JP2008126330A - Legged robot and control method thereof - Google Patents

Abstract

A legged robot capable of stably maintaining a standing state and a control method thereof.
A legged robot according to an aspect of the present invention is a legged robot including a leg portion 10 provided with an ankle joint, and a trunk portion 50 connected to the leg portion 10, wherein the legged robot receives a drive signal. In response, a motor for driving the ankle joint, a sensor 150 for measuring the state quantity of the torso, a target torque calculation unit 135 for calculating the target torque of the ankle joint according to an output from the sensor 150, and a limit for the target torque A limit value setting unit 136 in which a value is set, and a comparison unit 137 that compares the limit value with a target torque and outputs a drive signal corresponding to the limit value to the motor 142 when the target torque exceeds the limit value; , Are provided.
[Selection] Figure 2

Description

  The present invention relates to a legged robot having legs and a control method thereof.

  In recent years, a leg for walking is provided, the leg is driven, and a foot portion provided at the lower end of the leg is placed on the floor surface based on predetermined gait data to perform a walking motion. Legged mobile robots have been developed. (For example, Patent Document 1)

  In such a legged mobile robot, first, the foot portion of the leg is brought into contact with the floor surface to form a supporting leg, and then the floor surface is pushed by the back of the foot to raise the entire leg portion (the entire robot). Thus, the next walking motion is performed by driving the legs. The driven leg portion is a free leg, while the other leg portion is a support leg. In this way, a walking motion can be performed by alternately switching the free leg and the support leg.

In order to stably perform such a walking motion, it is necessary to control the position of the center of gravity of the entire robot and drive the leg portion. That is, in the case of a biped walking type legged walking type robot having leg portions on the left and right sides, the moment due to the center of gravity of the entire robot does not act on the ground contact surface of the foot portion of the supporting leg that contacts the floor surface to walk A point (ZMP = Zero Moment Point) is located.
JP-A-5-305580

  The leg portion of the legged mobile robot shown in Patent Document 1 includes a hip joint, a thigh, a knee joint, a lower leg, an ankle joint, and a foot. Then, the amount of movement of the waist is calculated from the deviation between the target center-of-gravity position of the trunk connected to the legs and the actual center-of-gravity position. And the joint angle to drive is calculated | required from the calculated movement amount of the waist, and it drives.

  By the way, the legged mobile robot has a problem in that it falls down when a disturbance occurs during standing. For example, when the floor surface F moves up and down or the robot 1 is pressed in the standing state, it falls down. In the legged mobile robot of Patent Document 1, the joint drive control value is corrected based on the absolute angle or the inclination angular velocity of the foot detected by the sensor. In this control, filter processing is performed on the output from the sensor. Then, a calculation for correcting the filtered output is performed. For this reason, in this legged mobile robot, a delay occurs in the filter processing and calculation, and it is difficult to increase the responsiveness. Therefore, the conventional legged mobile robot cannot cope with the disturbance, and the robot falls. Thus, the conventional control has a problem that it is difficult to stably maintain the standing state.

  The present invention has been made to solve such a problem, and an object of the present invention is to provide a legged robot capable of stably maintaining a standing state and a control method thereof.

A legged robot according to a first aspect of the present invention is a legged robot including a leg portion provided with an ankle joint and a torso portion connected to the leg portion, and the ankle according to a driving signal. An ankle joint drive unit that drives a joint; a sensor that measures a state quantity of the torso; a target torque calculation unit that calculates a target torque of the ankle joint according to an output from the sensor; and A limit value setting unit that sets a limit value, stores the set limit value, compares the limit value with the target torque, and responds to the limit value when the target torque exceeds the limit value And a comparison unit that outputs the drive signal to the ankle joint drive unit. Thereby, the standing state can be stably maintained.
The legged robot according to a second aspect of the present invention is characterized in that, in the legged robot, the state quantity of the body part is a value indicating a current state of the body part.

  A legged robot according to a third aspect of the present invention is the legged robot according to the above-described legged robot, wherein the target torque calculation unit is configured to calculate a state quantity of the body part measured by the sensor and a target state quantity of the body part. The target torque is calculated based on the difference. As a result, even when a disturbance occurs, it can be quickly controlled.

