in the foot (S4: YES), the controller 20 calculates the acceleration of the center of gravity at that time (S6). The acceleration calculation method will be described later. Next, the controller 20 obtains the velocity of the center of gravity realized by the obtained acceleration. The obtained speed is set as an initial value, the speed of the center of gravity after a predetermined time is given as a termination condition, (Equation 6) is solved, and the center of gravity trajectory is recalculated (S8). The term “after the predetermined time” is typically until the landing timing next to the swing leg (or two steps ahead). The method for obtaining the acceleration will be described later.

  When the center-of-gravity trajectory is calculated in step S8, the controller 20 calculates the target angle of each joint by inverse kinematics based on the center-of-gravity trajectory newly obtained in step S8 instead of the center-of-gravity trajectory used so far. Then, each joint is controlled so as to follow the target joint angle from which the actual joint angle is obtained (S12). The center-of-gravity trajectory obtained in step S8 is obtained from the acceleration of the center of gravity due to an unplanned inclination of the foot. That is, the initial period of the center of gravity trajectory reflects the acceleration of the center of gravity obtained in step S6, and the time-series data of the target angles of each leg joint obtained from the center of gravity trajectory corresponds to the acceleration of the center of gravity. It will change the attitude of. That is, the time-series data of the target angle of the leg joint thus obtained moves the entire leg in a direction to return the inclination of the foot with respect to the road surface.

  Further, the center-of-gravity velocity at the end of the center-of-gravity trajectory obtained in step S8 is a value given as a boundary condition. Therefore, if an appropriate value (for example, a predetermined walking speed) is given as the terminal speed, even if the acceleration (speed) is large at the initial stage of the center of gravity trajectory, the center of gravity speed at the terminal always has a known appropriate value (for example, the walking speed). ) Is guaranteed. That is, at the end of the center of gravity trajectory, it is guaranteed to return to a steady walking motion.

  As described above, the biped walking robot 10 of the embodiment returns the inclination of the foot to zero by positively allowing the acceleration of the center of gravity due to the inclination when the foot is inclined unscheduled with respect to the road surface. . Furthermore, although the calculated center of gravity trajectory is affected by acceleration caused by the inclination of the foot during the initial period, it always settles at a known speed at the end of the center of gravity trajectory, so that walking stability is guaranteed.

A method for obtaining the center-of-gravity acceleration caused by the inclination of the foot will be described. FIG. 3 shows an example of the posture of the robot 10 when the foot 14 is tilted. The solid line indicates the posture of the tilted robot, and the broken line indicates the posture that is planned on the original gait data. The actual ZMP at this time is in the position of the toes (because ZMP is the center of the reaction force that the foot receives from the ground. Position). An actual ZMP x-coordinate and z-coordinate are represented by (q _x , q _z ). When the foot is tilted unscheduled, the actual ZMP is different from the target ZMP. Based on this difference, the acceleration x _g ^{d (2)} of the center of gravity can be obtained by the following (Equation 7).

In (Expression 7), q _x ^d and q _z ^d indicate the target ZMP, and q _x indicates the actual ZMP.

Another method for obtaining the acceleration of the center of gravity will be described. As shown in FIG. 3, the center-of-gravity acceleration x _g ^{d (2)} can also be expressed by the following (Equation 8) using θ as the slope of the foot with respect to the road surface.

In equation (8), _{x r} is the rotational center of the robot 10, i.e., corresponding to the actual ZMP position. K ₁ and k ₂ are setting parameters (gains).

From (Expression 7) or (Expression 8), the acceleration of the center of gravity when the foot tilts unscheduled is obtained. As previously indicated, the ZMP equation (Equation 6) for obtaining the center of gravity trajectory gives velocity as the initial boundary condition. _Assuming that the center-of-gravity velocity at the current time is V _g0 and the time interval in ( _Equation 6) is Δt, the acceleration of the center of gravity x _g ^{d (2)} obtained by (Equation 7) or (Equation 8 ⁾ and (Equation 6) The relationship between the initial boundary condition _{Vg_init} and the initial boundary condition is expressed by the following ( _Equation 9).

V _{g_init in} ( _Equation 9) corresponds to the center-of-gravity initial speed when the center-of-gravity acceleration x _g ^{d (2)} is considered. In step S8, the controller 20 of the robot 10 solves ( _Equation 6) using V _{g_init} given by ( _Equation 9) as an initial boundary condition.

  A method of using the ground contact sensor in the case of a biped walking robot having a foot having a joint around the pitch axis at a position corresponding to the thumb ball and a toe swinging relative to the heel will be described. FIG. 3 shows a side view of a foot in which the toes 14a and the heels 14b are connected so as to be swingable by a joint 14c. The foot 14 includes three ground sensors 16a, 16b, and 16c. The ground sensors 16a and 16c are attached to the bottom surface of the toe 14a. The ground sensor 16a is disposed at the tip of the toe 14a, and the ground sensor 16c is disposed at the rear end of the toe 14a. The ground sensor 16b is disposed on the bottom surface of the flange 14b. 3A shows a situation where the entire foot is in contact with the ground, and FIG. 3B shows a situation where the toe 14a is bent with respect to the heel 14b and the toe 14a is in contact with the ground. ing. FIG. 3C shows a situation where the front of the foot protrudes from the step W.

  As shown in FIG. 3A, when the entire foot is in contact with the ground, the controller 20 measures the inclination of the foot using the earliest ground sensor 16a and the last ground sensor 16b. As shown in FIG. 3B, when the heel 14b is floating but the toe 14a is grounded, the controller 20 measures the inclination of the foot using the ground sensors 16a and 16c before and after the toe 14a. As shown in FIG. 3C, when the tip of the foot protrudes from the step W, the controller 20 measures the inclination of the foot using the ground sensor 16c behind the toe 16a and the ground sensor 16b of the heel 14b. To do.

  Specific examples of the present invention have been described in detail above, but these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications and changes of the specific examples illustrated above. The technical elements described in this specification or the drawings exhibit technical usefulness alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. In addition, the technology exemplified in this specification or the drawings can achieve a plurality of objects at the same time, and has technical usefulness by achieving one of the objects.

10: Biped walking robot 14: Foot 16a, 16b, 16c: Ground sensor 20: Controller

Claims (1)

A plurality of distance sensors that are disposed at least two places before and after the sole, and measure the distance to the road surface;
A time series data of the target angle of each joint of the leg from the target trajectory of the foot, and a controller for controlling each joint of the leg so as to follow the calculated time series data.
A step of calculating the acceleration of the center of gravity at that time when the distance sensor detects that the foot is tilted with respect to the road surface other than the scheduled takeoff timing on the target track;
The center-of- gravity velocity realized by the calculated center-of-gravity acceleration is set as an initial value, and the center-of- gravity trajectory is calculated by solving the ZMP equation with a predetermined center-of-gravity velocity at the end of a predetermined period as a termination condition. Steps,
Recalculating time series data of the target angle of the leg joint from the calculated trajectory of the center of gravity;
The biped walking robot is characterized by controlling each joint of the leg so as to follow the time series data of the recalculated target angle.

Images