  A legged robot according to a fourth aspect of the present invention is characterized in that, in the legged robot, the limit value is variable. Thereby, a standing state can be maintained more stably.

A control method for a legged robot according to a fifth aspect of the present invention is a control method for a legged robot including a leg portion provided with an ankle joint and a torso portion connected to the leg portion. A step of measuring a state quantity of the torso, a step of calculating a target torque of the ankle joint according to an output from the sensor, and comparing the limit value with the target torque; A step of outputting a drive signal corresponding to the limit value when the value is exceeded, and a step of driving the ankle joint based on the drive signal. Thereby, the standing state can be stably maintained.
A control method for a legged robot according to a sixth aspect of the present invention is characterized in that, in the above legged robot, the state quantity of the body part is a value indicating the current state of the body part. It is.

  A control method for a legged robot according to a seventh aspect of the present invention is the above-described control method, wherein in the step of calculating the target torque, the measured state quantity of the body part and the target state quantity of the body part are calculated. The target torque is calculated based on the difference between the two. As a result, even when a disturbance occurs, it can be quickly controlled.

  The legged robot according to an eighth aspect of the present invention is characterized in that, in the above legged robot, the limit value is variable. Thereby, a standing state can be maintained more stably.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the legged robot which can maintain a standing state stably, and its control method.

Embodiment 1 of the Invention
A legged mobile robot (hereinafter simply referred to as a robot) according to a first embodiment of the present invention will be described below with reference to FIG.

  FIG. 1 is a schematic diagram schematically showing the robot 1 as viewed from the front, and shows the robot 1 walking on the floor F. FIG. In FIG. 1, for convenience of explanation, the direction in which the robot 1 travels (front-rear direction) is the x-axis, and the direction in which the robot 1 travels is perpendicular to the horizontal direction (left-right direction), the y-axis. The direction (vertical direction) extending in the vertical direction from the plane to be performed is defined as the z axis, and a description will be given using a coordinate system including these three axes. That is, in FIG. 1, the x-axis indicates the depth direction of the paper surface, the y-axis indicates the left-right direction toward the paper surface, and the z-axis indicates the vertical direction in the paper surface.

  As shown in FIG. 1, the robot 1 includes a head 2, a trunk (torso) 3, a waist 4 coupled to the trunk 3, a right arm 5, a left arm 6 connected to the trunk 3, and a waist 4 is a biped walking robot that includes a leg portion 10 that is fixed to be freely pivotable with respect to 4. Details will be described below.

  The head 2 includes a pair of left and right imaging units (not shown) for visually imaging the environment around the robot 1, and the head 2 is parallel to the trunk 3 in the vertical direction. By rotating around the axis, the surrounding environment is imaged widely. Image data indicating the captured surrounding environment is transmitted to the control unit 130 described later, and is used as information for determining the operation of the robot 1.

  The trunk 3 houses therein a control unit 130 that controls the operation of the robot 1, a battery (not shown) for supplying power to a leg motor and the like. The control unit 130 includes a storage area that stores gait data for driving the leg 10 and moving the robot 1, an arithmetic processing unit that reads gait data stored in the storage area, and the leg 10. A motor driving unit for driving the motor. The configuration of the control unit 130 will be described later. Each of these components operates by being supplied with electric power from a battery (not shown) provided inside the trunk 3.

  The arithmetic processing unit reads out the gait data stored in the storage area and calculates the joint angle of the leg 10 necessary to realize the posture of the robot 1 specified by the read out gait data. And the signal based on the joint angle computed in this way is transmitted to a motor drive part.

  The motor drive unit specifies the drive amount of each motor for driving the leg portion based on the signal transmitted from the arithmetic processing unit, and the motor drive signal for driving the motor with these drive amounts is sent to each motor. Send to. As a result, the driving amount at each joint of the leg 10 is changed, and the movement of the robot 1 is controlled.

  In addition to instructing the motor to be driven based on the read gait data, the arithmetic processing unit also receives signals from sensors (not shown) such as gyros, accelerometers, and rotary encoders incorporated in the robot 1. In response, the motor drive amount is adjusted. Thus, the robot 1 can be maintained in a stable state by adjusting the joint angle of the leg 10 according to the external force acting on the robot 1 detected by the sensor, the posture of the robot 1, and the like.

  The right arm 5 and the left arm 6 are pivotally connected to the trunk 3, and perform the same movement as a human arm portion by driving joint portions provided at the elbow portion and the wrist portion. Can do. Further, the hand part connected to the tip of the wrist part has a hand structure for gripping an object (not shown), and by driving a plurality of finger joints incorporated in the hand structure, various kinds of parts are provided. It becomes possible to grip an object having a shape.

  The lumbar part 4 is connected to the trunk 3 so as to rotate, and the driving energy necessary for driving the leg part 10 can be obtained by combining the rotational action of the lumbar part 4 when performing a walking action. Can be reduced.

  A leg 10 (right leg 20 and left leg 30) for bipedal walking is composed of a right leg 20 and a left leg 30. Specifically, as shown in FIG. 2, the right leg 20 includes a right hip joint 21, an upper right thigh 22, a right knee joint 23, a right lower leg 24, a right ankle joint 25, and a right foot 26. A left hip joint 31, a left upper thigh 32, a left knee joint 33, a left lower thigh 34, a left ankle joint 35, and a left foot 36 are provided.

  The right leg 20 and the left leg 30 are each driven to a desired angle by transmitting a driving force from a motor (not shown) via a pulley and a belt (not shown). A desired movement can be made to a leg part.

  In this embodiment, when the leg 10 (the right leg 20 and the left leg 30) lifts the lower leg to the front side around the knee joint, the upper leg and the lower leg are placed rearward like a human leg. It is in an open state (a state where the lower leg rotates around the knee joint on the rear side of the extension line of the upper leg).

  The gait data stored in the storage area is associated with the amount of movement of the leg 10 specified by the signal sent from the operation unit (not shown), and the foot of the leg 10 (right foot 26, left foot). 36), the position of the tip (toe) and the position of the main body of the robot 1 are instructed over time in a coordinate system (for example, an xyz coordinate system) that defines a space in which the robot 1 moves. In addition, about the production | generation of gait data, since the method widely used conventionally can be used, description is abbreviate | omitted.

  Here, the control unit 130 will be described. The control unit 130 includes a CPU (Central Processing Unit) that is an arithmetic processing unit, a ROM (Read Only Memory) and RAM (Random Access Memory) that are storage areas, a communication interface, and the like, and performs various operations of the robot 1. Control. For example, the ROM stores a control program for control, various setting data, and the like. Then, the CPU reads the control program stored in the ROM and develops it in the RAM. Then, the program is executed in accordance with the setting data and the output from the sensor or the like.

  The trunk 3 is provided with a sensor 150. The sensor 150 is, for example, a gyro sensor or an acceleration sensor. The sensor 150 measures the inclination angular velocity and the inclination angle of the trunk part including the trunk 3 and the waist part 4. Alternatively, the sensor 150 measures the translational angular velocity of the body part. The number of sensors 150 is not limited to one and may be two or more. For example, the sensor 150 may include a gyro sensor that measures a tilt angle and a tilt acceleration, and an acceleration sensor that measures a translational acceleration. The control unit 130 performs feedback control based on the output from the sensor 150 when the robot 1 is standing. Thereby, even when a disturbance or the like occurs, the robot 1 can be prevented from falling.

  Next, control for maintaining the standing state in the robot 1 will be described with reference to FIGS. 2 and 3. FIG. 2 is a block diagram illustrating a configuration of the control unit 130. FIG. 3 is a side view schematically showing the configuration of the robot 1 when standing. In FIG. 3, only the leg portion 20 and the trunk portion 50 are shown for simplification of description. Here, the torso part 50 includes, for example, the trunk 3 and the waist part 4 of the robot 1 shown in FIG. Therefore, the trunk | drum 50 is connected with the leg part. In FIG. 3, since the right leg 20 and the left leg 30 are aligned, the joint angles of the right leg 20 and the left leg 30 are the same.

  The control unit 130 includes a target state quantity setting unit 132, an amplifier 133, a target torque calculation unit 135, a limit value setting unit 136, and a comparison unit 137. Then, the control unit 130 outputs a drive signal for driving the motor 141 and the motor 142. And the control part 130 is controlled so that the robot 1 in a standing state does not fall. Here, the robot 1 stands still on the inclined floor F. The control unit 130 performs feedback control so that the robot 1 does not fall down even when a disturbance such as the inclination of the inclined floor surface F changes or the floor surface F moves up and down occurs.

  Here, in the target state quantity setting unit 132, the target state quantity of the robot 1 is set. The target state quantity includes the target state quantity of the body part 50 and the target state quantities of places other than the body part 50. Examples of the target state quantity of the body part 50 include a body inclination angle, a body inclination angular velocity, and a body translational acceleration. On the other hand, the target state quantity other than the body part 50 includes the target angle of each joint other than the ankle joint. In the target state quantity setting unit 132, for example, target joint angles of the hip joint and knee joint of the leg 10 are set. Note that the target state amount in the standing state may be determined in advance or may be calculated by a predetermined program.

  The target state quantity other than the body part is input to the amplifier 133. Here, one amplifier 133 is shown, but actually, a plurality of amplifiers 133 are provided according to the degree of freedom of joints. The amplifier 133 is an amplifier for driving other than the ankle joint. That is, the amplifier 133 is an amplifier for driving a knee joint, a hip joint, and the like. The amplifier 133 is provided for each of the left leg 30 and the right leg 20.

  An appropriate gain is set for the amplifier 133. The amplifier 133 calculates a target torque based on the target joint angle and outputs a drive signal corresponding to the target torque of the motor 141. The drive signal output from the amplifier 133 is input to the motor 141. The motor 141 is, for example, a joint driving unit for driving a knee joint and a hip joint. The motor 141 is driven by a drive signal input from the amplifier 133. As a result, the joints other than the ankle joint are driven with the target torque, and the target joint angle is controlled. Actually, the motors 141 and the amplifiers 133 are provided in the number corresponding to the degrees of freedom of joints other than the ankle joint.

The target state quantity of the torso part 50 is input to the target torque calculator 135 that calculates the target torque of the ankle joint. Furthermore, the state amount of the body part 50 measured by the sensor 150 is input to the target torque calculation unit 135. The state quantity of the body part 50 is a value indicating the current state of the body part 50. When the state of the body part 50 changes, the state quantity also changes. Here, the fuselage tilt angle, the fuselage tilt angular velocity, and the fuselage translational acceleration are input to the target torque calculator 135 as state quantities. Accordingly, when the tilt angle, tilt angular velocity, and translational acceleration of the body portion 50 change, the state quantity corresponding to the change changes. Of course, the state quantity may be one of the trunk tilt angle, the trunk tilt angular velocity, and the trunk translational acceleration, or may be two or more. Furthermore, state quantities other than the fuselage tilt angle, the fuselage tilt angular velocity, and the fuselage translational acceleration may be used.
The target torque calculation unit 135 is a controller for calculating the target torque based on the state quantity of the body part 50 measured by the sensor 150 and the target state quantity. Specifically, the difference between the target value from the target state quantity setting unit 132 and the measured value from the sensor 150 is obtained for each of the trunk tilt angle, the trunk tilt angular velocity, and the trunk translational acceleration. Then, a predetermined feedback gain is applied to the difference between the target value and the measured value to obtain the target torque. Thereby, the target torque for the motor 142 of the ankle joint 25 is calculated. Then, the target torque calculation unit 135 performs feedback control according to the output from the sensor 150. Accordingly, the target torque of the ankle joint 25 changes in accordance with the change in the state of the body portion 50.

  Here, the target state quantity in the standing state is set in the target state quantity setting unit 132. That is, a target joint angle other than the ankle joint in the state where the robot 1 is standing still and the target tilt angle of the body unit 50 are set in the target state amount setting unit 132. Thereby, even if a disturbance occurs when standing still in the standing state, it can be prevented from falling. For example, when the inclination angle of the floor surface F changes in the standing state, a force is received from the floor surface F. Accordingly, the state of the robot 1 changes according to the change in the tilt angle. In this case, the fuselage tilt angle, the fuselage tilt angular velocity, and the fuselage translational acceleration measured by the sensor 150 are changed from the set target state values. The target torque calculation unit 135 calculates the target torque based on the difference between the target state quantity and the measurement value obtained by the sensor 150. Further, the angle of the joint other than the ankle joint changes from the set target joint angle. An angle other than the ankle joint may be measured with a rotary encoder or the like, and each joint angle may be controlled to return to the target joint angle. Thereby, even if the state quantity of the trunk | drum 50 approaches the target state quantity and a disturbance arises, it can prevent falling. And if the change of the inclination angle of the floor surface F converges, it will stop and will return to the original standing state. Thus, it is possible to operate quickly by performing feedback control based on the difference between the target state quantity and the measured value. Therefore, it can prevent falling and can maintain a standing state stably.

  Here, the target torque output from the target torque calculator 135 is input to the comparator 137. The input target torque is for driving the ankle joint. A limit value for the target torque of the ankle joint is set in the limit value setting unit 136. That is, the limit value setting unit 136 sets a limit value for the target torque and stores the set limit value. The limit value is different from the maximum torque that the motor 142 can output, and is lower than the maximum torque. The limit value set in the limit value setting unit 136 is input to the comparison unit 137. The comparison unit 137 compares the limit value with the target torque. If the target torque does not exceed the limit value, a drive signal corresponding to the target torque is output to the motor 142. On the other hand, when the target torque exceeds the limit value, a drive signal corresponding to the limit value is output to the motor 142. As a result, the motor 142 is driven with a torque corresponding to the limit value. That is, the motor 142 is driven using the limit value as the driving torque.

  As the limit value, an upper limit and a lower limit of the target torque are set. Here, the torque is positive or negative depending on the rotation direction of the ankle joint. That is, the rotation direction of the ankle joint is opposite between positive torque and negative torque. For example, in the limit value setting unit 136, a positive limit value is set as the upper limit value of the torque, and a negative limit value is set as the lower limit value of the torque. Thereby, the allowable range of the target torque is set. Therefore, it is possible to prevent the driving torque for driving the ankle joint from being outside the allowable range. That is, when the target torque is greater than or equal to the upper limit, the upper limit value is the drive torque, and when the target torque is less than or equal to the lower limit, the lower limit value is the drive torque. Here, limit values are set for the ankle joints of the left and right legs, respectively.

  When a limit value is set for the target torque, the ankle joint becomes soft. Therefore, even if the inclination angle of the floor surface F suddenly changes, the foot follows the floor surface F. That is, the soles are not peeled off from the floor surface F. In other words, a limit value is set such that the sole does not peel off the floor surface F even when a disturbance occurs. Since the foot follows the floor surface F, the torque of the ankle joint can be reliably transmitted to the floor surface F. That is, the posture of the body part 50 can be reliably controlled by the floor reaction force. Thereby, even when a disturbance occurs, the robot 1 can be prevented from falling.

  On the other hand, when the limit value is not set for the target torque of the ankle joint, if the inclination angle of the floor surface F changes abruptly, the change in the state quantity of the body portion 50 becomes large. In this case, the target torque becomes excessive. If the target torque is too large, the soles are peeled off from the floor surface F. Therefore, the torque of the ankle joint is not transmitted to the floor surface F, and it becomes difficult to control the posture. In order to control the posture, it is necessary to lower the feedback gain so that the sole does not peel off. However, when the feedback gain is lowered, the controllability is lowered.

  In the present embodiment, a limit value is provided for the target torque of the ankle joint. When the target torque exceeds the limit value, the ankle joint is driven with the limit value. Thereby, since the foot follows the floor surface F, it can prevent falling. That is, when the inclination angle of the body portion 50 changes due to disturbance of the floor surface F, control is performed to restore the inclination angle to its original value. At this time, the sole does not peel off the floor surface F. Further, when the change in the inclination angle is large, the control is performed with the limit value of the target torque of the ankle joint. For this reason, a feedback gain can be made high and controllability can be improved. Furthermore, since detailed setting of the feedback gain is unnecessary, the controller can be simplified. That is, it is not necessary to optimize the feedback gain in accordance with the disturbance frequency, so that the controller can be easily designed. Thus, the above control is effective for disturbances in a wide band from a high frequency to a low frequency. Furthermore, since there is no need to calculate the correction amount of the joint angle as in the prior art, the control delay can be reduced.

  Further, since the ankle joint is driven based on the target torque, even if a disturbance occurs, it can be quickly returned to the original state. The sensor 150 may be any sensor that can detect a change in the body portion 50 and is not limited to the above-described sensor. That is, when modeling with an inverted pendulum, it is possible to use the sensor 150 that can detect that the inverted pendulum is moving. The limit value can be set, for example, based on the existence region of the floor reaction force center point. That is, the position of the center point of the floor reaction force immediately before the toe or heel starts to float is set as the boundary of the region where the floor reaction force center point exists. Specifically, the limit value can be calculated by obtaining a vertical load and a horizontal load and using a balance formula of moments around the ankle at the boundary of the existing region.

  In the above description, the state in which both legs are aligned as shown in FIG. 3 has been described. However, the present invention is not limited to this state. For example, even when the left and right legs are open in the front-rear direction, the standing state can be maintained. Also, the number of legs is not limited to two, and may be one or three or more.

  Next, the configuration of the target torque calculation unit 135 will be described with reference to FIGS. 4 to 7 are block diagrams schematically showing the configuration of the target torque calculation unit 135. Here, FIGS. 4 to 7 show configuration examples of the target torque calculator 135, respectively. That is, FIGS. 4 to 7 show different configuration examples of the target torque calculator 135.

  First, a first configuration example will be described with reference to FIG. In FIG. 4, feedback control is performed based on the tilt angle and tilt angular velocity of the body portion 50. Specifically, the difference between the target body tilt angle set in the target state quantity setting unit 132 and the body tilt angle measured by the sensor 150 is taken. That is, the difference between the target state quantity and the measured value is calculated with respect to the tilt angle. Then, a feedback gain K1 is applied to the difference between the tilt angles. Then, the difference between the output value from the feedback gain K1 and the trunk inclination angular velocity measured by the sensor 150 is taken. Then, the target torque is obtained by multiplying the difference of the inclination angular velocity by the feedback gain K2. In this way, the target torque is repeatedly calculated, and feedback control is performed as described above.

  In the second configuration example shown in FIG. 5, feedback control is performed based on the tilt angle, tilt angular velocity, and translational acceleration of the body portion 50. Specifically, the difference between the target body tilt angle set in the target state quantity setting unit 132 and the body tilt angle measured by the sensor 150 is taken. Then, a feedback gain K1 is applied to the difference between the tilt angles. Then, the difference between the output value from the feedback gain K1 and the trunk inclination angular velocity measured by the sensor 150 is taken. Then, this difference is multiplied by a feedback gain K2. Furthermore, the difference between the target body translational acceleration set in the target state quantity setting unit 132 and the body translational acceleration measured by the sensor 150 is taken. Then, this difference is multiplied by a feedback gain K3. The sum of the output value from the feedback gain K2 and the output value from the feedback gain K3 is set as the target torque. As described above, in the second configuration example, the target torque is repeatedly calculated in consideration of the translational acceleration of the body portion 50, and the feedback control is performed as described above.

  In the third configuration example shown in FIG. 6, feedback control is performed based on the tilt angle, tilt angular velocity, and translational acceleration of the body portion 50 as in the second configuration example. Specifically, the difference between the target body tilt angle set in the target state quantity setting unit 132 and the body tilt angle measured by the sensor 150 is taken. Then, a feedback gain K1 is applied to the difference between the tilt angles. Then, the difference between the output value from the feedback gain K1 and the trunk inclination angular velocity measured by the sensor 150 is taken. Then, this difference is multiplied by a feedback gain K2. Furthermore, the difference between the target body translational acceleration set in the target state quantity setting unit 132 and the body translational acceleration measured by the sensor 150 is taken. Then, this difference is multiplied by a feedback gain K3. The calculation processing so far is the same as in the second configuration example. Furthermore, in the third configuration example, time integration is performed with respect to the tilt angle. That is, after performing time integration on the difference in tilt angle, the feedback gain K4 is applied. The sum of the output value of the feedback gain K4, the output value of the feedback gain K2, and the output value of the feedback gain K3 is set as the target torque. As described above, in the third configuration example, the target torque is repeatedly calculated in consideration of the translational acceleration of the body portion 50, and the feedback control is performed as described above.

  In the fourth configuration example shown in FIG. 7, feedback control is performed based on the tilt angle and tilt angular velocity of the body portion 50. Specifically, the difference between the target body tilt angle set in the target state quantity setting unit 132 and the body tilt angle measured by the sensor 150 is taken. Then, a feedback gain K1 is applied to the difference between the tilt angles. Further, the difference between the target body inclination angular velocity set in the target state quantity setting unit 132 and the body inclination angular velocity measured by the sensor 150 is taken. Then, a feedback gain K <b> 2 is applied to the difference in inclination angular velocity of the body portion 50. The sum of the output value of the feedback gain K1 and the output value of the feedback gain K2 is set as the target torque. In this way, the target torque is repeatedly calculated, and feedback control is performed as described above.

  In addition, you may change a limit value according to a condition or an environment. For example, when the inclination angle and material of the floor surface F are different, different limit values may be set. For example, the limit value may be changed when the floor surface F is horizontal and the inclination angle is 0 ° and when the floor surface F is inclined 45 °. Further, when the vertical load changes, the limit value may be changed according to the vertical load. Specifically, when the robot 1 is on an elevator or the like and the floor surface is accelerated or decelerated in the vertical direction, the vertical load applied to the robot 1 changes. In such a case, different limit values may be set at the time of ascending, descending, and stationary. Thus, by making the limit value variable, control according to the situation can be performed, and the standing state can be maintained more stably.

1 is an overall schematic diagram schematically showing an entire robot according to the present invention. It is a block diagram which shows the control part provided in the robot which concerns on this invention. It is a side view which shows typically the structure at the time of standing of the robot which concerns on this invention. It is a block diagram which shows the 1st structural example of the target torque calculation part in the robot which concerns on this invention. It is a block diagram which shows the 2nd structural example of the target torque calculation part in the robot which concerns on this invention. It is a block diagram which shows the 3rd structural example of the target torque calculation part in the robot which concerns on this invention. It is a block diagram which shows the 4th structural example of the target torque calculation part in the robot which concerns on this invention.

Explanation of symbols

1 robot, 2 heads, 3 trunks, 4 hips, 5 right arms, 6 left arms,
10 legs, 20 right leg, 21 right hip joint, 22 upper right thigh, 23 right knee joint,
24 right lower leg, 25 right ankle joint, 26 right foot,
30 Left leg 30, 31 Left hip joint, 32 Left upper leg, 33 Left knee joint, 34 Left lower leg,
35 left ankle joint, 36 left foot,
130 control unit, 132 target state quantity setting unit, 133 amplifier,
135 target torque calculation unit, 136 limit value setting unit, 137 comparison unit,
141 motor, 142 motor, 150 sensor,

Claims (8)

A legged robot including a leg portion provided with an ankle joint and a torso portion connected to the leg portion,
An ankle joint drive unit for driving the ankle joint according to a drive signal;
A sensor for measuring a state quantity of the body part;
A target torque calculator for calculating a target torque of the ankle joint according to an output from the sensor;
A limit value setting unit for setting a limit value for the target torque and storing the set limit value;
A comparison unit that compares the limit value with the target torque, and outputs a drive signal corresponding to the limit value to the ankle joint drive unit when the target torque exceeds the limit value;
Legged robot with   The legged robot according to claim 1, wherein the state quantity of the body part is a value indicating a current state of the body part.   The legged robot according to claim 1 or 2, wherein the target torque calculation unit calculates a target torque based on a difference between a state quantity of the body part measured by the sensor and a target state quantity of the body part. .   The legged robot according to claim 1, wherein the limit value is variable. A control method for a legged robot comprising a leg portion provided with an ankle joint and a torso portion connected to the leg portion,
Measuring a state quantity of the body part;
Calculating a target torque of the ankle joint according to an output from the sensor;
Comparing a limit value with the target torque, and outputting a drive signal corresponding to the limit value when the target torque exceeds the limit value;
And a step of driving the ankle joint based on the drive signal.   6. The control method for a legged robot according to claim 5, wherein the state quantity of the body part is a value indicating a current state of the body part.   The legged robot according to claim 5 or 6, wherein, in the step of calculating the target torque, the target torque is calculated based on a difference between the measured state quantity of the body part and a target state quantity of the body part. Control method.   8. The method for controlling a legged robot according to claim 5, wherein the limit value is variable.